wehub-resource-sync
caf324b09d
tests / check_code_quality (push) Waiting to run
tests / tests (ubuntu-latest, 3.10) (push) Blocked by required conditions
tests / tests (ubuntu-latest, 3.11) (push) Blocked by required conditions
Deploy "method_comparison" Gradio to Spaces / deploy (push) Waiting to run
Deploy "PEFT shop" Gradio app to Spaces / deploy (push) Waiting to run
tests on transformers main / tests (push) Waiting to run
tests / tests (ubuntu-latest, 3.12) (push) Blocked by required conditions
tests / tests (ubuntu-latest, 3.13) (push) Blocked by required conditions
tests / tests (windows-latest, 3.10) (push) Blocked by required conditions
tests / tests (windows-latest, 3.11) (push) Blocked by required conditions
tests / tests (windows-latest, 3.12) (push) Blocked by required conditions
tests / tests (windows-latest, 3.13) (push) Blocked by required conditions
Secret Leaks / trufflehog (push) Waiting to run
CI security linting / zizmor latest via Cargo (push) Waiting to run
Build documentation / build (push) Failing after 0s
532 lines
16 KiB
Plaintext
532 lines
16 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "d36e1e93-ae93-4a4e-93c6-68fd868d2882",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Using VeRA for sequence classification"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "ddfc0610-55f6-4343-a950-125ccf0f45ac",
|
|
"metadata": {},
|
|
"source": [
|
|
"In this example, we fine-tune Roberta on a sequence classification task using VeRA."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "45addd81-d4f3-4dfd-960d-3920d347f0a6",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Imports"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"id": "a9935ae2",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import torch\n",
|
|
"from torch.optim import AdamW\n",
|
|
"from torch.utils.data import DataLoader\n",
|
|
"from peft import (\n",
|
|
" get_peft_model,\n",
|
|
" VeraConfig,\n",
|
|
" PeftType,\n",
|
|
")\n",
|
|
"\n",
|
|
"import evaluate\n",
|
|
"from datasets import load_dataset\n",
|
|
"from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed, AutoConfig\n",
|
|
"from tqdm import tqdm"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "62c959bf-7cc2-49e0-b97e-4c10ec3b9bf3",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Parameters"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"id": "e3b13308",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"<torch._C.Generator at 0x7b8246b7c850>"
|
|
]
|
|
},
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"batch_size = 128\n",
|
|
"model_name_or_path = \"roberta-base\"\n",
|
|
"task = \"mrpc\"\n",
|
|
"peft_type = PeftType.VERA\n",
|
|
"device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
|
|
"num_epochs = 5 # for best results, increase this number\n",
|
|
"rank = 8 # for best results, increase this number\n",
|
|
"max_length = 128\n",
|
|
"torch.manual_seed(0)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"id": "0526f571",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"peft_config = VeraConfig(\n",
|
|
" task_type=\"SEQ_CLS\", \n",
|
|
" r=rank,\n",
|
|
" d_initial=0.1,\n",
|
|
" target_modules=[\"query\", \"value\", \"intermediate.dense\"],\n",
|
|
" save_projection=True,\n",
|
|
")\n",
|
|
"head_lr = 1e-2\n",
|
|
"vera_lr = 2e-2"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "c075c5d2-a457-4f37-a7f1-94fd0d277972",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Loading data"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"id": "7bb52cb4-d1c3-4b04-8bf0-f39ca88af139",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"if any(k in model_name_or_path for k in (\"gpt\", \"opt\", \"bloom\")):\n",
|
|
" padding_side = \"left\"\n",
|
|
"else:\n",
|
|
" padding_side = \"right\"\n",
|
|
"\n",
|
|
"tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, padding_side=padding_side)\n",
|
|
"if getattr(tokenizer, \"pad_token_id\") is None:\n",
|
|
" tokenizer.pad_token_id = tokenizer.eos_token_id"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 5,
|
|
"id": "e69c5e1f-d27b-4264-a41e-fc9b99d025e6",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"datasets = load_dataset(\"glue\", task)\n",
|
|
"metric = evaluate.load(\"glue\", task)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 6,
|
|
"id": "0209f778-c93b-40eb-a4e0-24c25db03980",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def tokenize_function(examples):\n",
|
|
" # max_length=None => use the model max length (it's actually the default)\n",
|
|
" outputs = tokenizer(examples[\"sentence1\"], examples[\"sentence2\"], truncation=True, max_length=max_length)\n",
|
|
" return outputs\n",
|
|
"\n",
|
|
"\n",
|
|
"tokenized_datasets = datasets.map(\n",
|
|
" tokenize_function,\n",
|
|
" batched=True,\n",
|
|
" remove_columns=[\"idx\", \"sentence1\", \"sentence2\"],\n",
|
|
")\n",
|
|
"\n",
|
|
"# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the\n",
|
|
"# transformers library\n",
|
|
"tokenized_datasets = tokenized_datasets.rename_column(\"label\", \"labels\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 7,
|
|
"id": "7453954e-982c-46f0-b09c-589776e6d6cb",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def collate_fn(examples):\n",
|
|
" return tokenizer.pad(examples, padding=\"longest\", return_tensors=\"pt\")\n",
|
|
"\n",
|
|
"\n",
|
|
"# Instantiate dataloaders.\n",
|
|
"train_dataloader = DataLoader(tokenized_datasets[\"train\"], shuffle=True, collate_fn=collate_fn, batch_size=batch_size)\n",
|
|
"eval_dataloader = DataLoader(\n",
|
|
" tokenized_datasets[\"validation\"], shuffle=False, collate_fn=collate_fn, batch_size=batch_size\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "f3b9b2e8-f415-4d0f-9fb4-436f1a3585ea",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Preparing the VeRA model"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 8,
|
|
"id": "2ed5ac74",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Some weights of RobertaForSequenceClassification were not initialized from the model checkpoint at roberta-base and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight']\n",
|
|
"You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"trainable params: 647,714 || all params: 125,294,884 || trainable%: 0.5170\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"model = AutoModelForSequenceClassification.from_pretrained(model_name_or_path, return_dict=True, max_length=None)\n",
|
|
"model = get_peft_model(model, peft_config)\n",
|
|
"model.print_trainable_parameters()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 9,
|
|
"id": "0d2d0381",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"optimizer = AdamW(\n",
|
|
" [\n",
|
|
" {\"params\": [p for n, p in model.named_parameters() if \"vera_lambda_\" in n], \"lr\": vera_lr},\n",
|
|
" {\"params\": [p for n, p in model.named_parameters() if \"classifier\" in n], \"lr\": head_lr},\n",
|
|
" ]\n",
|
|
")\n",
|
|
"\n",
|
|
"# Instantiate scheduler\n",
|
|
"lr_scheduler = get_linear_schedule_with_warmup(\n",
|
|
" optimizer=optimizer,\n",
|
|
" num_warmup_steps=0.06 * (len(train_dataloader) * num_epochs),\n",
|
|
" num_training_steps=(len(train_dataloader) * num_epochs),\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "c0dd5aa8-977b-4ac0-8b96-884b17bcdd00",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Training"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 10,
|
|
"id": "fa0e73be",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
" 0%| | 0/29 [00:00<?, ?it/s]You're using a RobertaTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
|
|
"100%|██████████| 29/29 [00:18<00:00, 1.58it/s]\n",
|
|
"100%|██████████| 4/4 [00:01<00:00, 3.52it/s]\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"epoch 0: {'accuracy': 0.7475490196078431, 'f1': 0.8367670364500792}\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"100%|██████████| 29/29 [00:17<00:00, 1.68it/s]\n",
|
|
"100%|██████████| 4/4 [00:01<00:00, 3.37it/s]\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"epoch 1: {'accuracy': 0.7671568627450981, 'f1': 0.8536209553158706}\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"100%|██████████| 29/29 [00:17<00:00, 1.66it/s]\n",
|
|
"100%|██████████| 4/4 [00:01<00:00, 3.33it/s]\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"epoch 2: {'accuracy': 0.8553921568627451, 'f1': 0.8959435626102292}\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"100%|██████████| 29/29 [00:17<00:00, 1.64it/s]\n",
|
|
"100%|██████████| 4/4 [00:01<00:00, 3.35it/s]\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"epoch 3: {'accuracy': 0.8823529411764706, 'f1': 0.9133574007220215}\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"100%|██████████| 29/29 [00:17<00:00, 1.63it/s]\n",
|
|
"100%|██████████| 4/4 [00:01<00:00, 3.17it/s]"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"epoch 4: {'accuracy': 0.8897058823529411, 'f1': 0.9183303085299456}\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"model.to(device)\n",
|
|
"for epoch in range(num_epochs):\n",
|
|
" model.train()\n",
|
|
" for step, batch in enumerate(tqdm(train_dataloader)):\n",
|
|
" batch.to(device)\n",
|
|
" outputs = model(**batch)\n",
|
|
" loss = outputs.loss\n",
|
|
" loss.backward()\n",
|
|
" optimizer.step()\n",
|
|
" lr_scheduler.step()\n",
|
|
" optimizer.zero_grad()\n",
|
|
"\n",
|
|
" model.eval()\n",
|
|
" for step, batch in enumerate(tqdm(eval_dataloader)):\n",
|
|
" batch.to(device)\n",
|
|
" with torch.no_grad():\n",
|
|
" outputs = model(**batch)\n",
|
|
" predictions = outputs.logits.argmax(dim=-1)\n",
|
|
" predictions, references = predictions, batch[\"labels\"]\n",
|
|
" metric.add_batch(\n",
|
|
" predictions=predictions,\n",
|
|
" references=references,\n",
|
|
" )\n",
|
|
"\n",
|
|
" eval_metric = metric.compute()\n",
|
|
" print(f\"epoch {epoch}:\", eval_metric)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "f2b2caca",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Share adapters on the 🤗 Hub"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 11,
|
|
"id": "7b23af6f-cf6e-486f-9d10-0eada95b631f",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"account_id = ... # your Hugging Face Hub account ID"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 12,
|
|
"id": "990b3c93",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"CommitInfo(commit_url='https://huggingface.co/BenjaminB/roberta-large-peft-vera/commit/d720cdc67b97df9cd1453b14d3e0fb17efc8779b', commit_message='Upload model', commit_description='', oid='d720cdc67b97df9cd1453b14d3e0fb17efc8779b', pr_url=None, pr_revision=None, pr_num=None)"
|
|
]
|
|
},
|
|
"execution_count": 12,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"model.push_to_hub(f\"{account_id}/roberta-large-peft-vera\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "9d140b26",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Load adapters from the Hub\n",
|
|
"\n",
|
|
"You can also directly load adapters from the Hub using the commands below:"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 13,
|
|
"id": "c283e028-b349-46b0-a20e-cde0ee5fbd7b",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import torch\n",
|
|
"from peft import PeftModel, PeftConfig\n",
|
|
"from transformers import AutoTokenizer"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 14,
|
|
"id": "320b10a0-4ea8-4786-9f3c-4670019c6b18",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Some weights of RobertaForSequenceClassification were not initialized from the model checkpoint at roberta-base and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight']\n",
|
|
"You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"peft_model_id = f\"{account_id}/roberta-large-peft-vera\"\n",
|
|
"config = PeftConfig.from_pretrained(peft_model_id)\n",
|
|
"inference_model = AutoModelForSequenceClassification.from_pretrained(config.base_model_name_or_path)\n",
|
|
"tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 15,
|
|
"id": "b3a94049-bc01-4f2e-8cf9-66daf24a4402",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"# Load the Vera model\n",
|
|
"inference_model = PeftModel.from_pretrained(inference_model, peft_model_id)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 16,
|
|
"id": "bd919fef-4e9a-4dc5-a957-7b879cfc5d38",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
" 0%| | 0/4 [00:00<?, ?it/s]You're using a RobertaTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
|
|
"100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.14it/s]"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"{'accuracy': 0.8480392156862745, 'f1': 0.8938356164383561}\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"inference_model.to(device)\n",
|
|
"inference_model.eval()\n",
|
|
"for step, batch in enumerate(tqdm(eval_dataloader)):\n",
|
|
" batch.to(device)\n",
|
|
" with torch.no_grad():\n",
|
|
" outputs = inference_model(**batch)\n",
|
|
" predictions = outputs.logits.argmax(dim=-1)\n",
|
|
" predictions, references = predictions, batch[\"labels\"]\n",
|
|
" metric.add_batch(\n",
|
|
" predictions=predictions,\n",
|
|
" references=references,\n",
|
|
" )\n",
|
|
"\n",
|
|
"eval_metric = metric.compute()\n",
|
|
"print(eval_metric)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "13cc0125-a052-4e26-b7ff-af0f763be34e",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3 (ipykernel)",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"vscode": {
|
|
"interpreter": {
|
|
"hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49"
|
|
}
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
}
|