wehub-resource-sync
7587 lines
406 KiB
Plaintext
7587 lines
406 KiB
Plaintext
{
|
||
"cells": [
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "aa3dfc3b",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.002313,
|
||
"end_time": "2026-06-13T02:51:06.527101+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:06.524788+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"# SEC Filing Signals: From 10-Q Text to Alpha Factors\n",
|
||
"\n",
|
||
"**Chapter 10: Text Feature Engineering**\n",
|
||
"\n",
|
||
"**Docker image**: `ml4t-gpu`\n",
|
||
"\n",
|
||
"**Section Reference**: See Section 10.5 for practitioner workflow and signal validation protocol\n",
|
||
"\n",
|
||
"> **GPU recommended**: this notebook runs FinBERT and sentence-transformer\n",
|
||
"> inference over thousands of MD&A passages (no training). On a GPU the\n",
|
||
"> MAX_SYMBOLS=50 default lands in ~5 minutes; CPU is 5–10× slower. For GPU:\n",
|
||
"> ```bash\n",
|
||
"> docker compose run --rm ml4t-gpu python 10_text_feature_engineering/09_filing_text_signals.py\n",
|
||
"> ```\n",
|
||
"\n",
|
||
"\n",
|
||
"## Purpose\n",
|
||
"\n",
|
||
"This notebook demonstrates how to construct alpha factors from SEC 10-Q filings.\n",
|
||
"Unlike headline-based signals (NB07), corporate filings provide dense, structured text\n",
|
||
"that reflects management's assessment of financial condition. We extract two complementary\n",
|
||
"signal types from MD&A sections:\n",
|
||
"\n",
|
||
"1. **Sentiment signals** via FinBERT (directional bias in management language)\n",
|
||
"2. **Semantic change signals** via sentence-transformer embeddings (quarter-over-quarter narrative shifts)\n",
|
||
"\n",
|
||
"The filing date provides a natural point-in-time anchor: the signal becomes available\n",
|
||
"when the SEC accepts the filing, not when the quarter ends.\n",
|
||
"\n",
|
||
"## Learning Objectives\n",
|
||
"\n",
|
||
"After completing this notebook, you will be able to:\n",
|
||
"- Load and explore SEC 10-Q MD&A text at scale\n",
|
||
"- Apply FinBERT sentiment scoring to long-form corporate text\n",
|
||
"- Compute document embeddings using sentence-transformers\n",
|
||
"- Construct a \"narrative change\" signal from sequential filing embeddings\n",
|
||
"- Join text signals to market data with point-in-time correctness\n",
|
||
"- Evaluate signal quality using IC, ICIR, and quintile analysis\n",
|
||
"\n",
|
||
"## Prerequisites\n",
|
||
"- Section 10.5 of the chapter (alpha-factor evaluation, point-in-time joins).\n",
|
||
"- SEC 10-Q MD&A panel produced by `data/equities/fundamentals/filings_download.py`.\n",
|
||
"\n",
|
||
"## Related Notebooks\n",
|
||
"- `04_bert_finetuning.py` / `06_finbert_cross_dataset.py` — FinBERT model details.\n",
|
||
"- `07_news_return_signals.py` — analogous workflow on headlines instead of filings."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 1,
|
||
"id": "1e252adc",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:06.531551Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:06.531445Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:08.381312Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:08.380800Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 1.852828,
|
||
"end_time": "2026-06-13T02:51:08.381936+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:06.529108+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"\"\"\"SEC Filing Text Signals - FinBERT sentiment and embedding-based alpha factors.\"\"\"\n",
|
||
"\n",
|
||
"import warnings\n",
|
||
"\n",
|
||
"import matplotlib.pyplot as plt\n",
|
||
"import numpy as np\n",
|
||
"import polars as pl\n",
|
||
"import torch\n",
|
||
"\n",
|
||
"from utils.paths import get_chapter_dir\n",
|
||
"from utils.reproducibility import set_global_seeds\n",
|
||
"\n",
|
||
"warnings.filterwarnings(\"ignore\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 2,
|
||
"id": "f9e9df6f",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:08.386266Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:08.386128Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:08.388340Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:08.387973Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.004954,
|
||
"end_time": "2026-06-13T02:51:08.388595+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:08.383641+00:00",
|
||
"status": "completed"
|
||
},
|
||
"tags": [
|
||
"parameters"
|
||
]
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"# Production defaults (0 = all symbols; full dataset has 477 companies, 6770 filings)\n",
|
||
"# FinBERT scoring is sequential per-filing, so runtime scales linearly.\n",
|
||
"# 50 symbols (~750 filings) runs in ~5 minutes on GPU; set to 0 for full dataset.\n",
|
||
"SEED = 42\n",
|
||
"MAX_SYMBOLS = 50\n",
|
||
"MAX_FILINGS = 0\n",
|
||
"BATCH_SIZE = 8\n",
|
||
"MAX_TOKENS = 512\n",
|
||
"EMBEDDING_MODEL = \"sentence-transformers/all-MiniLM-L6-v2\"\n",
|
||
"SENTIMENT_MODEL = \"yiyanghkust/finbert-tone\""
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 3,
|
||
"id": "b0703dfe",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:08.392258Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:08.392151Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:08.447732Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:08.447085Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.058032,
|
||
"end_time": "2026-06-13T02:51:08.448116+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:08.390084+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Device: cuda\n",
|
||
"GPU: NVIDIA GeForce RTX 3090\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"OUTPUT_DIR = get_chapter_dir(10) / \"output\" / \"filing_signals\"\n",
|
||
"OUTPUT_DIR.mkdir(parents=True, exist_ok=True)\n",
|
||
"\n",
|
||
"# Reproducibility — set_global_seeds covers Python random / NumPy / Torch.\n",
|
||
"set_global_seeds(SEED)\n",
|
||
"\n",
|
||
"device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
|
||
"print(f\"Device: {device}\")\n",
|
||
"if device.type == \"cuda\":\n",
|
||
" print(f\"GPU: {torch.cuda.get_device_name()}\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "84911f09",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.001547,
|
||
"end_time": "2026-06-13T02:51:08.451352+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:08.449805+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 1. Load SEC 10-Q MD&A Data\n",
|
||
"\n",
|
||
"The MD&A (Management's Discussion and Analysis) section is the most valuable narrative\n",
|
||
"section of quarterly filings. Unlike boilerplate Risk Factors that change slowly,\n",
|
||
"MD&A discusses current quarter performance and forward-looking outlook.\n",
|
||
"\n",
|
||
"Data comes from our SEC EDGAR download script\n",
|
||
"(`data/equities/fundamentals/filings_download.py --form 10-Q --universe sp500`),\n",
|
||
"which extracts MD&A sections from S&P 500 10-Q filings (2017-2021)."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 4,
|
||
"id": "d1cc260a",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:08.455311Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:08.455094Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:09.127521Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:09.127106Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.674933,
|
||
"end_time": "2026-06-13T02:51:09.127787+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:08.452854+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Loaded 6,770 MD&A sections from 477 companies\n",
|
||
"Date range: 2017-01-04 to 2021-12-22\n",
|
||
"Filtered to 50 symbols: 766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"text/html": [
|
||
"<div><style>\n",
|
||
".dataframe > thead > tr,\n",
|
||
".dataframe > tbody > tr {\n",
|
||
" text-align: right;\n",
|
||
" white-space: pre-wrap;\n",
|
||
"}\n",
|
||
"</style>\n",
|
||
"<small>shape: (5, 4)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>symbol</th><th>filing_date</th><th>period_end</th><th>word_count</th></tr><tr><td>str</td><td>date</td><td>date</td><td>u32</td></tr></thead><tbody><tr><td>"AEE"</td><td>2017-05-05</td><td>2017-03-31</td><td>8986</td></tr><tr><td>"AEE"</td><td>2017-08-04</td><td>2017-06-30</td><td>11308</td></tr><tr><td>"AEE"</td><td>2017-11-03</td><td>2017-09-30</td><td>12928</td></tr><tr><td>"AEE"</td><td>2018-05-09</td><td>2018-03-31</td><td>10119</td></tr><tr><td>"AEE"</td><td>2018-08-08</td><td>2018-06-30</td><td>12495</td></tr></tbody></table></div>"
|
||
],
|
||
"text/plain": [
|
||
"shape: (5, 4)\n",
|
||
"┌────────┬─────────────┬────────────┬────────────┐\n",
|
||
"│ symbol ┆ filing_date ┆ period_end ┆ word_count │\n",
|
||
"│ --- ┆ --- ┆ --- ┆ --- │\n",
|
||
"│ str ┆ date ┆ date ┆ u32 │\n",
|
||
"╞════════╪═════════════╪════════════╪════════════╡\n",
|
||
"│ AEE ┆ 2017-05-05 ┆ 2017-03-31 ┆ 8986 │\n",
|
||
"│ AEE ┆ 2017-08-04 ┆ 2017-06-30 ┆ 11308 │\n",
|
||
"│ AEE ┆ 2017-11-03 ┆ 2017-09-30 ┆ 12928 │\n",
|
||
"│ AEE ┆ 2018-05-09 ┆ 2018-03-31 ┆ 10119 │\n",
|
||
"│ AEE ┆ 2018-08-08 ┆ 2018-06-30 ┆ 12495 │\n",
|
||
"└────────┴─────────────┴────────────┴────────────┘"
|
||
]
|
||
},
|
||
"execution_count": 4,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"from data import load_sp500_10q_mda\n",
|
||
"\n",
|
||
"filings = load_sp500_10q_mda()\n",
|
||
"\n",
|
||
"print(f\"Loaded {len(filings):,} MD&A sections from {filings['symbol'].n_unique()} companies\")\n",
|
||
"print(f\"Date range: {filings['filing_date'].min()} to {filings['filing_date'].max()}\")\n",
|
||
"\n",
|
||
"if MAX_SYMBOLS > 0:\n",
|
||
" top_symbols = (\n",
|
||
" filings.group_by(\"symbol\")\n",
|
||
" .len()\n",
|
||
" .sort(\"len\", descending=True)\n",
|
||
" .head(MAX_SYMBOLS)[\"symbol\"]\n",
|
||
" .to_list()\n",
|
||
" )\n",
|
||
" filings = filings.filter(pl.col(\"symbol\").is_in(top_symbols))\n",
|
||
" print(f\"Filtered to {MAX_SYMBOLS} symbols: {len(filings):,} filings\")\n",
|
||
"\n",
|
||
"if MAX_FILINGS > 0 and len(filings) > MAX_FILINGS:\n",
|
||
" filings = filings.sort([\"filing_date\", \"symbol\"]).head(MAX_FILINGS)\n",
|
||
" print(f\"Reduced to first {MAX_FILINGS} filings for test run\")\n",
|
||
"\n",
|
||
"# Compute MD&A word count from the canonical `text` column.\n",
|
||
"filings = filings.with_columns(pl.col(\"text\").str.split(\" \").list.len().alias(\"word_count\"))\n",
|
||
"\n",
|
||
"filings.head(5).select([\"symbol\", \"filing_date\", \"period_end\", \"word_count\"])"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 5,
|
||
"id": "b9f0cfef",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:09.131805Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:09.131704Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:09.316509Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:09.316047Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.187615,
|
||
"end_time": "2026-06-13T02:51:09.317170+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:09.129555+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"MD&A word count statistics:\n",
|
||
"shape: (9, 2)\n",
|
||
"┌────────────┬──────────────┐\n",
|
||
"│ statistic ┆ value │\n",
|
||
"│ --- ┆ --- │\n",
|
||
"│ str ┆ f64 │\n",
|
||
"╞════════════╪══════════════╡\n",
|
||
"│ count ┆ 766.0 │\n",
|
||
"│ null_count ┆ 0.0 │\n",
|
||
"│ mean ┆ 10158.275457 │\n",
|
||
"│ std ┆ 7127.274687 │\n",
|
||
"│ min ┆ 1643.0 │\n",
|
||
"│ 25% ┆ 6008.0 │\n",
|
||
"│ 50% ┆ 8541.0 │\n",
|
||
"│ 75% ┆ 12016.0 │\n",
|
||
"│ max ┆ 66208.0 │\n",
|
||
"└────────────┴──────────────┘\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 1200x400 with 2 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Word count distribution\n",
|
||
"print(\"MD&A word count statistics:\")\n",
|
||
"print(filings[\"word_count\"].describe())\n",
|
||
"\n",
|
||
"fig, axes = plt.subplots(1, 2, figsize=(12, 4))\n",
|
||
"\n",
|
||
"axes[0].hist(filings[\"word_count\"].to_numpy(), bins=50, edgecolor=\"white\")\n",
|
||
"axes[0].set_xlabel(\"Word Count\")\n",
|
||
"axes[0].set_ylabel(\"Frequency\")\n",
|
||
"axes[0].set_title(\"MD&A Length Distribution\")\n",
|
||
"axes[0].axvline(filings[\"word_count\"].median(), color=\"red\", linestyle=\"--\", label=\"Median\")\n",
|
||
"axes[0].legend()\n",
|
||
"\n",
|
||
"# Filings per quarter\n",
|
||
"quarterly = (\n",
|
||
" filings.with_columns(\n",
|
||
" quarter=pl.col(\"filing_date\").dt.year().cast(pl.String)\n",
|
||
" + \"-Q\"\n",
|
||
" + pl.col(\"filing_date\").dt.quarter().cast(pl.String)\n",
|
||
" )\n",
|
||
" .group_by(\"quarter\")\n",
|
||
" .len()\n",
|
||
" .sort(\"quarter\")\n",
|
||
")\n",
|
||
"axes[1].bar(range(len(quarterly)), quarterly[\"len\"].to_numpy())\n",
|
||
"axes[1].set_xticks(range(0, len(quarterly), 4))\n",
|
||
"axes[1].set_xticklabels(quarterly[\"quarter\"].to_list()[::4], rotation=45)\n",
|
||
"axes[1].set_ylabel(\"Filings\")\n",
|
||
"axes[1].set_title(\"Filings per Quarter\")\n",
|
||
"\n",
|
||
"fig.tight_layout()\n",
|
||
"fig.show()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "de290418",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.001697,
|
||
"end_time": "2026-06-13T02:51:09.320843+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:09.319146+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 2. FinBERT Sentiment Scoring\n",
|
||
"\n",
|
||
"FinBERT processes text at the sentence level (max 512 tokens). For long MD&A sections\n",
|
||
"(median ~6,000 words), we use a chunking strategy:\n",
|
||
"\n",
|
||
"1. Split MD&A into sentences\n",
|
||
"2. Score each sentence with FinBERT (positive/negative/neutral probabilities)\n",
|
||
"3. Aggregate sentence scores to document-level sentiment\n",
|
||
"\n",
|
||
"This mirrors how analysts read filings: extracting overall tone from many paragraphs.\n",
|
||
"The aggregation captures both the **average sentiment** (management tone) and\n",
|
||
"**sentiment dispersion** (mixed signals within the same filing)."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 6,
|
||
"id": "80352fe3",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:09.325062Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:09.324971Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:21.393792Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:21.393404Z"
|
||
},
|
||
"lines_to_next_cell": 2,
|
||
"papermill": {
|
||
"duration": 12.071442,
|
||
"end_time": "2026-06-13T02:51:21.394115+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:09.322673+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Loading FinBERT: yiyanghkust/finbert-tone\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "30d42629908a44da9b09ab75d8bc0ba4",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"config.json: 0%| | 0.00/533 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "5bb5b12a9c1d4da6ad58cc4b4d811367",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"vocab.txt: 0.00B [00:00, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "b0bc8eca19704189b4ff8e8b0e631fa9",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"pytorch_model.bin: 0%| | 0.00/439M [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Device set to use cuda\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"FinBERT loaded\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline, set_seed\n",
|
||
"\n",
|
||
"set_seed(SEED)\n",
|
||
"print(f\"Loading FinBERT: {SENTIMENT_MODEL}\")\n",
|
||
"tokenizer = AutoTokenizer.from_pretrained(SENTIMENT_MODEL)\n",
|
||
"model = AutoModelForSequenceClassification.from_pretrained(SENTIMENT_MODEL)\n",
|
||
"model = model.to(device)\n",
|
||
"model.eval()\n",
|
||
"\n",
|
||
"sentiment_pipeline = pipeline(\n",
|
||
" \"sentiment-analysis\",\n",
|
||
" model=model,\n",
|
||
" tokenizer=tokenizer,\n",
|
||
" device=device,\n",
|
||
" truncation=True,\n",
|
||
" max_length=MAX_TOKENS,\n",
|
||
" batch_size=BATCH_SIZE,\n",
|
||
")\n",
|
||
"print(\"FinBERT loaded\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "50f1a7c2",
|
||
"metadata": {
|
||
"lines_to_next_cell": 2,
|
||
"papermill": {
|
||
"duration": 0.00183,
|
||
"end_time": "2026-06-13T02:51:21.398016+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:21.396186+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"### Chunking Strategy\n",
|
||
"\n",
|
||
"MD&A sections average ~6,000 words but FinBERT accepts only 512 tokens (~380 words).\n",
|
||
"We split into paragraphs and score each one, then aggregate."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 7,
|
||
"id": "abd712c0",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:21.402835Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:21.402644Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:21.405508Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:21.405017Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.00595,
|
||
"end_time": "2026-06-13T02:51:21.405747+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:21.399797+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"def chunk_text(text: str, max_chars: int = 1500) -> list[str]:\n",
|
||
" \"\"\"Split text into chunks suitable for FinBERT (roughly 512 tokens each).\"\"\"\n",
|
||
" paragraphs = [p.strip() for p in text.split(\"\\n\\n\") if len(p.strip()) > 50]\n",
|
||
" if not paragraphs:\n",
|
||
" # Fall back to sentence splitting\n",
|
||
" paragraphs = [s.strip() + \".\" for s in text.split(\".\") if len(s.strip()) > 30]\n",
|
||
"\n",
|
||
" chunks = []\n",
|
||
" current = \"\"\n",
|
||
" for para in paragraphs:\n",
|
||
" if len(current) + len(para) > max_chars and current:\n",
|
||
" chunks.append(current.strip())\n",
|
||
" current = para\n",
|
||
" else:\n",
|
||
" current = current + \"\\n\\n\" + para if current else para\n",
|
||
"\n",
|
||
" if current.strip():\n",
|
||
" chunks.append(current.strip())\n",
|
||
"\n",
|
||
" return chunks if chunks else [text[:max_chars]]"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 8,
|
||
"id": "2d664b74",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:21.410091Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:21.410007Z",
|
||
"iopub.status.idle": "2026-06-13T02:51:21.412364Z",
|
||
"shell.execute_reply": "2026-06-13T02:51:21.412057Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.005012,
|
||
"end_time": "2026-06-13T02:51:21.412590+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:21.407578+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"def score_document_sentiment(text: str) -> dict:\n",
|
||
" \"\"\"Score full MD&A document by aggregating chunk-level FinBERT predictions.\"\"\"\n",
|
||
" chunks = chunk_text(text)\n",
|
||
" if not chunks:\n",
|
||
" return {\"sentiment_mean\": 0.0, \"sentiment_std\": 0.0, \"n_chunks\": 0}\n",
|
||
"\n",
|
||
" # Score all chunks\n",
|
||
" results = sentiment_pipeline(chunks)\n",
|
||
"\n",
|
||
" # Convert labels to numeric: positive=+1, neutral=0, negative=-1\n",
|
||
" label_map = {\"Positive\": 1.0, \"Neutral\": 0.0, \"Negative\": -1.0}\n",
|
||
" scores = []\n",
|
||
" for r in results:\n",
|
||
" label = r[\"label\"]\n",
|
||
" confidence = r[\"score\"]\n",
|
||
" numeric = label_map.get(label, 0.0) * confidence\n",
|
||
" scores.append(numeric)\n",
|
||
"\n",
|
||
" scores_arr = np.array(scores)\n",
|
||
" return {\n",
|
||
" \"sentiment_mean\": float(scores_arr.mean()),\n",
|
||
" \"sentiment_std\": float(scores_arr.std()) if len(scores_arr) > 1 else 0.0,\n",
|
||
" \"sentiment_pos_pct\": float((scores_arr > 0).mean()),\n",
|
||
" \"sentiment_neg_pct\": float((scores_arr < 0).mean()),\n",
|
||
" \"n_chunks\": len(chunks),\n",
|
||
" }"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 9,
|
||
"id": "90494074",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:51:21.416731Z",
|
||
"iopub.status.busy": "2026-06-13T02:51:21.416665Z",
|
||
"iopub.status.idle": "2026-06-13T02:54:57.349564Z",
|
||
"shell.execute_reply": "2026-06-13T02:54:57.349091Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 215.935481,
|
||
"end_time": "2026-06-13T02:54:57.349891+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:51:21.414410+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Scoring 766 MD&A sections with FinBERT...\n",
|
||
"(This may take several minutes depending on GPU/CPU)\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "8da32d94f7bc48bda9bb6470d69d5b1f",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"model.safetensors: 0%| | 0.00/439M [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"You seem to be using the pipelines sequentially on GPU. In order to maximize efficiency please use a dataset\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 100/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 200/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 300/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 400/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 500/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 600/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 700/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Scored 766/766 filings\n",
|
||
"\n",
|
||
"Sentiment scoring complete: 766 filings scored\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"text/html": [
|
||
"<div><style>\n",
|
||
".dataframe > thead > tr,\n",
|
||
".dataframe > tbody > tr {\n",
|
||
" text-align: right;\n",
|
||
" white-space: pre-wrap;\n",
|
||
"}\n",
|
||
"</style>\n",
|
||
"<small>shape: (5, 7)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>sentiment_mean</th><th>sentiment_std</th><th>sentiment_pos_pct</th><th>sentiment_neg_pct</th><th>n_chunks</th><th>symbol</th><th>filing_date</th></tr><tr><td>f64</td><td>f64</td><td>f64</td><td>f64</td><td>i64</td><td>str</td><td>date</td></tr></thead><tbody><tr><td>0.057667</td><td>0.398735</td><td>0.115385</td><td>0.057692</td><td>52</td><td>"AEE"</td><td>2017-05-05</td></tr><tr><td>0.080763</td><td>0.411882</td><td>0.129032</td><td>0.048387</td><td>62</td><td>"AEE"</td><td>2017-08-04</td></tr><tr><td>0.051018</td><td>0.455604</td><td>0.150685</td><td>0.09589</td><td>73</td><td>"AEE"</td><td>2017-11-03</td></tr><tr><td>0.109907</td><td>0.438118</td><td>0.166667</td><td>0.05</td><td>60</td><td>"AEE"</td><td>2018-05-09</td></tr><tr><td>0.0589</td><td>0.44085</td><td>0.150685</td><td>0.082192</td><td>73</td><td>"AEE"</td><td>2018-08-08</td></tr></tbody></table></div>"
|
||
],
|
||
"text/plain": [
|
||
"shape: (5, 7)\n",
|
||
"┌────────────────┬───────────────┬───────────────┬───────────────┬──────────┬────────┬─────────────┐\n",
|
||
"│ sentiment_mean ┆ sentiment_std ┆ sentiment_pos ┆ sentiment_neg ┆ n_chunks ┆ symbol ┆ filing_date │\n",
|
||
"│ --- ┆ --- ┆ _pct ┆ _pct ┆ --- ┆ --- ┆ --- │\n",
|
||
"│ f64 ┆ f64 ┆ --- ┆ --- ┆ i64 ┆ str ┆ date │\n",
|
||
"│ ┆ ┆ f64 ┆ f64 ┆ ┆ ┆ │\n",
|
||
"╞════════════════╪═══════════════╪═══════════════╪═══════════════╪══════════╪════════╪═════════════╡\n",
|
||
"│ 0.057667 ┆ 0.398735 ┆ 0.115385 ┆ 0.057692 ┆ 52 ┆ AEE ┆ 2017-05-05 │\n",
|
||
"│ 0.080763 ┆ 0.411882 ┆ 0.129032 ┆ 0.048387 ┆ 62 ┆ AEE ┆ 2017-08-04 │\n",
|
||
"│ 0.051018 ┆ 0.455604 ┆ 0.150685 ┆ 0.09589 ┆ 73 ┆ AEE ┆ 2017-11-03 │\n",
|
||
"│ 0.109907 ┆ 0.438118 ┆ 0.166667 ┆ 0.05 ┆ 60 ┆ AEE ┆ 2018-05-09 │\n",
|
||
"│ 0.0589 ┆ 0.44085 ┆ 0.150685 ┆ 0.082192 ┆ 73 ┆ AEE ┆ 2018-08-08 │\n",
|
||
"└────────────────┴───────────────┴───────────────┴───────────────┴──────────┴────────┴─────────────┘"
|
||
]
|
||
},
|
||
"execution_count": 9,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Score all filings\n",
|
||
"print(f\"Scoring {len(filings):,} MD&A sections with FinBERT...\")\n",
|
||
"print(\"(This may take several minutes depending on GPU/CPU)\")\n",
|
||
"\n",
|
||
"sentiment_records = []\n",
|
||
"for i, row in enumerate(filings.iter_rows(named=True)):\n",
|
||
" scores = score_document_sentiment(row[\"text\"])\n",
|
||
" scores[\"symbol\"] = row[\"symbol\"]\n",
|
||
" scores[\"filing_date\"] = row[\"filing_date\"]\n",
|
||
" sentiment_records.append(scores)\n",
|
||
"\n",
|
||
" if (i + 1) % 100 == 0 or (i + 1) == len(filings):\n",
|
||
" print(f\" Scored {i + 1:,}/{len(filings):,} filings\")\n",
|
||
"\n",
|
||
"sentiment_df = pl.DataFrame(sentiment_records)\n",
|
||
"print(f\"\\nSentiment scoring complete: {len(sentiment_df):,} filings scored\")\n",
|
||
"sentiment_df.head(5)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 10,
|
||
"id": "67634b27",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:54:57.354691Z",
|
||
"iopub.status.busy": "2026-06-13T02:54:57.354616Z",
|
||
"iopub.status.idle": "2026-06-13T02:54:57.636075Z",
|
||
"shell.execute_reply": "2026-06-13T02:54:57.635620Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.284473,
|
||
"end_time": "2026-06-13T02:54:57.636537+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:54:57.352064+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 1500x400 with 3 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Sentiment distribution\n",
|
||
"fig, axes = plt.subplots(1, 3, figsize=(15, 4))\n",
|
||
"\n",
|
||
"axes[0].hist(sentiment_df[\"sentiment_mean\"].to_numpy(), bins=50, edgecolor=\"white\")\n",
|
||
"axes[0].set_xlabel(\"Mean Sentiment Score\")\n",
|
||
"axes[0].set_ylabel(\"Frequency\")\n",
|
||
"axes[0].set_title(\"MD&A Sentiment Distribution\")\n",
|
||
"axes[0].axvline(0, color=\"red\", linestyle=\"--\", alpha=0.5)\n",
|
||
"\n",
|
||
"axes[1].hist(sentiment_df[\"sentiment_std\"].to_numpy(), bins=50, edgecolor=\"white\")\n",
|
||
"axes[1].set_xlabel(\"Sentiment Std Dev\")\n",
|
||
"axes[1].set_title(\"Within-Filing Sentiment Dispersion\")\n",
|
||
"\n",
|
||
"axes[2].hist(\n",
|
||
" sentiment_df[\"sentiment_pos_pct\"].to_numpy(),\n",
|
||
" bins=30,\n",
|
||
" edgecolor=\"white\",\n",
|
||
" alpha=0.7,\n",
|
||
" label=\"Positive %\",\n",
|
||
")\n",
|
||
"axes[2].hist(\n",
|
||
" sentiment_df[\"sentiment_neg_pct\"].to_numpy(),\n",
|
||
" bins=30,\n",
|
||
" edgecolor=\"white\",\n",
|
||
" alpha=0.7,\n",
|
||
" label=\"Negative %\",\n",
|
||
")\n",
|
||
"axes[2].set_xlabel(\"Fraction of Chunks\")\n",
|
||
"axes[2].set_title(\"Positive vs Negative Chunk Fractions\")\n",
|
||
"axes[2].legend()\n",
|
||
"\n",
|
||
"fig.tight_layout()\n",
|
||
"fig.show()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "eb1c3315",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.004011,
|
||
"end_time": "2026-06-13T02:54:57.643707+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:54:57.639696+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 3. Document Embeddings and Narrative Change\n",
|
||
"\n",
|
||
"Beyond sentiment polarity, we capture **semantic content** using sentence-transformer\n",
|
||
"embeddings. The key signal is **narrative change**: how much the MD&A text shifts\n",
|
||
"from one quarter to the next.\n",
|
||
"\n",
|
||
"Intuition: a large semantic shift between consecutive filings suggests material\n",
|
||
"new information that the market may not have fully priced. This is analogous to\n",
|
||
"the \"news surprise\" factor in NB07, but applied to corporate disclosures."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 11,
|
||
"id": "4f7f46f3",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:54:57.648786Z",
|
||
"iopub.status.busy": "2026-06-13T02:54:57.648698Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:00.973148Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:00.972581Z"
|
||
},
|
||
"lines_to_next_cell": 2,
|
||
"papermill": {
|
||
"duration": 3.327674,
|
||
"end_time": "2026-06-13T02:55:00.973712+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:54:57.646038+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Loading embedding model: sentence-transformers/all-MiniLM-L6-v2\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "d0d927c6722746759fc47fde3d630940",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"modules.json: 0%| | 0.00/349 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "6800b40cfcd14f3aa11e6f6178a0b95b",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"config_sentence_transformers.json: 0%| | 0.00/116 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "f757974de24f4caa840617c9a7811dbb",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"README.md: 0.00B [00:00, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "8ab43d2c8d224c40a032b413b6ff4005",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"sentence_bert_config.json: 0%| | 0.00/53.0 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "1f1ab0a806c541058600646f04679323",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"config.json: 0%| | 0.00/612 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "b89a948cb97a49a6b7d511872e4a24b0",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"model.safetensors: 0%| | 0.00/90.9M [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "37f767c10e18439298b7e55ec09d1ef3",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"tokenizer_config.json: 0%| | 0.00/350 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "3663b30b0236459080df07facda4abbd",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"vocab.txt: 0.00B [00:00, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "4098c1c3d5e7438b8263dd26c22cba6e",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"tokenizer.json: 0.00B [00:00, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "f2c61446d2a140939dea99718207d9ab",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"special_tokens_map.json: 0%| | 0.00/112 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "e85c64b4c8b14b45b56b23c5b04bde80",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"config.json: 0%| | 0.00/190 [00:00<?, ?B/s]"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Embedding dimension: 384\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"from sentence_transformers import SentenceTransformer\n",
|
||
"\n",
|
||
"print(f\"Loading embedding model: {EMBEDDING_MODEL}\")\n",
|
||
"embed_model = SentenceTransformer(EMBEDDING_MODEL, device=str(device))\n",
|
||
"print(f\"Embedding dimension: {embed_model.get_sentence_embedding_dimension()}\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "0fe8627a",
|
||
"metadata": {
|
||
"lines_to_next_cell": 2,
|
||
"papermill": {
|
||
"duration": 0.002427,
|
||
"end_time": "2026-06-13T02:55:00.978801+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:00.976374+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"### Document Embedding Strategy\n",
|
||
"\n",
|
||
"Full MD&A texts are too long for a single embedding pass. We use **mean pooling\n",
|
||
"over chunk embeddings**: embed each paragraph/chunk, then average. This captures\n",
|
||
"the overall semantic content while respecting model token limits."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 12,
|
||
"id": "fbf78f57",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:00.984372Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:00.984148Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:00.986340Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:00.986014Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.00547,
|
||
"end_time": "2026-06-13T02:55:00.986621+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:00.981151+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"def embed_document(text: str, model: SentenceTransformer) -> np.ndarray:\n",
|
||
" \"\"\"Compute document embedding by mean-pooling chunk embeddings.\"\"\"\n",
|
||
" chunks = chunk_text(text, max_chars=1200)\n",
|
||
" if not chunks:\n",
|
||
" return np.zeros(model.get_sentence_embedding_dimension())\n",
|
||
"\n",
|
||
" chunk_embeddings = model.encode(chunks, show_progress_bar=False, batch_size=32)\n",
|
||
" return chunk_embeddings.mean(axis=0)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 13,
|
||
"id": "058b665c",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:00.992255Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:00.992171Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.116013Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.115419Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 47.127474,
|
||
"end_time": "2026-06-13T02:55:48.116626+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:00.989152+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Computing embeddings for 766 filings...\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 100/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 200/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 300/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 400/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 500/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 600/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 700/766 filings\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Embedded 766/766 filings\n",
|
||
"Embedding matrix: (766, 384)\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"# Compute embeddings for all filings\n",
|
||
"print(f\"Computing embeddings for {len(filings):,} filings...\")\n",
|
||
"\n",
|
||
"embeddings = []\n",
|
||
"for i, row in enumerate(filings.iter_rows(named=True)):\n",
|
||
" emb = embed_document(row[\"text\"], embed_model)\n",
|
||
" embeddings.append(emb)\n",
|
||
"\n",
|
||
" if (i + 1) % 100 == 0 or (i + 1) == len(filings):\n",
|
||
" print(f\" Embedded {i + 1:,}/{len(filings):,} filings\")\n",
|
||
"\n",
|
||
"embeddings_array = np.stack(embeddings)\n",
|
||
"print(f\"Embedding matrix: {embeddings_array.shape}\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 14,
|
||
"id": "ac32737f",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.122683Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.122544Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.132386Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.132046Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.013561,
|
||
"end_time": "2026-06-13T02:55:48.132977+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.119416+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Narrative change computed for 716 filings\n",
|
||
" (first filing per company has no prior quarter for comparison)\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"text/html": [
|
||
"<div><style>\n",
|
||
".dataframe > thead > tr,\n",
|
||
".dataframe > tbody > tr {\n",
|
||
" text-align: right;\n",
|
||
" white-space: pre-wrap;\n",
|
||
"}\n",
|
||
"</style>\n",
|
||
"<small>shape: (9, 2)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>statistic</th><th>value</th></tr><tr><td>str</td><td>f64</td></tr></thead><tbody><tr><td>"count"</td><td>716.0</td></tr><tr><td>"null_count"</td><td>0.0</td></tr><tr><td>"mean"</td><td>0.017853</td></tr><tr><td>"std"</td><td>0.015828</td></tr><tr><td>"min"</td><td>0.0</td></tr><tr><td>"25%"</td><td>0.007205</td></tr><tr><td>"50%"</td><td>0.013039</td></tr><tr><td>"75%"</td><td>0.023264</td></tr><tr><td>"max"</td><td>0.107054</td></tr></tbody></table></div>"
|
||
],
|
||
"text/plain": [
|
||
"shape: (9, 2)\n",
|
||
"┌────────────┬──────────┐\n",
|
||
"│ statistic ┆ value │\n",
|
||
"│ --- ┆ --- │\n",
|
||
"│ str ┆ f64 │\n",
|
||
"╞════════════╪══════════╡\n",
|
||
"│ count ┆ 716.0 │\n",
|
||
"│ null_count ┆ 0.0 │\n",
|
||
"│ mean ┆ 0.017853 │\n",
|
||
"│ std ┆ 0.015828 │\n",
|
||
"│ min ┆ 0.0 │\n",
|
||
"│ 25% ┆ 0.007205 │\n",
|
||
"│ 50% ┆ 0.013039 │\n",
|
||
"│ 75% ┆ 0.023264 │\n",
|
||
"│ max ┆ 0.107054 │\n",
|
||
"└────────────┴──────────┘"
|
||
]
|
||
},
|
||
"execution_count": 14,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Compute quarter-over-quarter narrative change (cosine distance)\n",
|
||
"# For each filing, compare its embedding to the previous quarter's filing for the same company\n",
|
||
"\n",
|
||
"# Sort by symbol and filing date\n",
|
||
"filing_order = (\n",
|
||
" filings.select([\"symbol\", \"filing_date\"]).with_row_index(\"idx\").sort([\"symbol\", \"filing_date\"])\n",
|
||
")\n",
|
||
"\n",
|
||
"narrative_changes = []\n",
|
||
"prev_emb_by_symbol = {}\n",
|
||
"\n",
|
||
"for row in filing_order.iter_rows(named=True):\n",
|
||
" idx = row[\"idx\"]\n",
|
||
" symbol = row[\"symbol\"]\n",
|
||
" emb = embeddings_array[idx]\n",
|
||
"\n",
|
||
" if symbol in prev_emb_by_symbol:\n",
|
||
" prev_emb = prev_emb_by_symbol[symbol]\n",
|
||
" # Cosine distance (0 = identical, 2 = opposite)\n",
|
||
" cos_sim = np.dot(emb, prev_emb) / (np.linalg.norm(emb) * np.linalg.norm(prev_emb) + 1e-8)\n",
|
||
" cos_dist = 1.0 - cos_sim\n",
|
||
" else:\n",
|
||
" cos_dist = None # No previous quarter\n",
|
||
"\n",
|
||
" narrative_changes.append(\n",
|
||
" {\n",
|
||
" \"symbol\": symbol,\n",
|
||
" \"filing_date\": row[\"filing_date\"],\n",
|
||
" \"narrative_change\": cos_dist,\n",
|
||
" }\n",
|
||
" )\n",
|
||
" prev_emb_by_symbol[symbol] = emb\n",
|
||
"\n",
|
||
"narrative_df = pl.DataFrame(narrative_changes)\n",
|
||
"print(\n",
|
||
" f\"Narrative change computed for {narrative_df.drop_nulls('narrative_change').height:,} filings\"\n",
|
||
")\n",
|
||
"print(\" (first filing per company has no prior quarter for comparison)\")\n",
|
||
"\n",
|
||
"narrative_df.drop_nulls(\"narrative_change\")[\"narrative_change\"].describe()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 15,
|
||
"id": "6e22458c",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.139543Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.139416Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.254439Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.253945Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.11863,
|
||
"end_time": "2026-06-13T02:55:48.254899+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.136269+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 800x400 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Distribution of narrative change\n",
|
||
"valid_changes = narrative_df.drop_nulls(\"narrative_change\")[\"narrative_change\"].to_numpy()\n",
|
||
"\n",
|
||
"fig, ax = plt.subplots(figsize=(8, 4))\n",
|
||
"ax.hist(valid_changes, bins=50, edgecolor=\"white\")\n",
|
||
"ax.set_xlabel(\"Cosine Distance (Quarter-over-Quarter)\")\n",
|
||
"ax.set_ylabel(\"Frequency\")\n",
|
||
"ax.set_title(\"MD&A Narrative Change Distribution\")\n",
|
||
"ax.axvline(\n",
|
||
" np.median(valid_changes),\n",
|
||
" color=\"red\",\n",
|
||
" linestyle=\"--\",\n",
|
||
" label=f\"Median: {np.median(valid_changes):.3f}\",\n",
|
||
")\n",
|
||
"ax.legend()\n",
|
||
"fig.tight_layout()\n",
|
||
"fig.show()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "09dea33e",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.002666,
|
||
"end_time": "2026-06-13T02:55:48.260777+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.258111+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 4. Combine Signals and Join to Market Data\n",
|
||
"\n",
|
||
"Two signal families are available:\n",
|
||
"- **Sentiment signals**: mean, std, positive/negative fractions\n",
|
||
"- **Narrative change**: cosine distance between consecutive filings\n",
|
||
"\n",
|
||
"We join these to AlgoSeek S&P 500 daily prices using the **filing_date** as the\n",
|
||
"point-in-time anchor. The signal becomes investable on the filing date itself\n",
|
||
"(SEC filings are public immediately upon acceptance)."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 16,
|
||
"id": "6128d537",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.266712Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.266627Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.271127Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.270803Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.00795,
|
||
"end_time": "2026-06-13T02:55:48.271362+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.263412+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Combined signals: (782, 8)\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"text/html": [
|
||
"<div><style>\n",
|
||
".dataframe > thead > tr,\n",
|
||
".dataframe > tbody > tr {\n",
|
||
" text-align: right;\n",
|
||
" white-space: pre-wrap;\n",
|
||
"}\n",
|
||
"</style>\n",
|
||
"<small>shape: (5, 8)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>sentiment_mean</th><th>sentiment_std</th><th>sentiment_pos_pct</th><th>sentiment_neg_pct</th><th>n_chunks</th><th>symbol</th><th>filing_date</th><th>narrative_change</th></tr><tr><td>f64</td><td>f64</td><td>f64</td><td>f64</td><td>i64</td><td>str</td><td>date</td><td>f64</td></tr></thead><tbody><tr><td>0.057667</td><td>0.398735</td><td>0.115385</td><td>0.057692</td><td>52</td><td>"AEE"</td><td>2017-05-05</td><td>null</td></tr><tr><td>0.080763</td><td>0.411882</td><td>0.129032</td><td>0.048387</td><td>62</td><td>"AEE"</td><td>2017-08-04</td><td>0.006356</td></tr><tr><td>0.051018</td><td>0.455604</td><td>0.150685</td><td>0.09589</td><td>73</td><td>"AEE"</td><td>2017-11-03</td><td>0.00282</td></tr><tr><td>0.109907</td><td>0.438118</td><td>0.166667</td><td>0.05</td><td>60</td><td>"AEE"</td><td>2018-05-09</td><td>0.00599</td></tr><tr><td>0.0589</td><td>0.44085</td><td>0.150685</td><td>0.082192</td><td>73</td><td>"AEE"</td><td>2018-08-08</td><td>0.003324</td></tr></tbody></table></div>"
|
||
],
|
||
"text/plain": [
|
||
"shape: (5, 8)\n",
|
||
"┌─────────────┬────────────┬────────────┬────────────┬──────────┬────────┬────────────┬────────────┐\n",
|
||
"│ sentiment_m ┆ sentiment_ ┆ sentiment_ ┆ sentiment_ ┆ n_chunks ┆ symbol ┆ filing_dat ┆ narrative_ │\n",
|
||
"│ ean ┆ std ┆ pos_pct ┆ neg_pct ┆ --- ┆ --- ┆ e ┆ change │\n",
|
||
"│ --- ┆ --- ┆ --- ┆ --- ┆ i64 ┆ str ┆ --- ┆ --- │\n",
|
||
"│ f64 ┆ f64 ┆ f64 ┆ f64 ┆ ┆ ┆ date ┆ f64 │\n",
|
||
"╞═════════════╪════════════╪════════════╪════════════╪══════════╪════════╪════════════╪════════════╡\n",
|
||
"│ 0.057667 ┆ 0.398735 ┆ 0.115385 ┆ 0.057692 ┆ 52 ┆ AEE ┆ 2017-05-05 ┆ null │\n",
|
||
"│ 0.080763 ┆ 0.411882 ┆ 0.129032 ┆ 0.048387 ┆ 62 ┆ AEE ┆ 2017-08-04 ┆ 0.006356 │\n",
|
||
"│ 0.051018 ┆ 0.455604 ┆ 0.150685 ┆ 0.09589 ┆ 73 ┆ AEE ┆ 2017-11-03 ┆ 0.00282 │\n",
|
||
"│ 0.109907 ┆ 0.438118 ┆ 0.166667 ┆ 0.05 ┆ 60 ┆ AEE ┆ 2018-05-09 ┆ 0.00599 │\n",
|
||
"│ 0.0589 ┆ 0.44085 ┆ 0.150685 ┆ 0.082192 ┆ 73 ┆ AEE ┆ 2018-08-08 ┆ 0.003324 │\n",
|
||
"└─────────────┴────────────┴────────────┴────────────┴──────────┴────────┴────────────┴────────────┘"
|
||
]
|
||
},
|
||
"execution_count": 16,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Merge sentiment and narrative change\n",
|
||
"signals = sentiment_df.join(\n",
|
||
" narrative_df.select([\"symbol\", \"filing_date\", \"narrative_change\"]),\n",
|
||
" on=[\"symbol\", \"filing_date\"],\n",
|
||
" how=\"left\",\n",
|
||
")\n",
|
||
"print(f\"Combined signals: {signals.shape}\")\n",
|
||
"signals.head(5)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 17,
|
||
"id": "b6589e58",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.277298Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.277206Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.357362Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.356947Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.084007,
|
||
"end_time": "2026-06-13T02:55:48.358129+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.274122+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Loaded 635,703 price observations for 638 symbols\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"# Load S&P 500 daily prices\n",
|
||
"from data import load_sp500_daily_bars\n",
|
||
"\n",
|
||
"prices = load_sp500_daily_bars()\n",
|
||
"print(f\"Loaded {len(prices):,} price observations for {prices['symbol'].n_unique()} symbols\")\n",
|
||
"\n",
|
||
"# Compute forward returns: return from day t to day t+N\n",
|
||
"price_returns = (\n",
|
||
" prices.sort([\"symbol\", \"timestamp\"])\n",
|
||
" .with_columns(\n",
|
||
" fwd_1d=(pl.col(\"close\").shift(-1) / pl.col(\"close\") - 1).over(\"symbol\"),\n",
|
||
" fwd_5d=(pl.col(\"close\").shift(-5) / pl.col(\"close\") - 1).over(\"symbol\"),\n",
|
||
" fwd_20d=(pl.col(\"close\").shift(-20) / pl.col(\"close\") - 1).over(\"symbol\"),\n",
|
||
" )\n",
|
||
" .select([\"symbol\", \"timestamp\", \"fwd_1d\", \"fwd_5d\", \"fwd_20d\"])\n",
|
||
")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 18,
|
||
"id": "ef5cd8a8",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.364588Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.364489Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.385636Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.385151Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.024822,
|
||
"end_time": "2026-06-13T02:55:48.385987+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.361165+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Evaluation dataset: 744 observations (50 symbols)\n",
|
||
"Date range: 2017-01-09 to 2021-12-02\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"# Join signals to prices using asof join\n",
|
||
"# Match each filing_date to the next trading day on or after that date\n",
|
||
"# This is PIT-correct: signal is available when the filing is accepted\n",
|
||
"\n",
|
||
"# Add trade_date column to prices so we can track which day was matched\n",
|
||
"prices_with_trade_date = price_returns.with_columns(trade_date=pl.col(\"timestamp\")).sort(\n",
|
||
" [\"symbol\", \"timestamp\"]\n",
|
||
")\n",
|
||
"\n",
|
||
"# Prepare signals: rename filing_date -> timestamp for the asof join key\n",
|
||
"signals_for_join = signals.rename({\"filing_date\": \"timestamp\"}).sort([\"symbol\", \"timestamp\"])\n",
|
||
"\n",
|
||
"# Asof join: for each signal date, find the nearest price date >= signal date\n",
|
||
"# strategy=\"forward\" means: match the next trading day on or after the signal date\n",
|
||
"eval_df = (\n",
|
||
" signals_for_join.join_asof(\n",
|
||
" prices_with_trade_date,\n",
|
||
" on=\"timestamp\",\n",
|
||
" by=\"symbol\",\n",
|
||
" strategy=\"forward\",\n",
|
||
" )\n",
|
||
" .rename({\"timestamp\": \"filing_date\"})\n",
|
||
" .drop_nulls([\"fwd_5d\"])\n",
|
||
")\n",
|
||
"\n",
|
||
"print(f\"Evaluation dataset: {len(eval_df):,} observations ({eval_df['symbol'].n_unique()} symbols)\")\n",
|
||
"print(f\"Date range: {eval_df['filing_date'].min()} to {eval_df['filing_date'].max()}\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "50e641b2",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.002802,
|
||
"end_time": "2026-06-13T02:55:48.391760+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.388958+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 5. Signal Evaluation: Information Coefficients\n",
|
||
"\n",
|
||
"We evaluate each signal using rank Information Coefficients (IC): the Spearman\n",
|
||
"correlation between signal values and subsequent returns. A good alpha factor\n",
|
||
"should show consistent, positive IC over time.\n",
|
||
"\n",
|
||
"The **IC** measures predictive power per cross-section (one date), while\n",
|
||
"**ICIR** (IC / std(IC)) measures consistency across dates."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 19,
|
||
"id": "04d3c2fc",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.398036Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.397903Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:48.399783Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:48.399510Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.005622,
|
||
"end_time": "2026-06-13T02:55:48.400153+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.394531+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"from scipy.stats import spearmanr\n",
|
||
"\n",
|
||
"signal_cols = [\n",
|
||
" \"sentiment_mean\",\n",
|
||
" \"sentiment_std\",\n",
|
||
" \"sentiment_pos_pct\",\n",
|
||
" \"sentiment_neg_pct\",\n",
|
||
" \"narrative_change\",\n",
|
||
"]\n",
|
||
"return_cols = [\"fwd_1d\", \"fwd_5d\", \"fwd_20d\"]"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "eae6322d",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.002678,
|
||
"end_time": "2026-06-13T02:55:48.405565+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.402887+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"We compute pooled Spearman ICs across all (filing, forward-return) pairs and\n",
|
||
"pair each pooled IC with a **cluster bootstrap** inference: resample whole\n",
|
||
"symbols (with replacement), recompute the pooled IC on each bootstrap\n",
|
||
"replicate, and report the 95% percentile interval and a two-sided bootstrap\n",
|
||
"p-value for the null IC=0. The cluster bootstrap requires a non-null cluster\n",
|
||
"id, so filings with a null `symbol` are dropped from `eval_df` before\n",
|
||
"computing both the point IC and the bootstrap; this shrinks `n_obs` per row\n",
|
||
"slightly versus the prior signal-only `drop_nulls` and is intentional.\n",
|
||
"\n",
|
||
"Why a cluster bootstrap on symbols? The pooled sample places the same firm\n",
|
||
"at multiple quarterly filings into one correlation. Returns are also\n",
|
||
"overlapping for the 5-day and 20-day horizons. The i.i.d. t-stat formula\n",
|
||
"`t = r·sqrt((n-2)/(1-r²))` would therefore overstate significance. Cluster\n",
|
||
"bootstrap on symbols preserves the within-firm dependence between filings\n",
|
||
"and overlapping returns. Treat the headline ICs as **screening** values;\n",
|
||
"the chapter's headline inference framework uses HAC on cross-sectional IC\n",
|
||
"series with adequate breadth (see NB07, NB08)."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 20,
|
||
"id": "a7571589",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:48.411507Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:48.411431Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:53.022334Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:53.021798Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 4.614466,
|
||
"end_time": "2026-06-13T02:55:53.022713+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:48.408247+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Signal Evaluation: Pooled ICs with cluster bootstrap (cluster=symbol)\n",
|
||
"======================================================================\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"shape: (15, 8)\n",
|
||
"┌───────────────────┬─────────┬─────────┬─────────┬─────────┬────────────────┬───────┬───────────┐\n",
|
||
"│ signal ┆ horizon ┆ ic ┆ ci95_lo ┆ ci95_hi ┆ p_cluster_boot ┆ n_obs ┆ n_symbols │\n",
|
||
"│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n",
|
||
"│ str ┆ str ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ i64 ┆ i64 │\n",
|
||
"╞═══════════════════╪═════════╪═════════╪═════════╪═════════╪════════════════╪═══════╪═══════════╡\n",
|
||
"│ narrative_change ┆ fwd_1d ┆ -0.1181 ┆ -0.193 ┆ -0.0446 ┆ 0.004 ┆ 693 ┆ 50 │\n",
|
||
"│ narrative_change ┆ fwd_20d ┆ 0.0506 ┆ -0.0188 ┆ 0.1218 ┆ 0.14 ┆ 692 ┆ 50 │\n",
|
||
"│ narrative_change ┆ fwd_5d ┆ -0.1496 ┆ -0.2283 ┆ -0.0719 ┆ 0.0 ┆ 693 ┆ 50 │\n",
|
||
"│ sentiment_mean ┆ fwd_1d ┆ -0.0698 ┆ -0.1374 ┆ -0.0026 ┆ 0.044 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_mean ┆ fwd_20d ┆ -0.1721 ┆ -0.2702 ┆ -0.0735 ┆ 0.0 ┆ 743 ┆ 50 │\n",
|
||
"│ sentiment_mean ┆ fwd_5d ┆ -0.0403 ┆ -0.1136 ┆ 0.0481 ┆ 0.37 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_neg_pct ┆ fwd_1d ┆ 0.0359 ┆ -0.04 ┆ 0.1146 ┆ 0.382 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_neg_pct ┆ fwd_20d ┆ 0.0743 ┆ -0.0157 ┆ 0.161 ┆ 0.092 ┆ 743 ┆ 50 │\n",
|
||
"│ sentiment_neg_pct ┆ fwd_5d ┆ 0.0027 ┆ -0.0839 ┆ 0.0881 ┆ 0.898 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_pos_pct ┆ fwd_1d ┆ -0.0507 ┆ -0.1184 ┆ 0.0118 ┆ 0.106 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_pos_pct ┆ fwd_20d ┆ -0.1392 ┆ -0.228 ┆ -0.0568 ┆ 0.0 ┆ 743 ┆ 50 │\n",
|
||
"│ sentiment_pos_pct ┆ fwd_5d ┆ -0.0345 ┆ -0.1086 ┆ 0.0432 ┆ 0.366 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_std ┆ fwd_1d ┆ -0.0188 ┆ -0.0862 ┆ 0.0475 ┆ 0.598 ┆ 744 ┆ 50 │\n",
|
||
"│ sentiment_std ┆ fwd_20d ┆ -0.0711 ┆ -0.1401 ┆ 0.0049 ┆ 0.068 ┆ 743 ┆ 50 │\n",
|
||
"│ sentiment_std ┆ fwd_5d ┆ -0.0472 ┆ -0.1113 ┆ 0.0239 ┆ 0.216 ┆ 744 ┆ 50 │\n",
|
||
"└───────────────────┴─────────┴─────────┴─────────┴─────────┴────────────────┴───────┴───────────┘\n",
|
||
"\n",
|
||
"Inference: cluster bootstrap with N_BOOT=1000 replicates; cluster = symbol.\n",
|
||
"Bootstrap CIs and p-values supersede the i.i.d. t-stat formula because\n",
|
||
"the pooled sample contains multiple filings per firm and overlapping\n",
|
||
"forward-return windows. Treat ICs as screening; HAC on cross-sectional\n",
|
||
"IC series (NB07, NB08) is the chapter's headline inference framework.\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"# Compute pooled ICs and cluster-bootstrap inference (cluster = symbol).\n",
|
||
"print(\"Signal Evaluation: Pooled ICs with cluster bootstrap (cluster=symbol)\")\n",
|
||
"print(\"=\" * 70)\n",
|
||
"\n",
|
||
"N_BOOT = 1000\n",
|
||
"# Single RNG shared across all (signal, horizon) iterations. Reproducibility\n",
|
||
"# of the table therefore depends on the iteration order of `signal_cols` ×\n",
|
||
"# `return_cols` — reorder either list and every subsequent pair gets a\n",
|
||
"# different bootstrap draw. If fewer than MIN_VALID_BOOT of N_BOOT replicates\n",
|
||
"# remain valid after NaN-drop, the percentile/p-value is suppressed (NaN) so\n",
|
||
"# unstable estimates from degenerate draws (e.g., a draw where one cluster\n",
|
||
"# dominates) do not leak into the table.\n",
|
||
"MIN_VALID_BOOT = 200\n",
|
||
"_boot_rng = np.random.default_rng(SEED)\n",
|
||
"\n",
|
||
"\n",
|
||
"def _pooled_spearman(sig_vals: np.ndarray, ret_vals: np.ndarray) -> float:\n",
|
||
" \"\"\"Pooled Spearman correlation, robust to constant-signal slices.\"\"\"\n",
|
||
" if len(sig_vals) < 5:\n",
|
||
" return np.nan\n",
|
||
" if np.std(sig_vals) == 0 or np.std(ret_vals) == 0:\n",
|
||
" return np.nan\n",
|
||
" return float(spearmanr(sig_vals, ret_vals)[0])\n",
|
||
"\n",
|
||
"\n",
|
||
"ic_results = []\n",
|
||
"for sig in signal_cols:\n",
|
||
" for ret in return_cols:\n",
|
||
" valid = eval_df.select([\"symbol\", sig, ret]).drop_nulls()\n",
|
||
" if valid.height < 20:\n",
|
||
" continue\n",
|
||
" symbols = valid[\"symbol\"].to_numpy()\n",
|
||
" sig_vals = valid[sig].to_numpy()\n",
|
||
" ret_vals = valid[ret].to_numpy()\n",
|
||
" n = len(sig_vals)\n",
|
||
"\n",
|
||
" ic_point = _pooled_spearman(sig_vals, ret_vals)\n",
|
||
"\n",
|
||
" # Cluster bootstrap: resample whole symbols with replacement.\n",
|
||
" unique_symbols = np.unique(symbols)\n",
|
||
" idx_by_symbol = {s: np.where(symbols == s)[0] for s in unique_symbols}\n",
|
||
" boot_ics = np.empty(N_BOOT)\n",
|
||
" for b in range(N_BOOT):\n",
|
||
" drawn = _boot_rng.choice(unique_symbols, size=len(unique_symbols), replace=True)\n",
|
||
" indices = np.concatenate([idx_by_symbol[s] for s in drawn])\n",
|
||
" boot_ics[b] = _pooled_spearman(sig_vals[indices], ret_vals[indices])\n",
|
||
" boot_ics = boot_ics[~np.isnan(boot_ics)]\n",
|
||
"\n",
|
||
" if boot_ics.size >= MIN_VALID_BOOT:\n",
|
||
" ci_lo = float(np.percentile(boot_ics, 2.5))\n",
|
||
" ci_hi = float(np.percentile(boot_ics, 97.5))\n",
|
||
" # Two-sided bootstrap p-value via percentile-method test inversion:\n",
|
||
" # the smallest α such that the (1-α) bootstrap CI excludes zero.\n",
|
||
" # Not a recentered-around-zero reflection test — for that, swap\n",
|
||
" # in `mean(|boot - boot.mean()| >= |ic_point|)`.\n",
|
||
" p_boot = 2.0 * min(\n",
|
||
" float(np.mean(boot_ics <= 0.0)),\n",
|
||
" float(np.mean(boot_ics >= 0.0)),\n",
|
||
" )\n",
|
||
" else:\n",
|
||
" ci_lo = ci_hi = p_boot = np.nan\n",
|
||
"\n",
|
||
" ic_results.append(\n",
|
||
" {\n",
|
||
" \"signal\": sig,\n",
|
||
" \"horizon\": ret,\n",
|
||
" \"ic\": round(ic_point, 4) if not np.isnan(ic_point) else np.nan,\n",
|
||
" \"ci95_lo\": round(ci_lo, 4) if not np.isnan(ci_lo) else np.nan,\n",
|
||
" \"ci95_hi\": round(ci_hi, 4) if not np.isnan(ci_hi) else np.nan,\n",
|
||
" \"p_cluster_boot\": round(p_boot, 4) if not np.isnan(p_boot) else np.nan,\n",
|
||
" \"n_obs\": n,\n",
|
||
" \"n_symbols\": int(len(unique_symbols)),\n",
|
||
" }\n",
|
||
" )\n",
|
||
"\n",
|
||
"if ic_results:\n",
|
||
" ic_summary = pl.DataFrame(ic_results).sort([\"signal\", \"horizon\"])\n",
|
||
" print(ic_summary)\n",
|
||
" print(\n",
|
||
" \"\\nInference: cluster bootstrap with N_BOOT=1000 replicates; cluster = symbol.\\n\"\n",
|
||
" \"Bootstrap CIs and p-values supersede the i.i.d. t-stat formula because\\n\"\n",
|
||
" \"the pooled sample contains multiple filings per firm and overlapping\\n\"\n",
|
||
" \"forward-return windows. Treat ICs as screening; HAC on cross-sectional\\n\"\n",
|
||
" \"IC series (NB07, NB08) is the chapter's headline inference framework.\"\n",
|
||
" )\n",
|
||
"else:\n",
|
||
" print(\"Insufficient data for IC computation (need more symbols/filings)\")\n",
|
||
" ic_summary = pl.DataFrame()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "07e5f4ee",
|
||
"metadata": {
|
||
"lines_to_next_cell": 2,
|
||
"papermill": {
|
||
"duration": 0.002778,
|
||
"end_time": "2026-06-13T02:55:53.028580+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.025802+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 6. Quintile Analysis\n",
|
||
"\n",
|
||
"Beyond IC, we examine whether signals produce economically meaningful return spreads\n",
|
||
"by sorting stocks into quintiles based on each signal and comparing average returns.\n",
|
||
"A strong signal should show a monotonic relationship between quintile rank and\n",
|
||
"subsequent returns."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 21,
|
||
"id": "9deb77c0",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:53.034907Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:53.034804Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:53.037447Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:53.037018Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.006446,
|
||
"end_time": "2026-06-13T02:55:53.037922+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.031476+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"def quintile_analysis(df: pl.DataFrame, signal_col: str, return_col: str) -> pl.DataFrame:\n",
|
||
" \"\"\"Sort into quintiles by signal, compute average return per quintile.\"\"\"\n",
|
||
" valid = df.drop_nulls([signal_col, return_col])\n",
|
||
" if valid.height < 25: # Need at least 5 per quintile\n",
|
||
" return pl.DataFrame()\n",
|
||
"\n",
|
||
" # Assign quintiles using qcut on the signal column\n",
|
||
" result = valid.with_columns(\n",
|
||
" quintile=pl.col(signal_col).qcut(5, labels=[\"Q1\", \"Q2\", \"Q3\", \"Q4\", \"Q5\"])\n",
|
||
" )\n",
|
||
"\n",
|
||
" # Average return per quintile\n",
|
||
" summary = (\n",
|
||
" result.group_by(\"quintile\")\n",
|
||
" .agg(\n",
|
||
" avg_return=pl.col(return_col).mean(),\n",
|
||
" std_return=pl.col(return_col).std(),\n",
|
||
" n_obs=pl.col(return_col).len(),\n",
|
||
" )\n",
|
||
" .sort(\"quintile\")\n",
|
||
" )\n",
|
||
"\n",
|
||
" return summary"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 22,
|
||
"id": "ac01eeaa",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:53.043913Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:53.043838Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:53.180987Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:53.180528Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.140749,
|
||
"end_time": "2026-06-13T02:55:53.181404+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.040655+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 1200x500 with 2 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"# Quintile analysis for key signals\n",
|
||
"key_signals = [\"sentiment_mean\", \"narrative_change\"]\n",
|
||
"fig, axes = plt.subplots(1, len(key_signals), figsize=(6 * len(key_signals), 5))\n",
|
||
"if len(key_signals) == 1:\n",
|
||
" axes = [axes]\n",
|
||
"\n",
|
||
"for i, sig in enumerate(key_signals):\n",
|
||
" ax = axes[i]\n",
|
||
" q_df = quintile_analysis(eval_df, sig, \"fwd_20d\")\n",
|
||
"\n",
|
||
" if q_df.height > 0:\n",
|
||
" quintiles = q_df[\"quintile\"].to_list()\n",
|
||
" returns = q_df[\"avg_return\"].to_numpy()\n",
|
||
" colors = [\"#d32f2f\" if r < 0 else \"#2e7d32\" for r in returns]\n",
|
||
"\n",
|
||
" ax.bar(range(len(quintiles)), returns * 100, color=colors, edgecolor=\"white\")\n",
|
||
" ax.set_xticks(range(len(quintiles)))\n",
|
||
" ax.set_xticklabels(quintiles)\n",
|
||
" ax.set_ylabel(\"Average 20-day Return (%)\")\n",
|
||
" ax.set_title(f\"{sig}: Quintile Returns\")\n",
|
||
" ax.axhline(0, color=\"black\", linewidth=0.5)\n",
|
||
"\n",
|
||
" # Long-short spread\n",
|
||
" if len(returns) >= 2:\n",
|
||
" spread = returns[-1] - returns[0]\n",
|
||
" ax.text(\n",
|
||
" 0.95,\n",
|
||
" 0.95,\n",
|
||
" f\"Q5-Q1: {spread * 100:.2f}%\",\n",
|
||
" transform=ax.transAxes,\n",
|
||
" ha=\"right\",\n",
|
||
" va=\"top\",\n",
|
||
" fontsize=10,\n",
|
||
" bbox=dict(boxstyle=\"round\", facecolor=\"wheat\", alpha=0.8),\n",
|
||
" )\n",
|
||
" else:\n",
|
||
" ax.text(0.5, 0.5, \"Insufficient data\", transform=ax.transAxes, ha=\"center\")\n",
|
||
" ax.set_title(f\"{sig}: Quintile Returns\")\n",
|
||
"\n",
|
||
"fig.tight_layout()\n",
|
||
"fig.show()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "9bd16033",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.002986,
|
||
"end_time": "2026-06-13T02:55:53.187544+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.184558+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## 7. Save Signals\n",
|
||
"\n",
|
||
"Save the computed signals for potential downstream use."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 23,
|
||
"id": "f63f73d7",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:53.194034Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:53.193927Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:53.198991Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:53.198567Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.008954,
|
||
"end_time": "2026-06-13T02:55:53.199337+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.190383+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Saved 744 signal observations to 10_text_feature_engineering/output/filing_signals/filing_signals.parquet\n",
|
||
"Saved IC summary to 10_text_feature_engineering/output/filing_signals/ic_summary.parquet\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"# Save evaluation dataset\n",
|
||
"eval_df.write_parquet(OUTPUT_DIR / \"filing_signals.parquet\")\n",
|
||
"print(f\"Saved {len(eval_df):,} signal observations to {OUTPUT_DIR / 'filing_signals.parquet'}\")\n",
|
||
"\n",
|
||
"# Save IC summary\n",
|
||
"if len(ic_summary) > 0:\n",
|
||
" ic_summary.write_parquet(OUTPUT_DIR / \"ic_summary.parquet\")\n",
|
||
" print(f\"Saved IC summary to {OUTPUT_DIR / 'ic_summary.parquet'}\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "0f45c369",
|
||
"metadata": {
|
||
"papermill": {
|
||
"duration": 0.002914,
|
||
"end_time": "2026-06-13T02:55:53.205307+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.202393+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"source": [
|
||
"## Key Takeaways\n",
|
||
"\n",
|
||
"1. **SEC filings provide dense, structured text** with natural PIT anchoring via filing dates.\n",
|
||
" MD&A sections average ~6,000 words per quarter — far richer than news headlines.\n",
|
||
"\n",
|
||
"2. **Chunking is essential for transformer models** that have 512-token limits.\n",
|
||
" Mean-pooled paragraph-level scores approximate full-document analysis.\n",
|
||
"\n",
|
||
"3. **Two complementary signal types emerge from the same text**:\n",
|
||
" - *Sentiment* captures directional management tone (optimistic vs cautious)\n",
|
||
" - *Narrative change* captures information novelty (quarter-over-quarter semantic shift)\n",
|
||
"\n",
|
||
"4. **Signal evaluation is screening-grade**. The pooled Spearman ICs above\n",
|
||
" are reported with cluster bootstrap (cluster=symbol) CIs and p-values\n",
|
||
" rather than the i.i.d. t-stat formula, because the pooled sample\n",
|
||
" contains multiple filings per firm and overlapping forward-return\n",
|
||
" windows. Treat the magnitudes as a screen; chapter-headline inference\n",
|
||
" uses HAC on per-date cross-sectional IC series (NB07, NB08), which\n",
|
||
" requires adequate breadth per date.\n",
|
||
"\n",
|
||
"5. **Filing signals complement headline signals** (NB07). News captures market reaction\n",
|
||
" speed; filings capture management's own assessment of financial condition.\n",
|
||
"\n",
|
||
"**Next**: See NB07/08 for news-based signal construction and evaluation.\n",
|
||
"**Book**: Section 10.5 discusses the full pre-train → adapt → fine-tune cascade\n",
|
||
"and signal validation protocol for production deployment."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 24,
|
||
"id": "e6da7dd8",
|
||
"metadata": {
|
||
"execution": {
|
||
"iopub.execute_input": "2026-06-13T02:55:53.211884Z",
|
||
"iopub.status.busy": "2026-06-13T02:55:53.211783Z",
|
||
"iopub.status.idle": "2026-06-13T02:55:53.214422Z",
|
||
"shell.execute_reply": "2026-06-13T02:55:53.213989Z"
|
||
},
|
||
"papermill": {
|
||
"duration": 0.006533,
|
||
"end_time": "2026-06-13T02:55:53.214696+00:00",
|
||
"exception": false,
|
||
"start_time": "2026-06-13T02:55:53.208163+00:00",
|
||
"status": "completed"
|
||
}
|
||
},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\n",
|
||
"======================================================================\n",
|
||
"NOTEBOOK COMPLETE: SEC Filing Text Signals\n",
|
||
"======================================================================\n",
|
||
"\n",
|
||
"Signals computed:\n",
|
||
" - sentiment_mean: FinBERT paragraph-level sentiment (mean across chunks)\n",
|
||
" - sentiment_std: Within-filing sentiment dispersion\n",
|
||
" - sentiment_pos_pct / sentiment_neg_pct: Fraction of positive/negative chunks\n",
|
||
" - narrative_change: Cosine distance to prior quarter's MD&A embedding\n",
|
||
"\n",
|
||
"Evaluation dataset: 744 filing-date observations\n",
|
||
"Symbols: 50\n",
|
||
"Date range: 2017-01-09 to 2021-12-02\n",
|
||
"Output: 10_text_feature_engineering/output/filing_signals\n",
|
||
"\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"print(\"\\n\" + \"=\" * 70)\n",
|
||
"print(\"NOTEBOOK COMPLETE: SEC Filing Text Signals\")\n",
|
||
"print(\"=\" * 70)\n",
|
||
"print(f\"\"\"\n",
|
||
"Signals computed:\n",
|
||
" - sentiment_mean: FinBERT paragraph-level sentiment (mean across chunks)\n",
|
||
" - sentiment_std: Within-filing sentiment dispersion\n",
|
||
" - sentiment_pos_pct / sentiment_neg_pct: Fraction of positive/negative chunks\n",
|
||
" - narrative_change: Cosine distance to prior quarter's MD&A embedding\n",
|
||
"\n",
|
||
"Evaluation dataset: {len(eval_df):,} filing-date observations\n",
|
||
"Symbols: {eval_df[\"symbol\"].n_unique()}\n",
|
||
"Date range: {eval_df[\"filing_date\"].min()} to {eval_df[\"filing_date\"].max()}\n",
|
||
"Output: {OUTPUT_DIR}\n",
|
||
"\"\"\")"
|
||
]
|
||
}
|
||
],
|
||
"metadata": {
|
||
"kernelspec": {
|
||
"display_name": "Python 3 (ipykernel)",
|
||
"language": "python",
|
||
"name": "python3"
|
||
},
|
||
"language_info": {
|
||
"codemirror_mode": {
|
||
"name": "ipython",
|
||
"version": 3
|
||
},
|
||
"file_extension": ".py",
|
||
"mimetype": "text/x-python",
|
||
"name": "python",
|
||
"nbconvert_exporter": "python",
|
||
"pygments_lexer": "ipython3",
|
||
"version": "3.14.5"
|
||
},
|
||
"papermill": {
|
||
"default_parameters": {},
|
||
"duration": 291.117902,
|
||
"end_time": "2026-06-13T02:55:55.934814+00:00",
|
||
"environment_variables": {},
|
||
"exception": null,
|
||
"input_path": "10_text_feature_engineering/09_filing_text_signals.ipynb",
|
||
"output_path": "10_text_feature_engineering/09_filing_text_signals.ipynb",
|
||
"parameters": {},
|
||
"start_time": "2026-06-13T02:51:04.816912+00:00",
|
||
"version": "2.7.0"
|
||
},
|
||
"widgets": {
|
||
"application/vnd.jupyter.widget-state+json": {
|
||
"state": {
|
||
"041bea3d3f97499cae14405735d220c9": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_cccb035e21b14c5baf1a4723971fd596",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_ac31941f9dcf473da8fe1246a70744bd",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 232k/? [00:00<00:00, 35.3MB/s]"
|
||
}
|
||
},
|
||
"05bf7d756db543f6b73966b7f2a5700e": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_0b48fff756f243c69fab86c8ed82f458",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_1f528c04091b417cb6e025e6c31fc626",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 349/349 [00:00<00:00, 107kB/s]"
|
||
}
|
||
},
|
||
"0a5f9b391a9642f1a111f45f0fa86839": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"0a9a86036827452c976a7da0bdf3e57a": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"0b48fff756f243c69fab86c8ed82f458": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"0c827a3eaba646fdb757cb1083297c5c": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"0e9351543fb74964963a90ed206b05fb": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"103642927efe402f94ff695bccfb3d43": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"10f206020bd846b4ac0e74436171fb7d": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"13ca2b35791e4d82885bc5497c5c0fe7": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"151b410c2ed247a781b9f35f66af68c8": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_c1fe7dee4a714c1eb06a12fda678c6ec",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_c1806c93132f40f79ad6424a5924f7eb",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "config.json: 100%"
|
||
}
|
||
},
|
||
"157192d643d94f3fbe841f255ce1c63c": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"18ccd19228294873bdf02038e7a0c7ae": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_3399a275f3b74d7dad414b15a9e6c47b",
|
||
"max": 53.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_86fb20cbb43f4fdf8f721f980fa0bf22",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 53.0
|
||
}
|
||
},
|
||
"1a16547c83c34628a4389fcaa811150d": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"1c04ad2d704043d4a9ede47941c36e44": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"1cd410b2397b43538fff763e422767ff": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"1cde3a11abea486198cb59a150923b50": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"1d01f72f08d44979a9871705d63d1a84": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"1e33be888533477bbdcb1da2c99e169d": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_5a886b59cc6b483ca3aa53fca557ae80",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_2caf791917784d0899ae376abc7a1b6a",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 612/612 [00:00<00:00, 300kB/s]"
|
||
}
|
||
},
|
||
"1ed9e3c4e76945ba95b8bb4ea6cf9812": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": "20px"
|
||
}
|
||
},
|
||
"1f1ab0a806c541058600646f04679323": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_c1ab4c4749bb4ab8a8ddba6ec3beacf9",
|
||
"IPY_MODEL_f8712c90b52743bab73ebf0607d46111",
|
||
"IPY_MODEL_1e33be888533477bbdcb1da2c99e169d"
|
||
],
|
||
"layout": "IPY_MODEL_e28f73edee9a4f97ade18febf04cc39e",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"1f528c04091b417cb6e025e6c31fc626": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"2107cbd2a74c46f4ae72650fdcbb8245": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_a8ebc3141609490192161ed96160d0d2",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_c97cd9f91a3d43f5aa299e6eb50c93d8",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "special_tokens_map.json: 100%"
|
||
}
|
||
},
|
||
"243a571ede2547e4af8c9564c3996b7b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"27fd2ea99773482b93c50969c4c5bd20": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_ce489db414d04a74bf64c5ec7258464f",
|
||
"max": 90868376.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_a922715981554dccb1700b46cedf7581",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 90868376.0
|
||
}
|
||
},
|
||
"29e998a62af14781996bcfe913bedfec": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"2caf791917784d0899ae376abc7a1b6a": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"2cf4d3ad9f254347906f55419f203796": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"2dad658cc78942d18eb40d162f69ee5a": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"30d42629908a44da9b09ab75d8bc0ba4": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_151b410c2ed247a781b9f35f66af68c8",
|
||
"IPY_MODEL_fea3a1d379374bd8a17be342f6831019",
|
||
"IPY_MODEL_6b3898058d334870bf50218f12ca778f"
|
||
],
|
||
"layout": "IPY_MODEL_33072f85638144edace6fff248bd4564",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"33072f85638144edace6fff248bd4564": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"3399a275f3b74d7dad414b15a9e6c47b": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"33de0be0bfa34647b678a7fdc1e28fcd": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"3663b30b0236459080df07facda4abbd": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_3ce8e7d6c33d465bbbee6d38fceda5dc",
|
||
"IPY_MODEL_830b4f506f4e4d9dbabd2447916cec87",
|
||
"IPY_MODEL_041bea3d3f97499cae14405735d220c9"
|
||
],
|
||
"layout": "IPY_MODEL_396220cab490484691710fd3db3ffe26",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"36aa21f9987b4dfeb3e15856958e2652": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"37f767c10e18439298b7e55ec09d1ef3": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_c416a7c370f9411faff88a4251096533",
|
||
"IPY_MODEL_c8df645eb9754a0eaa04fd025f875486",
|
||
"IPY_MODEL_534e4dd686d846dabc9ef8b6bcfa661d"
|
||
],
|
||
"layout": "IPY_MODEL_ce173c1b249a46a297cc77fa4ad31cc5",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"3896cf409c9a4740a499b9d4bf9fc6ed": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"396220cab490484691710fd3db3ffe26": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"3b953d35cb2c4cd1920c5089d708d75a": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"3c8afb886aec46fc997818cadd573c14": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": "20px"
|
||
}
|
||
},
|
||
"3c961458b28848b68589195d20999deb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"3ce8e7d6c33d465bbbee6d38fceda5dc": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_917d0cff0d544b2bb06bcea5bca54582",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_f532695830674d029919cad4cd3821ee",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "vocab.txt: "
|
||
}
|
||
},
|
||
"3eaa006192eb43e0915f50f9c8903134": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"3f54980476c941c090bfbf1086b46108": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_59bc5136b65743538147fdccddac1c5c",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_82a5c33eada14a3f812ec3459dc45a0d",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 439M/439M [00:08<00:00, 50.1MB/s]"
|
||
}
|
||
},
|
||
"40148fcc47af4098b427bd904df1c677": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"4098c1c3d5e7438b8263dd26c22cba6e": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_86728d2336d747cda9e8a595b70250f8",
|
||
"IPY_MODEL_9c5d93f0ad7d4b74bb6bc7315b3a5090",
|
||
"IPY_MODEL_b8a2eac4e5b14eaeb3513b573e7f2a88"
|
||
],
|
||
"layout": "IPY_MODEL_41dbc446839046be9bde9e462aec6ff3",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"41dbc446839046be9bde9e462aec6ff3": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"44252facc4a847ebb54c86494311564a": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"468539c90f2b44ffa1e2ac0607b004c4": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"4bf29fbbf313467ebce20df8e6a5773c": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"4d7a2e459864496a9f2127505568fa6b": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"52db467ead144274ae757919a0667d45": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"534e4dd686d846dabc9ef8b6bcfa661d": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_b54b433e525c4489b7f6b50ece608332",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_94ea0077a4494ae59681ee28ad1c1d5c",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 350/350 [00:00<00:00, 160kB/s]"
|
||
}
|
||
},
|
||
"5699d951e5b94eef9d36df852159d12c": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"58723d4d7ca24c9c9241009b75cc21ee": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"59bc5136b65743538147fdccddac1c5c": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"5a886b59cc6b483ca3aa53fca557ae80": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"5b56e9d94ce14194b298472357270f99": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"5bb5b12a9c1d4da6ad58cc4b4d811367": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_baf681d28081466cb65a8d4fffb754f4",
|
||
"IPY_MODEL_7a125061a39548b3bd0cb4fde286f9be",
|
||
"IPY_MODEL_8de2630c5d4146b881010c9e43538901"
|
||
],
|
||
"layout": "IPY_MODEL_def9192762a24618b4b04d3e86b83b1e",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"5bbce0f7fad84f2d84b8c379a97e68a2": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"610a4d9b264b45aba4bfd0e6100294ca": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_ebd54ff84bc6412f9fb3fa233524396a",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_96ae6a6e57184137ab16f5a34fa842c2",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "sentence_bert_config.json: 100%"
|
||
}
|
||
},
|
||
"627bee62dd694b5e8cc2e1727cb8f43a": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": "20px"
|
||
}
|
||
},
|
||
"6800b40cfcd14f3aa11e6f6178a0b95b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_c704f421077849a9b8450b65e81bc0ee",
|
||
"IPY_MODEL_dee6094526dc483188867d2fcd85cc55",
|
||
"IPY_MODEL_e99fea5eadf5423696fbc5c480f7de52"
|
||
],
|
||
"layout": "IPY_MODEL_fc6122155e944b569411930d47dd21f9",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"6b3898058d334870bf50218f12ca778f": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_1cd410b2397b43538fff763e422767ff",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_7331a68bdf5946bf9c481a7d789f7726",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 533/533 [00:00<00:00, 199kB/s]"
|
||
}
|
||
},
|
||
"6c16015abe394a528a2de7254cd29179": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_0e9351543fb74964963a90ed206b05fb",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_c06aa3791376446e8d4238af309d48f1",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "modules.json: 100%"
|
||
}
|
||
},
|
||
"6e0910ab432745c7aa8928cad0fd6ca4": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"70b863c1f28348d4a4f7955d0feb6382": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_40148fcc47af4098b427bd904df1c677",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_ccbd4437df9a4e4b834f26c37325cfd7",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 439M/439M [00:08<00:00, 74.6MB/s]"
|
||
}
|
||
},
|
||
"71a2d4f576a84db4b41dfee24bf77e34": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_3eaa006192eb43e0915f50f9c8903134",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_99cd082336644368a52150595fca7ae7",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "config.json: 100%"
|
||
}
|
||
},
|
||
"724f279f082843979ee4cb9c7f27bea3": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"7331a68bdf5946bf9c481a7d789f7726": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"771b1d43a186411eba8fbfdcd9299834": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_0c827a3eaba646fdb757cb1083297c5c",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_5b56e9d94ce14194b298472357270f99",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "README.md: "
|
||
}
|
||
},
|
||
"7a125061a39548b3bd0cb4fde286f9be": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_1ed9e3c4e76945ba95b8bb4ea6cf9812",
|
||
"max": 1.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_52db467ead144274ae757919a0667d45",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 1.0
|
||
}
|
||
},
|
||
"7d8cdc38d9bc4d57a7e9e308d0f073b9": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_d2246b0bf9104f2fb741a8a3ad40c293",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_bcf92c3ea7694247a462ad64aba8f70b",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 112/112 [00:00<00:00, 58.1kB/s]"
|
||
}
|
||
},
|
||
"82a5c33eada14a3f812ec3459dc45a0d": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"830b4f506f4e4d9dbabd2447916cec87": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_3c8afb886aec46fc997818cadd573c14",
|
||
"max": 1.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_2cf4d3ad9f254347906f55419f203796",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 1.0
|
||
}
|
||
},
|
||
"841c0a14ba464ff8bf592b3c25c32940": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_990be9ed1aea438c97b7c30062666876",
|
||
"max": 112.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_10f206020bd846b4ac0e74436171fb7d",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 112.0
|
||
}
|
||
},
|
||
"84ede3ed960a4242ba62db3ffa02ca28": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"86728d2336d747cda9e8a595b70250f8": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_ae932fe5c7324042a5ecf9164fe745f3",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_1cde3a11abea486198cb59a150923b50",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "tokenizer.json: "
|
||
}
|
||
},
|
||
"8690665dcd654a6da77eb7a7fd475544": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"86fb20cbb43f4fdf8f721f980fa0bf22": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"884405b31ef44760a0a3e7f9280c2169": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"8953e0b1018647e29aebfab6987c1a20": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"8ab43d2c8d224c40a032b413b6ff4005": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_610a4d9b264b45aba4bfd0e6100294ca",
|
||
"IPY_MODEL_18ccd19228294873bdf02038e7a0c7ae",
|
||
"IPY_MODEL_9a8180a4f81f4ca58815877afd419ec8"
|
||
],
|
||
"layout": "IPY_MODEL_dcc6ae9472a14c8190f5dea7040be0e1",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"8c52f82e78bd44079d3719ed899e5445": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_3b953d35cb2c4cd1920c5089d708d75a",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_157192d643d94f3fbe841f255ce1c63c",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 10.5k/? [00:00<00:00, 2.36MB/s]"
|
||
}
|
||
},
|
||
"8da32d94f7bc48bda9bb6470d69d5b1f": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_bd65752195c94262aac59688a17767ab",
|
||
"IPY_MODEL_d8f32400ceb4438b877639d2ea4e6515",
|
||
"IPY_MODEL_70b863c1f28348d4a4f7955d0feb6382"
|
||
],
|
||
"layout": "IPY_MODEL_efae760e0db347dfa50997bbeb6d2be5",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"8de2630c5d4146b881010c9e43538901": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_97f00d2f1eae423da7245df5dfe46d47",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_a33590ec5b0c4fd29911a31c215b6672",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 226k/? [00:00<00:00, 10.4MB/s]"
|
||
}
|
||
},
|
||
"917d0cff0d544b2bb06bcea5bca54582": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"9312d86bab8f45929368fdd69be285fc": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_c3412ebd5fd54f2c8bcf56385c4a4f6e",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_36aa21f9987b4dfeb3e15856958e2652",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "pytorch_model.bin: 100%"
|
||
}
|
||
},
|
||
"94ea0077a4494ae59681ee28ad1c1d5c": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"96ae6a6e57184137ab16f5a34fa842c2": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"97f00d2f1eae423da7245df5dfe46d47": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"97fd2c21d7e0407b8fa1ad51c6c12c1b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"98a34f6e3fc44b17bfd5645f9c53e501": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"990be9ed1aea438c97b7c30062666876": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"99cd082336644368a52150595fca7ae7": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"9a8180a4f81f4ca58815877afd419ec8": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_c21c003648ad4aedb23fb77817e2a73b",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_724f279f082843979ee4cb9c7f27bea3",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 53.0/53.0 [00:00<00:00, 14.8kB/s]"
|
||
}
|
||
},
|
||
"9c5d93f0ad7d4b74bb6bc7315b3a5090": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_627bee62dd694b5e8cc2e1727cb8f43a",
|
||
"max": 1.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_d8061368c28b45528fe99eb9b7b0eda7",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 1.0
|
||
}
|
||
},
|
||
"9c8b1daaa592434292d177835ef3568b": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"9d3ed7e8a032489f9edcbd7a59b315ee": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"a33590ec5b0c4fd29911a31c215b6672": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"a529059455ce49db8fc28679bb2e1442": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"a8ebc3141609490192161ed96160d0d2": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"a922715981554dccb1700b46cedf7581": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"ac31941f9dcf473da8fe1246a70744bd": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"ad03e03717884518bd51fc580d221912": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_a529059455ce49db8fc28679bb2e1442",
|
||
"max": 439101405.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_5699d951e5b94eef9d36df852159d12c",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 439101405.0
|
||
}
|
||
},
|
||
"ad5a109c605847329e07867a65066989": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"ae932fe5c7324042a5ecf9164fe745f3": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"af34708ea8214c328f2da90072713e8b": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"b0bc8eca19704189b4ff8e8b0e631fa9": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_9312d86bab8f45929368fdd69be285fc",
|
||
"IPY_MODEL_ad03e03717884518bd51fc580d221912",
|
||
"IPY_MODEL_3f54980476c941c090bfbf1086b46108"
|
||
],
|
||
"layout": "IPY_MODEL_d989ce561fa14b8d84d2f85ae7d2f03d",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"b54b433e525c4489b7f6b50ece608332": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"b89a948cb97a49a6b7d511872e4a24b0": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_c1de165dc8d441858cd39517fd412243",
|
||
"IPY_MODEL_27fd2ea99773482b93c50969c4c5bd20",
|
||
"IPY_MODEL_fabc148caf9f4b408a84c59aed12cd79"
|
||
],
|
||
"layout": "IPY_MODEL_6e0910ab432745c7aa8928cad0fd6ca4",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"b8a2eac4e5b14eaeb3513b573e7f2a88": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_af34708ea8214c328f2da90072713e8b",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_33de0be0bfa34647b678a7fdc1e28fcd",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 466k/? [00:00<00:00, 67.3MB/s]"
|
||
}
|
||
},
|
||
"badb2e9d3b8c43408f3adb55289fac8d": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"baf681d28081466cb65a8d4fffb754f4": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_fc6a5883be5f45cfbffb683f5623e832",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_f90b0f37182545cea89ebd55b185c231",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "vocab.txt: "
|
||
}
|
||
},
|
||
"bbfd1b3ce9104e86bf934ab18b051ab3": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"bcd18598ea364d69afaa4fb74e6d6fb7": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": "20px"
|
||
}
|
||
},
|
||
"bcf92c3ea7694247a462ad64aba8f70b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"bd65752195c94262aac59688a17767ab": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_58723d4d7ca24c9c9241009b75cc21ee",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_c422d82cb0a448c5a0b78b424ce9a682",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "model.safetensors: 100%"
|
||
}
|
||
},
|
||
"bff4338063bc4894953258ef71618099": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"c06aa3791376446e8d4238af309d48f1": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"c1806c93132f40f79ad6424a5924f7eb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"c1ab4c4749bb4ab8a8ddba6ec3beacf9": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_bbfd1b3ce9104e86bf934ab18b051ab3",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_13ca2b35791e4d82885bc5497c5c0fe7",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "config.json: 100%"
|
||
}
|
||
},
|
||
"c1de165dc8d441858cd39517fd412243": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_8953e0b1018647e29aebfab6987c1a20",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_243a571ede2547e4af8c9564c3996b7b",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "model.safetensors: 100%"
|
||
}
|
||
},
|
||
"c1fe7dee4a714c1eb06a12fda678c6ec": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"c21c003648ad4aedb23fb77817e2a73b": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"c3412ebd5fd54f2c8bcf56385c4a4f6e": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"c416a7c370f9411faff88a4251096533": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_badb2e9d3b8c43408f3adb55289fac8d",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_3c961458b28848b68589195d20999deb",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "tokenizer_config.json: 100%"
|
||
}
|
||
},
|
||
"c422d82cb0a448c5a0b78b424ce9a682": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"c704f421077849a9b8450b65e81bc0ee": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_bff4338063bc4894953258ef71618099",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_103642927efe402f94ff695bccfb3d43",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": "config_sentence_transformers.json: 100%"
|
||
}
|
||
},
|
||
"c8df645eb9754a0eaa04fd025f875486": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_ad5a109c605847329e07867a65066989",
|
||
"max": 350.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_0a5f9b391a9642f1a111f45f0fa86839",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 350.0
|
||
}
|
||
},
|
||
"c97cd9f91a3d43f5aa299e6eb50c93d8": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"cbe08d622d424897b7850e295574571e": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"ccbd4437df9a4e4b834f26c37325cfd7": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"cccb035e21b14c5baf1a4723971fd596": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"ce173c1b249a46a297cc77fa4ad31cc5": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"ce489db414d04a74bf64c5ec7258464f": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"d0d927c6722746759fc47fde3d630940": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_6c16015abe394a528a2de7254cd29179",
|
||
"IPY_MODEL_eb2532812f074b18a237a4330524ba9b",
|
||
"IPY_MODEL_05bf7d756db543f6b73966b7f2a5700e"
|
||
],
|
||
"layout": "IPY_MODEL_fdff8b6359674244993aed0a29cd772f",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"d2246b0bf9104f2fb741a8a3ad40c293": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"d8061368c28b45528fe99eb9b7b0eda7": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"d8abf141495041dc9ff47e03e9b71747": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_29e998a62af14781996bcfe913bedfec",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_98a34f6e3fc44b17bfd5645f9c53e501",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 190/190 [00:00<00:00, 79.4kB/s]"
|
||
}
|
||
},
|
||
"d8f32400ceb4438b877639d2ea4e6515": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_cbe08d622d424897b7850e295574571e",
|
||
"max": 439044180.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_884405b31ef44760a0a3e7f9280c2169",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 439044180.0
|
||
}
|
||
},
|
||
"d989ce561fa14b8d84d2f85ae7d2f03d": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"dcc6ae9472a14c8190f5dea7040be0e1": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"dee6094526dc483188867d2fcd85cc55": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_3896cf409c9a4740a499b9d4bf9fc6ed",
|
||
"max": 116.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_8690665dcd654a6da77eb7a7fd475544",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 116.0
|
||
}
|
||
},
|
||
"def9192762a24618b4b04d3e86b83b1e": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"e28f73edee9a4f97ade18febf04cc39e": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"e367aa02bab640d28f502929c2e453cb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"e85c64b4c8b14b45b56b23c5b04bde80": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_71a2d4f576a84db4b41dfee24bf77e34",
|
||
"IPY_MODEL_efde5658aac140ff8a1bb7c48d5ab304",
|
||
"IPY_MODEL_d8abf141495041dc9ff47e03e9b71747"
|
||
],
|
||
"layout": "IPY_MODEL_84ede3ed960a4242ba62db3ffa02ca28",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"e99fea5eadf5423696fbc5c480f7de52": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_9c8b1daaa592434292d177835ef3568b",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_5bbce0f7fad84f2d84b8c379a97e68a2",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 116/116 [00:00<00:00, 45.5kB/s]"
|
||
}
|
||
},
|
||
"eb2532812f074b18a237a4330524ba9b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_1d01f72f08d44979a9871705d63d1a84",
|
||
"max": 349.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_e367aa02bab640d28f502929c2e453cb",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 349.0
|
||
}
|
||
},
|
||
"ebd54ff84bc6412f9fb3fa233524396a": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"efae760e0db347dfa50997bbeb6d2be5": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"efde5658aac140ff8a1bb7c48d5ab304": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_fad7146fdaf44523b642440b5314ad1d",
|
||
"max": 190.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_97fd2c21d7e0407b8fa1ad51c6c12c1b",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 190.0
|
||
}
|
||
},
|
||
"f0e2a0fd17a24941b94453aef173d624": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_bcd18598ea364d69afaa4fb74e6d6fb7",
|
||
"max": 1.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_44252facc4a847ebb54c86494311564a",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 1.0
|
||
}
|
||
},
|
||
"f2c61446d2a140939dea99718207d9ab": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_2107cbd2a74c46f4ae72650fdcbb8245",
|
||
"IPY_MODEL_841c0a14ba464ff8bf592b3c25c32940",
|
||
"IPY_MODEL_7d8cdc38d9bc4d57a7e9e308d0f073b9"
|
||
],
|
||
"layout": "IPY_MODEL_2dad658cc78942d18eb40d162f69ee5a",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"f532695830674d029919cad4cd3821ee": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"f757974de24f4caa840617c9a7811dbb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_771b1d43a186411eba8fbfdcd9299834",
|
||
"IPY_MODEL_f0e2a0fd17a24941b94453aef173d624",
|
||
"IPY_MODEL_8c52f82e78bd44079d3719ed899e5445"
|
||
],
|
||
"layout": "IPY_MODEL_4d7a2e459864496a9f2127505568fa6b",
|
||
"tabbable": null,
|
||
"tooltip": null
|
||
}
|
||
},
|
||
"f8712c90b52743bab73ebf0607d46111": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_1a16547c83c34628a4389fcaa811150d",
|
||
"max": 612.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_4bf29fbbf313467ebce20df8e6a5773c",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 612.0
|
||
}
|
||
},
|
||
"f90b0f37182545cea89ebd55b185c231": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "StyleView",
|
||
"background": null,
|
||
"description_width": "",
|
||
"font_size": null,
|
||
"text_color": null
|
||
}
|
||
},
|
||
"fabc148caf9f4b408a84c59aed12cd79": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_0a9a86036827452c976a7da0bdf3e57a",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_9d3ed7e8a032489f9edcbd7a59b315ee",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": " 90.9M/90.9M [00:01<00:00, 70.1MB/s]"
|
||
}
|
||
},
|
||
"fad7146fdaf44523b642440b5314ad1d": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"fc6122155e944b569411930d47dd21f9": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"fc6a5883be5f45cfbffb683f5623e832": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"fdff8b6359674244993aed0a29cd772f": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border_bottom": null,
|
||
"border_left": null,
|
||
"border_right": null,
|
||
"border_top": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"fea3a1d379374bd8a17be342f6831019": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_module_version": "2.0.0",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "2.0.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "2.0.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_allow_html": false,
|
||
"layout": "IPY_MODEL_1c04ad2d704043d4a9ede47941c36e44",
|
||
"max": 533.0,
|
||
"min": 0.0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_468539c90f2b44ffa1e2ac0607b004c4",
|
||
"tabbable": null,
|
||
"tooltip": null,
|
||
"value": 533.0
|
||
}
|
||
}
|
||
},
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
}
|
||
},
|
||
"jupytext": {
|
||
"cell_metadata_filter": "tags,-all"
|
||
},
|
||
"ml4t_provenance": {
|
||
"source_py_blob": "1c8fdf459c94c6cb4a00e4dd191724194dd9b1b6",
|
||
"executed_at": "2026-06-13T02:55:56.389800+00:00",
|
||
"executor": "gpu-docker",
|
||
"production": true,
|
||
"parameters": {},
|
||
"notes": "executed-state finalization batch (2026-06-13 run)"
|
||
}
|
||
},
|
||
"nbformat": 4,
|
||
"nbformat_minor": 5
|
||
}
|