wehub/labmlai--annotated_deep_learning_paper_implementations
1
0
Fork 0
Code Issues Pull Requests Actions Deployments Models Agent Notes Packages Projects Releases Wiki Activity

chore: import upstream snapshot with attribution

Browse Source
This commit is contained in:
wehub-resource-sync
2026-07-13 12:19:01 +08:00
commit 3b90d1192f
2172 changed files with 594509 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
translate_cache/experiments/__init__.ja.json
+1
View File
@@ -0,0 +1 @@
{}
translate_cache/experiments/__init__.si.json
+1
View File
@@ -0,0 +1 @@
{}
translate_cache/experiments/__init__.zh.json
+1
View File
@@ -0,0 +1 @@
{}
translate_cache/experiments/arithmetic_dataset.ja.json
+53
View File
@@ -0,0 +1,53 @@
{
"<h2>Arithmetic Dataset</h2>\n<p>This creates arithmetic addition problems and solutions with workings. We&#x27;ve only implemented addition so far.</p>\n<p>It&#x27;s based on a character level tokenization.</p>\n": "<h2>\u7b97\u8853\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</h2>\n<p>\u3053\u308c\u306b\u3088\u308a\u3001\u7b97\u8853\u52a0\u7b97\u306e\u554f\u984c\u3068\u89e3\u6cd5\u304c\u751f\u6210\u3055\u308c\u307e\u3059\u3002\u4eca\u306e\u3068\u3053\u308d\u3001\u8ffd\u52a0\u3092\u5b9f\u88c5\u3057\u305f\u3060\u3051\u3067\u3059\u3002</p>\n<p>\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u306e\u30c8\u30fc\u30af\u30f3\u5316\u306b\u57fa\u3065\u3044\u3066\u3044\u307e\u3059\u3002</p>\n",
"<h2>Arithmetic Task Experiment Configurations</h2>\n": "<h2>\u7b97\u8853\u30bf\u30b9\u30af\u5b9f\u9a13\u69cb\u6210</h2>\n",
"<h3>Evaluation</h3>\n<p>We use the sampling function to evaluate the model on a set of problems</p>\n": "<h3>\u8a55\u4fa1</h3>\n<p>\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u95a2\u6570\u3092\u4f7f\u7528\u3057\u3066\u3001\u4e00\u9023\u306e\u554f\u984c\u306b\u3064\u3044\u3066\u30e2\u30c7\u30eb\u3092\u8a55\u4fa1\u3057\u307e\u3059\u3002</p>\n",
"<p> </p>\n": "<p></p>\n",
"<p> Code to test generated problems</p>\n": "<p>\u751f\u6210\u3055\u308c\u305f\u554f\u984c\u3092\u30c6\u30b9\u30c8\u3059\u308b\u30b3\u30fc\u30c9</p>\n",
"<p> Decode a list of token ids</p>\n": "<p>\u30c8\u30fc\u30af\u30f3 ID \u306e\u30ea\u30b9\u30c8\u3092\u30c7\u30b3\u30fc\u30c9\u3059\u308b</p>\n",
"<p> Encode a given string</p>\n": "<p>\u4e0e\u3048\u3089\u308c\u305f\u6587\u5b57\u5217\u3092\u30a8\u30f3\u30b3\u30fc\u30c9\u3059\u308b</p>\n",
"<p> Generate multiple problems and pack them into a sequence.</p>\n": "<p>\u8907\u6570\u306e\u554f\u984c\u3092\u751f\u6210\u3057\u3001\u305d\u308c\u3089\u3092\u30b7\u30fc\u30b1\u30f3\u30b9\u306b\u307e\u3068\u3081\u307e\u3059\u3002</p>\n",
"<p> Generates an integer with <span translate=no>_^_0_^_</span> number of digits</p>\n": "<p><span translate=no>_^_0_^_</span>\u6841\u6570\u306e\u6574\u6570\u3092\u751f\u6210\u3057\u307e\u3059</p>\n",
"<p> Generates the workings for <span translate=no>_^_0_^_</span>. For example for <span translate=no>_^_1_^_</span> it generates <span translate=no>_^_2_^_</span>.</p>\n": "<p>\u306e\u4f5c\u696d\u3092\u751f\u6210\u3057\u307e\u3059\u3002<span translate=no>_^_0_^_</span>\u305f\u3068\u3048\u3070<span translate=no>_^_1_^_</span>\u3001\u751f\u6210\u3059\u308b\u5834\u5408\u306a\u3069\u3067\u3059<span translate=no>_^_2_^_</span>\u3002</p>\n",
"<p> Get a input and target pair for auto-regressive modelling</p>\n": "<p>\u81ea\u5df1\u56de\u5e30\u30e2\u30c7\u30ea\u30f3\u30b0\u306e\u5165\u529b\u3068\u30bf\u30fc\u30b2\u30c3\u30c8\u306e\u30da\u30a2\u3092\u53d6\u5f97</p>\n",
"<p> Get arithmetic problem and answer. This is used for evaluation.</p>\n": "<p>\u7b97\u8853\u554f\u984c\u3092\u51fa\u3057\u3066\u3001\u7b54\u3048\u3092\u51fa\u3057\u3066\u304f\u3060\u3055\u3044\u3002\u3053\u308c\u306f\u8a55\u4fa1\u306b\u4f7f\u7528\u3055\u308c\u307e\u3059\u3002</p>\n",
"<p> Number of sequences per epoch</p>\n": "<p>\u30a8\u30dd\u30c3\u30af\u3042\u305f\u308a\u306e\u30b7\u30fc\u30b1\u30f3\u30b9\u6570</p>\n",
"<p> Training data loader</p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
"<p><em>This is based on code by <a href=\"https://twitter.com/gharik\">Georges Harik (@gharik)</a>.</em></p>\n": "<p><em><a href=\"https://twitter.com/gharik\">\u3053\u308c\u306f\u30b8\u30e7\u30eb\u30b8\u30e5\u30fb\u30cf\u30ea\u30af</a> (@gharik) \u306e\u30b3\u30fc\u30c9\u306b\u57fa\u3065\u3044\u3066\u3044\u307e\u3059\u3002</em></p>\n",
"<p>Add the next token to the input </p>\n": "<p>\u6b21\u306e\u30c8\u30fc\u30af\u30f3\u3092\u5165\u529b\u306b\u8ffd\u52a0\u3057\u307e\u3059</p>\n",
"<p>Character to token id </p>\n": "<p>\u6587\u5b57\u304b\u3089\u30c8\u30fc\u30af\u30f3 ID \u3078</p>\n",
"<p>Collect the problems only </p>\n": "<p>\u554f\u984c\u3060\u3051\u96c6\u3081\u3088\u3046</p>\n",
"<p>Count the number of correct answers </p>\n": "<p>\u6b63\u89e3\u306e\u6570\u3092\u6570\u3048\u308b</p>\n",
"<p>Create a dataset to generate problems </p>\n": "<p>\u554f\u984c\u3092\u751f\u6210\u3059\u308b\u305f\u3081\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u4f5c\u6210</p>\n",
"<p>Create a tensor with only the initial token </p>\n": "<p>\u6700\u521d\u306e\u30c8\u30fc\u30af\u30f3\u306e\u307f\u3067\u30c6\u30f3\u30bd\u30eb\u3092\u4f5c\u6210</p>\n",
"<p>Discard everything after the answer in the results </p>\n": "<p>\u7d50\u679c\u306e\u56de\u7b54\u306e\u5f8c\u306b\u7d9a\u304f\u3082\u306e\u306f\u3059\u3079\u3066\u7834\u68c4\u3057\u3066\u304f\u3060\u3055\u3044</p>\n",
"<p>Find which sequences have finished </p>\n": "<p>\u3069\u306e\u30b7\u30fc\u30b1\u30f3\u30b9\u304c\u7d42\u4e86\u3057\u305f\u304b\u8abf\u3079\u308b</p>\n",
"<p>Get a set of problems and answers </p>\n": "<p>\u4e00\u9023\u306e\u554f\u984c\u3068\u56de\u7b54\u3092\u5165\u624b</p>\n",
"<p>Get the answers </p>\n": "<p>\u7b54\u3048\u3092\u30b2\u30c3\u30c8</p>\n",
"<p>Get the model output </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u53d6\u5f97</p>\n",
"<p>Get the model prediction (greedy) </p>\n": "<p>\u30e2\u30c7\u30eb\u4e88\u6e2c\u3092\u53d6\u5f97 (\u6b32\u5f35\u308a)</p>\n",
"<p>Get the sampled results </p>\n": "<p>\u30b5\u30f3\u30d7\u30eb\u7d50\u679c\u3092\u53d6\u5f97</p>\n",
"<p>If all the sequences have completed we skip this </p>\n": "<p>\u3059\u3079\u3066\u306e\u30b7\u30fc\u30b1\u30f3\u30b9\u304c\u5b8c\u4e86\u3057\u305f\u3089\u3053\u308c\u3092\u30b9\u30ad\u30c3\u30d7\u3057\u307e\u3059\u3002</p>\n",
"<p>Log a sample </p>\n": "<p>\u30b5\u30f3\u30d7\u30eb\u3092\u30ed\u30b0\u306b\u8a18\u9332\u3059\u308b</p>\n",
"<p>Log the score </p>\n": "<p>\u30b9\u30b3\u30a2\u3092\u8a18\u9332\u3059\u308b</p>\n",
"<p>Make a problem with a pre_explanation or not</p>\n<p>Creates an arithmetic addition problem with workings and answer.</p>\n": "<p>pre_explanation \u3067\u554f\u984c\u3092\u8d77\u3053\u3059\u304b\u3057\u306a\u3044\u304b</p>\n<p>\u8a08\u7b97\u3068\u89e3\u3092\u542b\u3080\u7b97\u8853\u52a0\u7b97\u554f\u984c\u3092\u4f5c\u6210\u3057\u307e\u3059\u3002</p>\n",
"<p>Maximum number of digits per operand integer </p>\n": "<p>\u30aa\u30da\u30e9\u30f3\u30c9\u6574\u6570\u3042\u305f\u308a\u306e\u6700\u5927\u6841\u6570</p>\n",
"<p>Move to device </p>\n": "<p>\u30c7\u30d0\u30a4\u30b9\u306b\u79fb\u52d5</p>\n",
"<p>No need of a validation dataset </p>\n": "<p>\u691c\u8a3c\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306f\u4e0d\u8981</p>\n",
"<p>Number of problems in evaluation </p>\n": "<p>\u8a55\u4fa1\u4e2d\u306e\u554f\u984c\u306e\u6570</p>\n",
"<p>Number of sequences that have completed </p>\n": "<p>\u5b8c\u4e86\u3057\u305f\u30b7\u30fc\u30b1\u30f3\u30b9\u306e\u6570</p>\n",
"<p>Number of times to run evaluations per epoch </p>\n": "<p>\u30a8\u30dd\u30c3\u30af\u3054\u3068\u306b\u8a55\u4fa1\u3092\u5b9f\u884c\u3059\u308b\u56de\u6570</p>\n",
"<p>Number of tokens in the vocabulary </p>\n": "<p>\u30dc\u30ad\u30e3\u30d6\u30e9\u30ea\u30fc\u306e\u30c8\u30fc\u30af\u30f3\u306e\u6570</p>\n",
"<p>Number of training sequences per epoch </p>\n": "<p>\u30a8\u30dd\u30c3\u30af\u3042\u305f\u308a\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30b7\u30fc\u30b1\u30f3\u30b9\u306e\u6570</p>\n",
"<p>Override with the question </p>\n": "<p>\u8cea\u554f\u3067\u4e0a\u66f8\u304d</p>\n",
"<p>Sample upto sequence length </p>\n": "<p>\u30b7\u30fc\u30b1\u30f3\u30b9\u9577\u307e\u3067\u306e\u30b5\u30f3\u30d7\u30eb</p>\n",
"<p>Sampled results </p>\n": "<p>\u30b5\u30f3\u30d7\u30eb\u7d50\u679c</p>\n",
"<p>Skip if all have finished </p>\n": "<p>\u3059\u3079\u3066\u7d42\u4e86\u3057\u305f\u3089\u30b9\u30ad\u30c3\u30d7</p>\n",
"<p>Skip in the first epoch </p>\n": "<p>\u6700\u521d\u306e\u30a8\u30dd\u30c3\u30af\u3092\u30b9\u30ad\u30c3\u30d7</p>\n",
"<p>Token id of the new line character - this marks end of the answer </p>\n": "<p>\u6539\u884c\u6587\u5b57\u306e\u30c8\u30fc\u30af\u30f3ID-\u3053\u308c\u3067\u56de\u7b54\u306e\u6700\u5f8c\u306b\u306a\u308a\u307e\u3059</p>\n",
"<p>Token id to string </p>\n": "<p>\u30c8\u30fc\u30af\u30f3 ID \u3092\u6587\u5b57\u5217\u306b</p>\n",
"<p>Training data loader </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the sequence length of generated math problems. We fill as many problems as possible upto this length :max_digits: is the maximum number of digits in the operand integers :n_sequences: is the number of sequences per epoch</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u751f\u6210\u3055\u308c\u305f\u6570\u5b66\u554f\u984c\u306e\u30b7\u30fc\u30b1\u30f3\u30b9\u9577\u3067\u3059\u3002\u3053\u306e\u9577\u3055\u307e\u3067\u3067\u304d\u308b\u3060\u3051\u591a\u304f\u306e\u554f\u984c\u3092\u89e3\u304d\u307e\u3059\u3002max_digits: \u306f\u30aa\u30da\u30e9\u30f3\u30c9\u6574\u6570\u306e\u6700\u5927\u6841\u6570:n_sequences: \u306f\u30a8\u30dd\u30c3\u30af\u3042\u305f\u308a\u306e\u30b7\u30fc\u30b1\u30f3\u30b9\u6570</li></ul>\n",
"Arithmetic Dataset": "\u7b97\u8853\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8",
"This creates arithmetic problems.": "\u3053\u308c\u306f\u7b97\u8853\u4e0a\u306e\u554f\u984c\u3092\u5f15\u304d\u8d77\u3053\u3057\u307e\u3059\u3002"
}
translate_cache/experiments/arithmetic_dataset.si.json
+53
View File
@@ -0,0 +1,53 @@
{
"<h2>Arithmetic Dataset</h2>\n<p>This creates arithmetic addition problems and solutions with workings. We&#x27;ve only implemented addition so far.</p>\n<p>It&#x27;s based on a character level tokenization.</p>\n": "<h2>\u0d85\u0d82\u0d9a\u0d9c\u0dab\u0dd2\u0dad \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h2>\n<p>\u0db8\u0dd9\u0dba\u0d85\u0d82\u0d9a \u0d9c\u0dab\u0dd2\u0dad\u0db8\u0dba \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0dc3\u0dc4 \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0d9a\u0dcf\u0dbb\u0dd2\u0dad\u0dca\u0dc0\u0dba \u0dc3\u0db8\u0d9f \u0dc0\u0dd2\u0dc3\u0db3\u0dd4\u0db8\u0dca \u0db1\u0dd2\u0dbb\u0dca\u0db8\u0dcf\u0dab\u0dba \u0d9a\u0dbb\u0dba\u0dd2. \u0d85\u0db4\u0dd2 \u0db8\u0dd9\u0dad\u0dd9\u0d9a\u0dca \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dbb \u0d87\u0dad\u0dca\u0dad\u0dda \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0db4\u0db8\u0dab\u0dd2. </p>\n<p>\u0d91\u0dba\u0db4\u0daf\u0db1\u0db8\u0dca \u0dc0\u0dd3 \u0d87\u0dad\u0dca\u0dad\u0dda \u0da0\u0dbb\u0dd2\u0dad \u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dd3\u0d9a\u0dbb\u0dab\u0dba \u0db8\u0dad \u0dba. </p>\n",
"<h2>Arithmetic Task Experiment Configurations</h2>\n": "<h2>\u0d85\u0d82\u0d9a\u0d9c\u0dab\u0dd2\u0dad \u0d9a\u0dcf\u0dbb\u0dca\u0dba \u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8\u0dda \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca</h2>\n",
"<h3>Evaluation</h3>\n<p>We use the sampling function to evaluate the model on a set of problems</p>\n": "<h3>\u0d87\u0d9c\u0dba\u0dd3\u0db8</h3>\n<p>\u0d9c\u0dd0\u0da7\u0dc5\u0dd4\u0dc3\u0db8\u0dd6\u0dc4\u0dba\u0d9a\u0dca \u0db8\u0dad \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba \u0d87\u0d9c\u0dba\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d85\u0db4\u0dd2 \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2 \u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db8\u0dd4</p>\n",
"<p> </p>\n": "<p> </p>\n",
"<p> Code to test generated problems</p>\n": "<p> \u0da2\u0db1\u0db1\u0dba\u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0db4\u0dbb\u0dd3\u0d9a\u0dca\u0dc2\u0dcf \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d9a\u0dda\u0dad\u0dba</p>\n",
"<p> Decode a list of token ids</p>\n": "<p> \u0da7\u0ddd\u0d9a\u0db1\u0dca\u0dc4\u0dd0\u0db3\u0dd4\u0db1\u0dd4\u0db8\u0dca\u0db4\u0dad\u0dca \u0dbd\u0dd0\u0dba\u0dd2\u0dc3\u0dca\u0dad\u0dd4\u0dc0\u0d9a\u0dca \u0dc0\u0dd2\u0d9a\u0dda\u0dad\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1</p>\n",
"<p> Encode a given string</p>\n": "<p> \u0daf\u0dd3\u0d87\u0dad\u0dd2 \u0db1\u0dd6\u0dbd\u0d9a\u0dca \u0d9a\u0dda\u0dad\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1</p>\n",
"<p> Generate multiple problems and pack them into a sequence.</p>\n": "<p> \u0db6\u0dc4\u0dd4\u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0da2\u0db1\u0db1\u0dba \u0d9a\u0dbb \u0d92\u0dc0\u0dcf \u0d85\u0db1\u0dd4\u0db4\u0dd2\u0dc5\u0dd2\u0dc0\u0dd9\u0dbd\u0d9a\u0da7 \u0d87\u0dc3\u0dd4\u0dbb\u0dd4\u0db8\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1. </p>\n",
"<p> Generates an integer with <span translate=no>_^_0_^_</span> number of digits</p>\n": "<p> \u0d89\u0dbd\u0d9a\u0dca\u0d9a\u0db8\u0dca <span translate=no>_^_0_^_</span> \u0d9c\u0dab\u0db1 \u0dc3\u0dc4\u0dd2\u0dad \u0db4\u0dd6\u0dbb\u0dca\u0dab \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0dc0\u0d9a\u0dca \u0da2\u0db1\u0db1\u0dba \u0d9a\u0dbb\u0dba\u0dd2</p>\n",
"<p> Generates the workings for <span translate=no>_^_0_^_</span>. For example for <span translate=no>_^_1_^_</span> it generates <span translate=no>_^_2_^_</span>.</p>\n": "<p> \u0dc3\u0db3\u0dc4\u0dcf\u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0d9a\u0dcf\u0dbb\u0dd2\u0dad\u0dca\u0dc0\u0dba \u0da2\u0db1\u0db1\u0dba \u0d9a\u0dbb\u0dba\u0dd2 <span translate=no>_^_0_^_</span>. \u0d8b\u0daf\u0dcf\u0dc4\u0dbb\u0dab\u0dba\u0d9a\u0dca \u0dbd\u0dd9\u0dc3 <span translate=no>_^_1_^_</span> \u0d91\u0dba \u0da2\u0db1\u0db1\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf <span translate=no>_^_2_^_</span>. </p>\n",
"<p> Get a input and target pair for auto-regressive modelling</p>\n": "<p> \u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0d9c\u0dcf\u0db8\u0dd3 \u0d86\u0d9a\u0dd8\u0dad\u0dd2 \u0db1\u0dd2\u0dbb\u0dca\u0db8\u0dcf\u0dab\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0d86\u0daf\u0dcf\u0db1 \u0dc3\u0dc4 \u0d89\u0dbd\u0d9a\u0dca\u0d9a \u0dba\u0dd4\u0d9c\u0dbd\u0dba\u0d9a\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1</p>\n",
"<p> Get arithmetic problem and answer. This is used for evaluation.</p>\n": "<p> \u0d85\u0d82\u0d9a\u0d9c\u0dab\u0dd2\u0dad\u0db8\u0dba \u0d9c\u0dd0\u0da7\u0dc5\u0dd4\u0dc0 \u0dc3\u0dc4 \u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dd4 \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1. \u0db8\u0dd9\u0dba \u0d87\u0d9c\u0dba\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0dc0\u0dda. </p>\n",
"<p> Number of sequences per epoch</p>\n": "<p> \u0d91\u0db4\u0ddd\u0da0\u0dca\u0d91\u0d9a\u0d9a\u0da7 \u0d85\u0db1\u0dd4\u0db4\u0dd2\u0dc5\u0dd2\u0dc0\u0dd9\u0dbd\u0dc0\u0dbd\u0dca \u0d9c\u0dab\u0db1</p>\n",
"<p> Training data loader</p>\n": "<p> \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8</p>\n",
"<p><em>This is based on code by <a href=\"https://twitter.com/gharik\">Georges Harik (@gharik)</a>.</em></p>\n": "<p><em>\u0db8\u0dd9\u0dba <a href=\"https://twitter.com/gharik\">\u0da2\u0ddd\u0dbb\u0dca\u0da2\u0dc3\u0dca \u0dc4\u0dcf\u0dbb\u0dd2\u0d9a\u0dca (@gharik)</a>\u0dc0\u0dd2\u0dc3\u0dd2\u0db1\u0dca \u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0d9a\u0dda\u0dad\u0dba \u0db8\u0dad \u0db4\u0daf\u0db1\u0db8\u0dca \u0dc0\u0dda. </em></p>\n",
"<p>Add the next token to the input </p>\n": "<p>\u0d86\u0daf\u0dcf\u0db1\u0dba\u0da7\u0d8a\u0dc5\u0d9f \u0da7\u0ddd\u0d9a\u0db1\u0dba \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Character to token id </p>\n": "<p>\u0d85\u0d9a\u0dca\u0dc2\u0dbb\u0da7\u0ddd\u0d9a\u0db1\u0dca \u0dc4\u0dd0\u0db3\u0dd4\u0db1\u0dd4\u0db8\u0dca\u0db4\u0dad\u0da7 </p>\n",
"<p>Collect the problems only </p>\n": "<p>\u0d9c\u0dd0\u0da7\u0dc5\u0dd4\u0db4\u0db8\u0dab\u0d9a\u0dca \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Count the number of correct answers </p>\n": "<p>\u0db1\u0dd2\u0dc0\u0dd0\u0dbb\u0daf\u0dd2\u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dd4 \u0d9c\u0dab\u0db1 \u0d9c\u0dab\u0db1\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Create a dataset to generate problems </p>\n": "<p>\u0d9c\u0dd0\u0da7\u0dc5\u0dd4\u0da2\u0db1\u0db1\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba\u0d9a\u0dca \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Create a tensor with only the initial token </p>\n": "<p>\u0d86\u0dbb\u0db8\u0dca\u0db7\u0d9a\u0da7\u0ddd\u0d9a\u0db1\u0dba \u0db4\u0db8\u0dab\u0d9a\u0dca \u0dc3\u0dc4\u0dd2\u0dad \u0da7\u0dd9\u0db1\u0dca\u0dc3\u0dbb\u0dba\u0d9a\u0dca \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Discard everything after the answer in the results </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0db5\u0dbd\u0dc0\u0dbd\u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dd9\u0db1\u0dca \u0db4\u0dc3\u0dd4 \u0dc3\u0dd2\u0dba\u0dbd\u0dca\u0dbd \u0d89\u0dc0\u0dad\u0dbd\u0db1\u0dca\u0db1 </p>\n",
"<p>Find which sequences have finished </p>\n": "<p>\u0d9a\u0dd4\u0db8\u0db1\u0d85\u0db1\u0dd4\u0db4\u0dd2\u0dc5\u0dd2\u0dc0\u0dd9\u0dbd\u0dc0\u0dbd\u0dca \u0d85\u0dc0\u0dc3\u0db1\u0dca \u0d9a\u0dbb \u0d87\u0dad\u0dca\u0daf\u0dd0\u0dba\u0dd2 \u0dc3\u0ddc\u0dba\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Get a set of problems and answers </p>\n": "<p>\u0d9c\u0dd0\u0da7\u0dc5\u0dd4\u0dc3\u0dc4 \u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dd4 \u0dc3\u0db8\u0dd6\u0dc4\u0dba\u0d9a\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Get the answers </p>\n": "<p>\u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dd4\u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Get the model output </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Get the model prediction (greedy) </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0d85\u0db1\u0dcf\u0dc0\u0dd0\u0d9a\u0dd2\u0dba \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 (\u0d9a\u0dd1\u0daf\u0dbb) </p>\n",
"<p>Get the sampled results </p>\n": "<p>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0db5\u0dbd \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>If all the sequences have completed we skip this </p>\n": "<p>\u0dc3\u0dd2\u0dba\u0dbd\u0dd4\u0db8\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dba\u0db1\u0dca \u0dc3\u0db8\u0dca\u0db4\u0dd6\u0dbb\u0dca\u0dab \u0d9a\u0dbb \u0d87\u0dad\u0dca\u0db1\u0db8\u0dca \u0d85\u0db4\u0dd2 \u0db8\u0dd9\u0dba \u0db8\u0d9f \u0dc4\u0dbb\u0dd2\u0db8\u0dd4 </p>\n",
"<p>Log a sample </p>\n": "<p>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba\u0d9a\u0dca\u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Log the score </p>\n": "<p>\u0dbd\u0d9a\u0dd4\u0dab\u0dd4\u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Make a problem with a pre_explanation or not</p>\n<p>Creates an arithmetic addition problem with workings and answer.</p>\n": "<p>\u0db4\u0dd6\u0dbb\u0dca\u0dc0\u0db4\u0dd0\u0dc4\u0dd0\u0daf\u0dd2\u0dbd\u0dd2 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0d9a\u0dca \u0dc3\u0db8\u0d9f \u0d9c\u0dd0\u0da7\u0dc5\u0dd4\u0dc0\u0d9a\u0dca \u0d87\u0dad\u0dd2 \u0d9a\u0dbb\u0db1\u0dca\u0db1 \u0dc4\u0ddd \u0db1\u0dd0\u0dad</p>\n<p>workings\u0dc4\u0dcf \u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dd4 \u0dc3\u0db8\u0d9c \u0d85\u0d82\u0d9a \u0d9c\u0dab\u0dd2\u0dad\u0db8\u0dba \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d9c\u0dd0\u0da7\u0dbd\u0dd4\u0dc0\u0d9a\u0dca \u0db1\u0dd2\u0dbb\u0dca\u0db8\u0dcf\u0dab\u0dba \u0d9a\u0dbb\u0dba\u0dd2. </p>\n",
"<p>Maximum number of digits per operand integer </p>\n": "<p>\u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0d9a\u0dcf\u0dbb\u0dd3\u0db4\u0dd6\u0dbb\u0dca\u0dab \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0dc0\u0d9a\u0da7 \u0d8b\u0db4\u0dbb\u0dd2\u0db8 \u0d89\u0dbd\u0d9a\u0dca\u0d9a\u0db8\u0dca \u0d9c\u0dab\u0db1 </p>\n",
"<p>Move to device </p>\n": "<p>\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba\u0dc0\u0dd9\u0dad \u0d9c\u0dd9\u0db1 \u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>No need of a validation dataset </p>\n": "<p>\u0dc0\u0dbd\u0d82\u0d9c\u0dd4\u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba\u0d9a\u0dca \u0d85\u0dc0\u0dc1\u0dca\u0dba \u0db1\u0ddc\u0dc0\u0dda </p>\n",
"<p>Number of problems in evaluation </p>\n": "<p>\u0d87\u0d9c\u0dba\u0dd3\u0db8\u0dda\u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0d9c\u0dab\u0db1 </p>\n",
"<p>Number of sequences that have completed </p>\n": "<p>\u0dc3\u0db8\u0dca\u0db4\u0dd6\u0dbb\u0dca\u0dab\u0d9a\u0dbb \u0d87\u0dad\u0dd2 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8 \u0d9c\u0dab\u0db1 </p>\n",
"<p>Number of times to run evaluations per epoch </p>\n": "<p>\u0d91\u0db4\u0ddd\u0da0\u0dca\u0d91\u0d9a\u0d9a\u0da7 \u0d87\u0d9c\u0dba\u0dd3\u0db8\u0dca \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0dc0\u0dcf\u0dbb \u0d9c\u0dab\u0db1 </p>\n",
"<p>Number of tokens in the vocabulary </p>\n": "<p>\u0dc0\u0da0\u0db1\u0db8\u0dcf\u0dbd\u0dcf\u0dc0\u0dda \u0da7\u0ddd\u0d9a\u0db1 \u0d9c\u0dab\u0db1 </p>\n",
"<p>Number of training sequences per epoch </p>\n": "<p>\u0d91\u0db4\u0ddd\u0da0\u0dca\u0d91\u0d9a\u0d9a\u0da7 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d85\u0db1\u0dd4\u0db4\u0dd2\u0dc5\u0dd2\u0dc0\u0dd9\u0dbd \u0d9c\u0dab\u0db1 </p>\n",
"<p>Override with the question </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dca\u0db1\u0dba\u0dc3\u0db8\u0d9f \u0d85\u0db7\u0dd2\u0db6\u0dc0\u0dcf \u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Sample upto sequence length </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0daf\u0dd2\u0d9c \u0daf\u0d9a\u0dca\u0dc0\u0dcf \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba </p>\n",
"<p>Sampled results </p>\n": "<p>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0db4\u0dca\u0dbb\u0dad\u0dd2. \u0dbd </p>\n",
"<p>Skip if all have finished </p>\n": "<p>\u0dc3\u0dd2\u0dba\u0dbd\u0dca\u0dbd\u0d85\u0dc0\u0dc3\u0db1\u0dca \u0dc0\u0dd3 \u0d87\u0dad\u0dca\u0db1\u0db8\u0dca \u0db8\u0d9f \u0dc4\u0dbb\u0dd2\u0db1\u0dca\u0db1 </p>\n",
"<p>Skip in the first epoch </p>\n": "<p>\u0db4\u0dc5\u0db8\u0dd4\u0d91\u0db4\u0ddd\u0da0\u0dca \u0d91\u0d9a\u0dda \u0db8\u0d9f \u0dc4\u0dbb\u0dd2\u0db1\u0dca\u0db1 </p>\n",
"<p>Token id of the new line character - this marks end of the answer </p>\n": "<p>\u0db1\u0dc0\u0dbb\u0dda\u0d9b\u0dcf \u0d85\u0d9a\u0dca\u0dc2\u0dbb\u0dba\u0dd9\u0dc4\u0dd2 \u0da7\u0ddd\u0d9a\u0db1\u0dca \u0dc4\u0dd0\u0db3\u0dd4\u0db1\u0dd4\u0db8\u0dca\u0db4\u0dad - \u0db8\u0dd9\u0dba \u0db4\u0dd2\u0dc5\u0dd2\u0dad\u0dd4\u0dbb\u0dda \u0d85\u0dc0\u0dc3\u0dcf\u0db1\u0dba \u0dc3\u0dbd\u0d9a\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0dba\u0dd2 </p>\n",
"<p>Token id to string </p>\n": "<p>\u0db1\u0dd6\u0dbd\u0da7\u0da7\u0ddd\u0d9a\u0db1\u0dca \u0dc4\u0dd0\u0db3\u0dd4\u0db1\u0dd4\u0db8\u0dca\u0db4\u0dad </p>\n",
"<p>Training data loader </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 </p>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the sequence length of generated math problems. We fill as many problems as possible upto this length :max_digits: is the maximum number of digits in the operand integers :n_sequences: is the number of sequences per epoch</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0dba\u0db1\u0dd4 \u0da2\u0db1\u0db1\u0dba \u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0d9c\u0dab\u0dd2\u0dad \u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0dc0\u0dbd \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0daf\u0dd2\u0d9c\u0dba\u0dd2. \u0db8\u0dd9\u0db8 \u0daf\u0dd2\u0d9c \u0daf\u0d9a\u0dca\u0dc0\u0dcf \u0d85\u0db4\u0dd2 \u0dc4\u0dd0\u0d9a\u0dd2 \u0dad\u0dbb\u0db8\u0dca \u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0db4\u0dd4\u0dbb\u0dc0\u0db1\u0dca\u0db1\u0dd9\u0db8\u0dd4: max_digits: \u0dba\u0db1\u0dd4 \u0d94\u0db4\u0dd9\u0dbb\u0db1\u0dca\u0da9\u0dca \u0dc3\u0d82\u0d9b\u0dca\u200d\u0dba\u0dcf\u0dc0\u0dda \u0d8b\u0db4\u0dbb\u0dd2\u0db8 \u0d89\u0dbd\u0d9a\u0dca\u0d9a\u0db8\u0dca \u0dc3\u0d82\u0d9b\u0dca\u200d\u0dba\u0dcf\u0dc0 \u0dc0\u0dda: n_sequences: \u0dba\u0db1\u0dd4 \u0d91\u0db4\u0ddd\u0da0\u0dca \u0d91\u0d9a\u0d9a\u0da7 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u200d\u0dbb\u0db8 \u0d9c\u0dab\u0db1</li></ul>\u0dc0\u0dda\n",
"Arithmetic Dataset": "\u0d85\u0d82\u0d9a \u0d9c\u0dab\u0dd2\u0dad \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba",
"This creates arithmetic problems.": "\u0db8\u0dd9\u0dba \u0d85\u0d82\u0d9a \u0d9c\u0dab\u0dd2\u0dad \u0d9c\u0dd0\u0da7\u0dc5\u0dd4 \u0db1\u0dd2\u0dbb\u0dca\u0db8\u0dcf\u0dab\u0dba \u0d9a\u0dbb\u0dba\u0dd2."
}
translate_cache/experiments/arithmetic_dataset.zh.json
+53
View File
@@ -0,0 +1,53 @@
{
"<h2>Arithmetic Dataset</h2>\n<p>This creates arithmetic addition problems and solutions with workings. We&#x27;ve only implemented addition so far.</p>\n<p>It&#x27;s based on a character level tokenization.</p>\n": "<h2>\u7b97\u672f\u6570\u636e\u96c6</h2>\n<p>\u8fd9\u4f1a\u4ea7\u751f\u7b97\u672f\u52a0\u6cd5\u95ee\u9898\u548c\u8fd0\u4f5c\u89e3\u51b3\u65b9\u6848\u3002\u5230\u76ee\u524d\u4e3a\u6b62\uff0c\u6211\u4eec\u53ea\u5b9e\u65bd\u4e86\u52a0\u6cd5\u3002</p>\n<p>\u5b83\u57fa\u4e8e\u89d2\u8272\u7ea7\u522b\u7684\u6807\u8bb0\u5316\u3002</p>\n",
"<h2>Arithmetic Task Experiment Configurations</h2>\n": "<h2>\u7b97\u672f\u4efb\u52a1\u5b9e\u9a8c\u914d\u7f6e</h2>\n",
"<h3>Evaluation</h3>\n<p>We use the sampling function to evaluate the model on a set of problems</p>\n": "<h3>\u8bc4\u4f30</h3>\n<p>\u6211\u4eec\u4f7f\u7528\u91c7\u6837\u51fd\u6570\u6765\u8bc4\u4f30\u4e00\u7ec4\u95ee\u9898\u7684\u6a21\u578b</p>\n",
"<p> </p>\n": "<p></p>\n",
"<p> Code to test generated problems</p>\n": "<p>\u7528\u4e8e\u6d4b\u8bd5\u751f\u6210\u7684\u95ee\u9898\u7684\u4ee3\u7801</p>\n",
"<p> Decode a list of token ids</p>\n": "<p>\u89e3\u7801\u4ee4\u724c ID \u5217\u8868</p>\n",
"<p> Encode a given string</p>\n": "<p>\u5bf9\u7ed9\u5b9a\u5b57\u7b26\u4e32\u8fdb\u884c\u7f16\u7801</p>\n",
"<p> Generate multiple problems and pack them into a sequence.</p>\n": "<p>\u751f\u6210\u591a\u4e2a\u95ee\u9898\u5e76\u5c06\u5b83\u4eec\u6253\u5305\u6210\u4e00\u4e2a\u5e8f\u5217\u3002</p>\n",
"<p> Generates an integer with <span translate=no>_^_0_^_</span> number of digits</p>\n": "<p>\u751f\u6210\u4e00\u4e2a\u5305\u542b\u4f4d<span translate=no>_^_0_^_</span>\u6570\u7684\u6574\u6570</p>\n",
"<p> Generates the workings for <span translate=no>_^_0_^_</span>. For example for <span translate=no>_^_1_^_</span> it generates <span translate=no>_^_2_^_</span>.</p>\n": "<p>\u751f\u6210\u7684\u5de5\u4f5c\u539f\u7406<span translate=no>_^_0_^_</span>\u3002\u4f8b\u5982\uff0c<span translate=no>_^_1_^_</span>\u5b83\u4f1a\u751f\u6210<span translate=no>_^_2_^_</span>\u3002</p>\n",
"<p> Get a input and target pair for auto-regressive modelling</p>\n": "<p>\u83b7\u53d6\u81ea\u52a8\u56de\u5f52\u5efa\u6a21\u7684\u8f93\u5165\u548c\u76ee\u6807\u5bf9</p>\n",
"<p> Get arithmetic problem and answer. This is used for evaluation.</p>\n": "<p>\u83b7\u53d6\u7b97\u672f\u95ee\u9898\u548c\u7b54\u6848\u3002\u8fd9\u7528\u4e8e\u8bc4\u4f30\u3002</p>\n",
"<p> Number of sequences per epoch</p>\n": "<p>\u6bcf\u4e2a\u7eaa\u5143\u7684\u5e8f\u5217\u6570</p>\n",
"<p> Training data loader</p>\n": "<p>\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p><em>This is based on code by <a href=\"https://twitter.com/gharik\">Georges Harik (@gharik)</a>.</em></p>\n": "<p><em>\u8fd9\u662f\u57fa\u4e8e<a href=\"https://twitter.com/gharik\">\u4e54\u6cbb\u00b7\u54c8\u91cc\u514b\uff08@gharik\uff09</a>\u7684\u4ee3\u7801\u3002</em></p>\n",
"<p>Add the next token to the input </p>\n": "<p>\u5c06\u4e0b\u4e00\u4e2a\u4ee4\u724c\u6dfb\u52a0\u5230\u8f93\u5165\u4e2d</p>\n",
"<p>Character to token id </p>\n": "<p>\u5b57\u7b26\u5230\u4ee4\u724c ID</p>\n",
"<p>Collect the problems only </p>\n": "<p>\u53ea\u6536\u96c6\u95ee\u9898</p>\n",
"<p>Count the number of correct answers </p>\n": "<p>\u8ba1\u7b97\u6b63\u786e\u7b54\u6848\u7684\u6570\u91cf</p>\n",
"<p>Create a dataset to generate problems </p>\n": "<p>\u521b\u5efa\u6570\u636e\u96c6\u4ee5\u751f\u6210\u95ee\u9898</p>\n",
"<p>Create a tensor with only the initial token </p>\n": "<p>\u4ec5\u4f7f\u7528\u521d\u59cb\u4ee4\u724c\u521b\u5efa\u5f20\u91cf</p>\n",
"<p>Discard everything after the answer in the results </p>\n": "<p>\u4e22\u5f03\u7ed3\u679c\u4e2d\u7b54\u6848\u540e\u7684\u6240\u6709\u5185\u5bb9</p>\n",
"<p>Find which sequences have finished </p>\n": "<p>\u627e\u51fa\u54ea\u4e9b\u5e8f\u5217\u5df2\u5b8c\u6210</p>\n",
"<p>Get a set of problems and answers </p>\n": "<p>\u83b7\u53d6\u4e00\u7cfb\u5217\u95ee\u9898\u548c\u7b54\u6848</p>\n",
"<p>Get the answers </p>\n": "<p>\u5f97\u5230\u7b54\u6848</p>\n",
"<p>Get the model output </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u8f93\u51fa</p>\n",
"<p>Get the model prediction (greedy) </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u9884\u6d4b\uff08\u8d2a\u5a6a\uff09</p>\n",
"<p>Get the sampled results </p>\n": "<p>\u83b7\u53d6\u62bd\u6837\u7ed3\u679c</p>\n",
"<p>If all the sequences have completed we skip this </p>\n": "<p>\u5982\u679c\u6240\u6709\u7684\u5e8f\u5217\u90fd\u5b8c\u6210\u4e86\uff0c\u6211\u4eec\u5c31\u8df3\u8fc7\u8fd9\u4e2a</p>\n",
"<p>Log a sample </p>\n": "<p>\u8bb0\u5f55\u6837\u672c</p>\n",
"<p>Log the score </p>\n": "<p>\u8bb0\u5f55\u5206\u6570</p>\n",
"<p>Make a problem with a pre_explanation or not</p>\n<p>Creates an arithmetic addition problem with workings and answer.</p>\n": "<p>\u4e0d\u7ba1\u662f\u5426\u7528 pre_explansion \u95ee\u95ee\u9898</p>\n<p>\u7528\u8fd0\u4f5c\u548c\u7b54\u6848\u521b\u5efa\u7b97\u672f\u52a0\u6cd5\u95ee\u9898\u3002</p>\n",
"<p>Maximum number of digits per operand integer </p>\n": "<p>\u6bcf\u4e2a\u64cd\u4f5c\u6570\u6574\u6570\u7684\u6700\u5927\u4f4d\u6570</p>\n",
"<p>Move to device </p>\n": "<p>\u79fb\u81f3\u8bbe\u5907</p>\n",
"<p>No need of a validation dataset </p>\n": "<p>\u4e0d\u9700\u8981\u9a8c\u8bc1\u6570\u636e\u96c6</p>\n",
"<p>Number of problems in evaluation </p>\n": "<p>\u8bc4\u4f30\u4e2d\u7684\u95ee\u9898\u6570\u91cf</p>\n",
"<p>Number of sequences that have completed </p>\n": "<p>\u5df2\u5b8c\u6210\u7684\u5e8f\u5217\u6570</p>\n",
"<p>Number of times to run evaluations per epoch </p>\n": "<p>\u6bcf\u4e2a\u7eaa\u5143\u8fd0\u884c\u8bc4\u4f30\u7684\u6b21\u6570</p>\n",
"<p>Number of tokens in the vocabulary </p>\n": "<p>\u8bcd\u6c47\u8868\u4e2d\u7684\u4ee3\u5e01\u6570\u91cf</p>\n",
"<p>Number of training sequences per epoch </p>\n": "<p>\u6bcf\u4e2a\u7eaa\u5143\u7684\u8bad\u7ec3\u5e8f\u5217\u6570</p>\n",
"<p>Override with the question </p>\n": "<p>\u7528\u95ee\u9898\u8986\u76d6</p>\n",
"<p>Sample upto sequence length </p>\n": "<p>\u6837\u672c\u76f4\u81f3\u5e8f\u5217\u957f\u5ea6</p>\n",
"<p>Sampled results </p>\n": "<p>\u62bd\u6837\u7ed3\u679c</p>\n",
"<p>Skip if all have finished </p>\n": "<p>\u5982\u679c\u5168\u90e8\u5b8c\u6210\uff0c\u5219\u8df3\u8fc7</p>\n",
"<p>Skip in the first epoch </p>\n": "<p>\u8df3\u8fc7\u7b2c\u4e00\u4e2a\u7eaa\u5143</p>\n",
"<p>Token id of the new line character - this marks end of the answer </p>\n": "<p>\u6362\u884c\u7b26\u7684\u6807\u8bb0 ID-\u8fd9\u6807\u5fd7\u7740\u7b54\u6848\u7684\u7ed3\u675f</p>\n",
"<p>Token id to string </p>\n": "<p>\u4ee4\u724c ID \u8f6c\u6362\u4e3a\u5b57\u7b26\u4e32</p>\n",
"<p>Training data loader </p>\n": "<p>\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the sequence length of generated math problems. We fill as many problems as possible upto this length :max_digits: is the maximum number of digits in the operand integers :n_sequences: is the number of sequences per epoch</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u662f\u751f\u6210\u7684\u6570\u5b66\u95ee\u9898\u7684\u5e8f\u5217\u957f\u5ea6\u3002\u6211\u4eec\u5c3d\u53ef\u80fd\u591a\u5730\u586b\u5199\u95ee\u9898\uff0c\u76f4\u5230\u8fd9\u4e2a\u957f\u5ea6\uff1amax_digits: \u662f\u64cd\u4f5c\u6570\u4e2d\u7684\u6700\u5927\u4f4d\u6570\u6574\u6570:n_sequences: \u662f\u6bcf\u4e2a\u7eaa\u5143\u7684\u5e8f\u5217\u6570</li></ul>\n",
"Arithmetic Dataset": "\u7b97\u672f\u6570\u636e\u96c6",
"This creates arithmetic problems.": "\u8fd9\u4f1a\u4ea7\u751f\u7b97\u672f\u95ee\u9898\u3002"
}
translate_cache/experiments/cifar10.ja.json
+24
View File
@@ -0,0 +1,24 @@
{
"<h1>CIFAR10 Experiment</h1>\n": "<h1>CIFAR10 \u5b9f\u9a13</h1>\n",
"<h2>Configurations</h2>\n<p>This extends from CIFAR 10 dataset configurations from <a href=\"https://github.com/labmlai/labml/tree/master/helpers\"><span translate=no>_^_0_^_</span></a> and <a href=\"mnist.html\"><span translate=no>_^_1_^_</span></a>.</p>\n": "<h2>\u30b3\u30f3\u30d5\u30a3\u30ae\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3</h2>\n<p>\u3053\u308c\u306f\u3001\u304a\u3088\u3073\u306e CIFAR 10 \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u69cb\u6210\u3092\u62e1\u5f35\u3057\u305f\u3082\u306e\u3067\u3059<a href=\"https://github.com/labmlai/labml/tree/master/helpers\"><span translate=no>_^_0_^_</span></a>\u3002<a href=\"mnist.html\"><span translate=no>_^_1_^_</span></a></p>\n",
"<h3>Augmented CIFAR 10 train dataset</h3>\n": "<h3>\u62e1\u5f35\u3055\u308c\u305f CIFAR 10 \u30c8\u30ec\u30a4\u30f3\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</h3>\n",
"<h3>Non-augmented CIFAR 10 validation dataset</h3>\n": "<h3>\u62e1\u5f35\u3055\u308c\u3066\u3044\u306a\u3044 CIFAR 10 \u691c\u8a3c\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</h3>\n",
"<h3>VGG model for CIFAR-10 classification</h3>\n": "<h3>CIFAR-10 \u5206\u985e\u7528\u306e VGG \u30e2\u30c7\u30eb</h3>\n",
"<p> </p>\n": "<p></p>\n",
"<p> Convolution and activation combined</p>\n": "<p>\u30b3\u30f3\u30dc\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u3068\u30a2\u30af\u30c6\u30a3\u30d9\u30fc\u30b7\u30e7\u30f3\u306e\u7d44\u307f\u5408\u308f\u305b</p>\n",
"<p>5 <span translate=no>_^_0_^_</span> pooling layers will produce a output of size <span translate=no>_^_1_^_</span>. CIFAR 10 image size is <span translate=no>_^_2_^_</span> </p>\n": "<p><span translate=no>_^_0_^_</span><span translate=no>_^_1_^_</span>5\u3064\u306e\u30d7\u30fc\u30ea\u30f3\u30b0\u30ec\u30a4\u30e4\u30fc\u3067\u30b5\u30a4\u30ba\u306e\u51fa\u529b\u304c\u5f97\u3089\u308c\u307e\u3059\u3002CIFAR 10 \u306e\u753b\u50cf\u30b5\u30a4\u30ba\u306f <span translate=no>_^_2_^_</span></p>\n",
"<p>Convolution, Normalization and Activation layers </p>\n": "<p>\u30b3\u30f3\u30dc\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u3001\u30ce\u30fc\u30de\u30e9\u30a4\u30bc\u30fc\u30b7\u30e7\u30f3\u3001\u30a2\u30af\u30c6\u30a3\u30d9\u30fc\u30b7\u30e7\u30f3\u30ec\u30a4\u30e4\u30fc</p>\n",
"<p>Create a sequential model with the layers </p>\n": "<p>\u30ec\u30a4\u30e4\u30fc\u3092\u542b\u3080\u30b7\u30fc\u30b1\u30f3\u30b7\u30e3\u30eb\u30e2\u30c7\u30eb\u306e\u4f5c\u6210</p>\n",
"<p>Final linear layer </p>\n": "<p>\u6700\u7d42\u7dda\u5f62\u30ec\u30a4\u30e4\u30fc</p>\n",
"<p>Final logits layer </p>\n": "<p>\u6700\u7d42\u30ed\u30b8\u30c3\u30c8\u30ec\u30a4\u30e4\u30fc</p>\n",
"<p>Max pooling at end of each block </p>\n": "<p>\u5404\u30d6\u30ed\u30c3\u30af\u7d42\u4e86\u6642\u306e\u6700\u5927\u30d7\u30fc\u30ea\u30f3\u30b0</p>\n",
"<p>Number of channels in each layer in each block </p>\n": "<p>\u5404\u30d6\u30ed\u30c3\u30af\u306e\u5404\u30ec\u30a4\u30e4\u30fc\u306e\u30c1\u30e3\u30f3\u30cd\u30eb\u6570</p>\n",
"<p>Pad and crop </p>\n": "<p>\u30d1\u30c3\u30c9\u3068\u30af\u30ed\u30c3\u30d7</p>\n",
"<p>RGB channels </p>\n": "<p>RGB \u30c1\u30e3\u30f3\u30cd\u30eb</p>\n",
"<p>Random horizontal flip </p>\n": "<p>\u30e9\u30f3\u30c0\u30e0\u6c34\u5e73\u53cd\u8ee2</p>\n",
"<p>Reshape for classification layer </p>\n": "<p>\u5206\u985e\u30ec\u30a4\u30e4\u30fc\u306e\u5f62\u72b6\u3092\u5909\u66f4</p>\n",
"<p>The VGG layers </p>\n": "<p>VGG \u30ec\u30a4\u30e4\u30fc</p>\n",
"<p>Use CIFAR10 dataset by default </p>\n": "<p>\u30c7\u30d5\u30a9\u30eb\u30c8\u3067 CIFAR10 \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u4f7f\u7528</p>\n",
"CIFAR10 Experiment": "CIFAR10 \u5b9f\u9a13",
"This is a reusable trainer for CIFAR10 dataset": "\u3053\u308c\u306fCIFAR10\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u7528\u306e\u518d\u5229\u7528\u53ef\u80fd\u306a\u30c8\u30ec\u30fc\u30ca\u30fc\u3067\u3059"
}
translate_cache/experiments/cifar10.si.json
+24
View File
@@ -0,0 +1,24 @@
{
"<h1>CIFAR10 Experiment</h1>\n": "<h1>CIFAR10\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf</h1>\n",
"<h2>Configurations</h2>\n<p>This extends from CIFAR 10 dataset configurations from <a href=\"https://github.com/labmlai/labml/tree/master/helpers\"><span translate=no>_^_0_^_</span></a> and <a href=\"mnist.html\"><span translate=no>_^_1_^_</span></a>.</p>\n": "<h2>\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dca</h2>\n<p>\u0db8\u0dd9\u0db8CIFAR \u0dc3\u0dd2\u0da7 \u0dc0\u0dd2\u0dc4\u0dd2\u0daf\u0dd3 10 \u0daf\u0dad\u0dca\u0dad \u0dc3\u0db8\u0dd4\u0daf\u0dcf\u0dba \u0db8\u0dcf\u0db1\u0d9a\u0dbb\u0dab <a href=\"https://github.com/labmlai/labml/tree/master/helpers\"><span translate=no>_^_0_^_</span></a> \u0dc4\u0dcf <a href=\"mnist.html\"><span translate=no>_^_1_^_</span></a> . </p>\n",
"<h3>Augmented CIFAR 10 train dataset</h3>\n": "<h3>\u0dc0\u0dd0\u0da9\u0dd2\u0daf\u0dd2\u0dba\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0db1 \u0dbd\u0daf CIFAR 10 \u0daf\u0dd4\u0db8\u0dca\u0dbb\u0dd2\u0dba \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h3>\n",
"<h3>Non-augmented CIFAR 10 validation dataset</h3>\n": "<h3>\u0dc0\u0dd0\u0da9\u0dd2\u0daf\u0dd2\u0dba\u0dd4\u0dab\u0dd4 \u0db1\u0ddc\u0d9a\u0dbb\u0db1 \u0dbd\u0daf CIFAR 10 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h3>\n",
"<h3>VGG model for CIFAR-10 classification</h3>\n": "<h3>CIFA-10\u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab\u0dba \u0dc3\u0db3\u0dc4\u0dcf VGG \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba</h3>\n",
"<p> </p>\n": "<p> </p>\n",
"<p> Convolution and activation combined</p>\n": "<p> \u0dc3\u0db8\u0dca\u0db8\u0dd4\u0dad\u0dd2\u0dba\u0dc3\u0dc4 \u0dc3\u0d9a\u0dca\u0dbb\u0dd2\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d92\u0d9a\u0dcf\u0db6\u0daf\u0dca\u0db0</p>\n",
"<p>5 <span translate=no>_^_0_^_</span> pooling layers will produce a output of size <span translate=no>_^_1_^_</span>. CIFAR 10 image size is <span translate=no>_^_2_^_</span> </p>\n": "<p><span translate=no>_^_0_^_</span> \u0dad\u0da7\u0dcf\u0d9a \u0dc3\u0dca\u0dae\u0dbb 5 \u0d9a\u0dca \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba\u0dda \u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0d9a\u0dca \u0db1\u0dd2\u0db4\u0daf\u0dc0\u0db1\u0dd4 \u0d87\u0dad <span translate=no>_^_1_^_</span>. CIFAR 10 \u0dbb\u0dd6\u0db4 \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba <span translate=no>_^_2_^_</span> </p>\n",
"<p>Convolution, Normalization and Activation layers </p>\n": "<p>\u0dc3\u0db8\u0dca\u0db8\u0dd4\u0dad\u0dd2\u0dba, \u0dc3\u0dcf\u0db8\u0dcf\u0db1\u0dca\u0dba\u0d9a\u0dbb\u0dab\u0dba \u0dc3\u0dc4 \u0dc3\u0d9a\u0dca\u0dbb\u0dd2\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0dc3\u0dca\u0dae\u0dbb </p>\n",
"<p>Create a sequential model with the layers </p>\n": "<p>\u0dc3\u0dca\u0dae\u0dbb\u0dc3\u0db8\u0d9f \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0d9a\u0dca \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Final linear layer </p>\n": "<p>\u0d85\u0dc0\u0dc3\u0dcf\u0db1\u0dbb\u0dda\u0d9b\u0dd3\u0dba \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
"<p>Final logits layer </p>\n": "<p>\u0d85\u0dc0\u0dc3\u0db1\u0dca\u0db4\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
"<p>Max pooling at end of each block </p>\n": "<p>\u0d91\u0d9a\u0dca\u0d91\u0d9a\u0dca \u0d9a\u0ddc\u0da7\u0dc3 \u0d85\u0dc0\u0dc3\u0dcf\u0db1\u0dba\u0dda \u0db8\u0dd0\u0d9a\u0dca\u0dc3\u0dca \u0dad\u0da7\u0dcf\u0d9a </p>\n",
"<p>Number of channels in each layer in each block </p>\n": "<p>\u0d91\u0d9a\u0dca\u0d91\u0d9a\u0dca \u0d9a\u0ddc\u0da7\u0dc3\u0dd9\u0dc4\u0dd2 \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0dc3\u0dca\u0dae\u0dbb\u0dba\u0dda \u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1 </p>\n",
"<p>Pad and crop </p>\n": "<p>\u0db4\u0dd1\u0da9\u0dca\u0dc3\u0dc4 \u0db6\u0ddd\u0d9c </p>\n",
"<p>RGB channels </p>\n": "<p>RGB\u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf </p>\n",
"<p>Random horizontal flip </p>\n": "<p>\u0d85\u0dc4\u0db9\u0dd4\u0dad\u0dd2\u0dbb\u0dc3\u0dca \u0db4\u0dd9\u0dbb\u0dc5\u0dd3\u0db8 </p>\n",
"<p>Reshape for classification layer </p>\n": "<p>\u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab\u0dc3\u0dca\u0dae\u0dbb\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd0\u0dc0\u0dad \u0dc3\u0d9a\u0dc3\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>The VGG layers </p>\n": "<p>VGG\u0dc3\u0dca\u0dae\u0dbb </p>\n",
"<p>Use CIFAR10 dataset by default </p>\n": "<p>\u0db4\u0dd9\u0dbb\u0db1\u0dd2\u0db8\u0dd2\u0dba\u0dd9\u0db1\u0dcaCIFAR10 \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"CIFAR10 Experiment": "CIFAR10 \u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf",
"This is a reusable trainer for CIFAR10 dataset": "\u0db8\u0dd9\u0dba CIFAR10 \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd0\u0dc0\u0dad \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dc5 \u0dc4\u0dd0\u0d9a\u0dd2 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4\u0dc0\u0d9a\u0dd2"
}
translate_cache/experiments/cifar10.zh.json
+24
View File
@@ -0,0 +1,24 @@
{
"<h1>CIFAR10 Experiment</h1>\n": "<h1>CIFAR10 \u5b9e\u9a8c</h1>\n",
"<h2>Configurations</h2>\n<p>This extends from CIFAR 10 dataset configurations from <a href=\"https://github.com/labmlai/labml/tree/master/helpers\"><span translate=no>_^_0_^_</span></a> and <a href=\"mnist.html\"><span translate=no>_^_1_^_</span></a>.</p>\n": "<h2>\u914d\u7f6e</h2>\n<p>\u8fd9\u662f\u4ece\u548c\u5f00\u59cb\u7684 CIFAR 10 \u6570\u636e\u96c6\u914d\u7f6e\u6269\u5c55<a href=\"https://github.com/labmlai/labml/tree/master/helpers\"><span translate=no>_^_0_^_</span></a>\u800c\u6765\u7684<a href=\"mnist.html\"><span translate=no>_^_1_^_</span></a>\u3002</p>\n",
"<h3>Augmented CIFAR 10 train dataset</h3>\n": "<h3>\u589e\u5f3a\u7684 CIFAR 10 \u8bad\u7ec3\u6570\u636e\u96c6</h3>\n",
"<h3>Non-augmented CIFAR 10 validation dataset</h3>\n": "<h3>\u975e\u589e\u5f3a CIFAR 10 \u9a8c\u8bc1\u6570\u636e\u96c6</h3>\n",
"<h3>VGG model for CIFAR-10 classification</h3>\n": "<h3>\u7528\u4e8e CIFAR-10 \u5206\u7c7b\u7684 VGG \u6a21\u578b</h3>\n",
"<p> </p>\n": "<p></p>\n",
"<p> Convolution and activation combined</p>\n": "<p>\u5377\u79ef\u548c\u6fc0\u6d3b\u76f8\u7ed3\u5408</p>\n",
"<p>5 <span translate=no>_^_0_^_</span> pooling layers will produce a output of size <span translate=no>_^_1_^_</span>. CIFAR 10 image size is <span translate=no>_^_2_^_</span> </p>\n": "<p>5 \u4e2a<span translate=no>_^_0_^_</span>\u6c60\u5316\u56fe\u5c42\u5c06\u751f\u6210\u5927\u5c0f\u4e3a size \u7684\u8f93\u51fa<span translate=no>_^_1_^_</span>\u3002CIFAR 10 \u56fe\u50cf\u5927\u5c0f\u4e3a<span translate=no>_^_2_^_</span></p>\n",
"<p>Convolution, Normalization and Activation layers </p>\n": "<p>\u5377\u79ef\u3001\u5f52\u4e00\u5316\u548c\u6fc0\u6d3b\u5c42</p>\n",
"<p>Create a sequential model with the layers </p>\n": "<p>\u4f7f\u7528\u5c42\u521b\u5efa\u987a\u5e8f\u6a21\u578b</p>\n",
"<p>Final linear layer </p>\n": "<p>\u6700\u540e\u7684\u7ebf\u6027\u5c42</p>\n",
"<p>Final logits layer </p>\n": "<p>\u6700\u540e\u7684 logits \u5c42</p>\n",
"<p>Max pooling at end of each block </p>\n": "<p>\u6bcf\u4e2a\u533a\u5757\u672b\u7aef\u7684\u6700\u5927\u6c60\u6570</p>\n",
"<p>Number of channels in each layer in each block </p>\n": "<p>\u6bcf\u4e2a\u533a\u5757\u4e2d\u6bcf\u5c42\u7684\u901a\u9053\u6570</p>\n",
"<p>Pad and crop </p>\n": "<p>\u586b\u5145\u548c\u88c1\u526a</p>\n",
"<p>RGB channels </p>\n": "<p>RGB \u901a\u9053</p>\n",
"<p>Random horizontal flip </p>\n": "<p>\u968f\u673a\u6c34\u5e73\u7ffb\u8f6c</p>\n",
"<p>Reshape for classification layer </p>\n": "<p>\u4fee\u6539\u5206\u7c7b\u56fe\u5c42\u7684\u5f62\u72b6</p>\n",
"<p>The VGG layers </p>\n": "<p>VGG \u5c42</p>\n",
"<p>Use CIFAR10 dataset by default </p>\n": "<p>\u9ed8\u8ba4\u4f7f\u7528 CIFAR10 \u6570\u636e\u96c6</p>\n",
"CIFAR10 Experiment": "CIFAR10 \u5b9e\u9a8c",
"This is a reusable trainer for CIFAR10 dataset": "\u8fd9\u662f CIFAR10 \u6570\u636e\u96c6\u7684\u53ef\u91cd\u590d\u4f7f\u7528\u7684\u8bad\u7ec3\u5668"
}
translate_cache/experiments/mnist.ja.json
+32
View File
@@ -0,0 +1,32 @@
{
"<h1>MNIST Experiment</h1>\n": "<h1>MNIST \u5b9f\u9a13</h1>\n",
"<h3>Default optimizer configurations</h3>\n": "<h3>\u30c7\u30d5\u30a9\u30eb\u30c8\u306e\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc\u69cb\u6210</h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u521d\u671f\u5316</h3>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u307e\u305f\u306f\u691c\u8a3c\u30b9\u30c6\u30c3\u30d7</h3>\n",
"<p> <a id=\"MNISTConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n": "<p><a id=\"MNISTConfigs\"></a></p>\n<h2>\u30c8\u30ec\u30fc\u30ca\u30fc\u69cb\u6210</h2>\n",
"<p>Accuracy function </p>\n": "<p>\u7cbe\u5ea6\u6a5f\u80fd</p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u30e2\u30b8\u30e5\u30fc\u30eb\u51fa\u529b\u3092\u30ed\u30b0\u306b\u8a18\u9332\u3059\u308b\u30d5\u30c3\u30af\u3092\u8ffd\u52a0</p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u30b9\u30c6\u30fc\u30c8\u30e2\u30b8\u30e5\u30fc\u30eb\u3068\u3057\u3066\u7cbe\u5ea6\u3092\u8ffd\u52a0\u3057\u3066\u304f\u3060\u3055\u3044\u3002\u3053\u306e\u540d\u524d\u306f\u3001RNN \u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3068\u691c\u8a3c\u306e\u9593\u306e\u72b6\u614b\u3092\u4fdd\u5b58\u3059\u308b\u305f\u3081\u306e\u3082\u306e\u306a\u306e\u3067\u3001\u304a\u305d\u3089\u304f\u308f\u304b\u308a\u306b\u304f\u3044\u3067\u3057\u3087\u3046\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u7cbe\u5ea6\u6307\u6a19\u306e\u7d71\u8a08\u60c5\u5831\u304c\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u7528\u3068\u691c\u8a3c\u7528\u306b\u5225\u3005\u306b\u4fdd\u6301\u3055\u308c\u307e\u3059\u3002</p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u7cbe\u5ea6\u306e\u8a08\u7b97\u3068\u8a18\u9332</p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u640d\u5931\u306e\u8a08\u7b97\u3068\u8a18\u9332</p>\n",
"<p>Calculate gradients </p>\n": "<p>\u52fe\u914d\u306e\u8a08\u7b97</p>\n",
"<p>Classification model </p>\n": "<p>\u5206\u985e\u30e2\u30c7\u30eb</p>\n",
"<p>Clear the gradients </p>\n": "<p>\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3\u3092\u30af\u30ea\u30a2</p>\n",
"<p>Get model outputs. </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u53d6\u5f97\u3057\u307e\u3059\u3002</p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u5404\u30a8\u30dd\u30c3\u30af\u306e\u6700\u5f8c\u306e\u30d0\u30c3\u30c1\u3067\u30e2\u30c7\u30eb\u30d1\u30e9\u30e1\u30fc\u30bf\u3068\u52fe\u914d\u3092\u8a18\u9332\u3057\u307e\u3059</p>\n",
"<p>Loss function </p>\n": "<p>\u640d\u5931\u95a2\u6570</p>\n",
"<p>Move data to the device </p>\n": "<p>\u30c7\u30fc\u30bf\u3092\u30c7\u30d0\u30a4\u30b9\u306b\u79fb\u52d5</p>\n",
"<p>Number of epochs to train for </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u5bfe\u8c61\u30a8\u30dd\u30c3\u30af\u306e\u6570</p>\n",
"<p>Number of times to switch between training and validation within an epoch </p>\n": "<p>1 \u3064\u306e\u30a8\u30dd\u30c3\u30af\u5185\u3067\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3068\u691c\u8a3c\u3092\u5207\u308a\u66ff\u3048\u308b\u56de\u6570</p>\n",
"<p>Optimizer </p>\n": "<p>\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc</p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u8ffd\u8de1\u3057\u305f\u30e1\u30c8\u30ea\u30af\u30b9\u3092\u4fdd\u5b58\u3059\u308b</p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u30c8\u30e9\u30c3\u30ab\u30fc\u69cb\u6210\u3092\u8a2d\u5b9a</p>\n",
"<p>Take optimizer step </p>\n": "<p>\u6700\u9069\u5316\u306e\u4e00\u6b69\u3092\u8e0f\u307f\u51fa\u3059</p>\n",
"<p>Train the model </p>\n": "<p>\u30e2\u30c7\u30eb\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0</p>\n",
"<p>Training device </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30d0\u30a4\u30b9</p>\n",
"<p>Training/Evaluation mode </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0/\u8a55\u4fa1\u30e2\u30fc\u30c9</p>\n",
"<p>Update global step (number of samples processed) when in training mode </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30e2\u30fc\u30c9\u6642\u306b\u30b0\u30ed\u30fc\u30d0\u30eb\u30b9\u30c6\u30c3\u30d7 (\u51e6\u7406\u3055\u308c\u305f\u30b5\u30f3\u30d7\u30eb\u6570) \u3092\u66f4\u65b0</p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u30ad\u30e3\u30d7\u30c1\u30e3\u3059\u308b\u304b\u3069\u3046\u304b</p>\n",
"MNIST Experiment": "MNIST \u5b9f\u9a13",
"This is a reusable trainer for MNIST dataset": "\u3053\u308c\u306fMNIST\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u518d\u5229\u7528\u53ef\u80fd\u306a\u30c8\u30ec\u30fc\u30ca\u30fc\u3067\u3059"
}
translate_cache/experiments/mnist.si.json
+32
View File
@@ -0,0 +1,32 @@
{
"<h1>MNIST Experiment</h1>\n": "<h1>MNIST\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8</h1>\n",
"<h3>Default optimizer configurations</h3>\n": "<h3>\u0db4\u0dd9\u0dbb\u0db1\u0dd2\u0db8\u0dd2\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca</h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u0d86\u0dbb\u0db8\u0dca\u0db7\u0d9a\u0d9a\u0dbb\u0dab\u0dba</h3>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0\u0dc4\u0ddd \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db4\u0dd2\u0dba\u0dc0\u0dbb</h3>\n",
"<p> <a id=\"MNISTConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n": "<p> <a id=\"MNISTConfigs\"></a></p>\n<h2>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4\u0db8\u0dcf\u0db1\u0d9a\u0dbb\u0dab</h2>\n",
"<p>Accuracy function </p>\n": "<p>\u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u0db8\u0ddc\u0da9\u0dd2\u0dba\u0dd4\u0dbd\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0d9a\u0ddc\u0d9a\u0dca\u0d9a\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u0dbb\u0dcf\u0da2\u0dca\u0dba\u0db8\u0ddc\u0da9\u0dd2\u0dba\u0dd4\u0dbd\u0dba\u0d9a\u0dca \u0dbd\u0dd9\u0dc3 \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc0 \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1. RNs \u0dc3\u0db3\u0dc4\u0dcf \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0dad\u0dbb \u0dbb\u0dcf\u0da2\u0dca\u0dba\u0dba\u0db1\u0dca \u0d9c\u0db6\u0da9\u0dcf \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0daf\u0dc4\u0dc3\u0dca \u0d9a\u0dbb\u0db1 \u0db6\u0dd0\u0dc0\u0dd2\u0db1\u0dca \u0db1\u0db8 \u0db6\u0ddc\u0dc4\u0ddd \u0dc0\u0dd2\u0da7 \u0dc0\u0dca\u0dba\u0dcf\u0d9a\u0dd6\u0dbd \u0dc0\u0dda. \u0db8\u0dd9\u0dba \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf \u0db8\u0dd9\u0da7\u0dca\u0dbb\u0dd2\u0d9a\u0dca \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dd9\u0db1\u0db8 \u0dad\u0db6\u0dcf \u0d9c\u0db1\u0dd3. </p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u0d9c\u0dab\u0db1\u0dba\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc0\u0dba </p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u0d9c\u0dab\u0db1\u0dba\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dbd\u0ddc\u0d9c\u0dca \u0dc0\u0dd3\u0db8 </p>\n",
"<p>Calculate gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d9c\u0dab\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Classification model </p>\n": "<p>\u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba </p>\n",
"<p>Clear the gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d89\u0dc0\u0dad\u0dca </p>\n",
"<p>Get model outputs. </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1. </p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u0dc3\u0dd1\u0db8\u0dba\u0dd4\u0d9c\u0dbd\u0dba\u0d9a\u0db8 \u0d85\u0dc0\u0dc3\u0dcf\u0db1 \u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dda \u0d86\u0daf\u0dbb\u0dca\u0dc1 \u0db4\u0dbb\u0dcf\u0db8\u0dd2\u0dad\u0dd3\u0db1\u0dca \u0dc3\u0dc4 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Loss function </p>\n": "<p>\u0db4\u0dcf\u0da9\u0dd4\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
"<p>Move data to the device </p>\n": "<p>\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba\u0dc0\u0dd9\u0dad \u0daf\u0dad\u0dca\u0dad \u0d9c\u0dd9\u0db1\u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Number of epochs to train for </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0d91\u0db4\u0ddc\u0da0\u0dca \u0d9c\u0dab\u0db1 </p>\n",
"<p>Number of times to switch between training and validation within an epoch </p>\n": "<p>\u0d91\u0db4\u0ddd\u0da0\u0dca\u0dad\u0dd4\u0dc5 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0dad\u0dbb \u0db8\u0dcf\u0dbb\u0dd4 \u0dc0\u0dd3\u0db8\u0da7 \u0dc0\u0dcf\u0dbb \u0d9c\u0dab\u0db1 </p>\n",
"<p>Optimizer </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0d9a\u0dbb\u0dab\u0dba </p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u0dbd\u0dd4\u0dc4\u0dd4\u0db6\u0dd0\u0db3\u0d87\u0dad\u0dd2 \u0db4\u0dca\u0dbb\u0db8\u0dd2\u0dad\u0dd2\u0d9a \u0dc3\u0dd4\u0dbb\u0d9a\u0dd2\u0db1\u0dca\u0db1 </p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u0da7\u0dca\u0dbb\u0dd0\u0d9a\u0dbb\u0dca\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca \u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 </p>\n",
"<p>Take optimizer step </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab\u0db4\u0dd2\u0dba\u0dc0\u0dbb \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Train the model </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Training device </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba </p>\n",
"<p>Training/Evaluation mode </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0/\u0d87\u0d9c\u0dba\u0dd3\u0db8\u0dca\u0db8\u0dcf\u0daf\u0dd2\u0dbd\u0dd2\u0dba </p>\n",
"<p>Update global step (number of samples processed) when in training mode </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0db4\u0dca\u0dbb\u0d9a\u0dcf\u0dbb\u0dba\u0dda\u0daf\u0dd3 \u0d9c\u0ddd\u0dbd\u0dd3\u0dba \u0db4\u0dd2\u0dba\u0dc0\u0dbb (\u0dc3\u0dd0\u0d9a\u0dc3\u0dd6 \u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd \u0d9c\u0dab\u0db1) \u0dba\u0dcf\u0dc0\u0dad\u0dca\u0d9a\u0dcf\u0dbd\u0dd3\u0db1 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0d9c\u0dca\u0dbb\u0dc4\u0dab\u0dba \u0d9a\u0dbb \u0d9c\u0dad \u0dba\u0dd4\u0dad\u0dd4\u0daf \u0dba\u0db1\u0dca\u0db1 </p>\n",
"MNIST Experiment": "MNIST \u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8",
"This is a reusable trainer for MNIST dataset": "\u0db8\u0dd9\u0db8 MNIST \u0daf\u0dad\u0dca\u0dad \u0dc3\u0db8\u0dd4\u0daf\u0dcf\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd0\u0dc0\u0dad \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbd \u0dc4\u0dd0\u0d9a\u0dd2 \u0dc0\u0db1 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4 \u0dc0\u0dda"
}
translate_cache/experiments/mnist.zh.json
+32
View File
@@ -0,0 +1,32 @@
{
"<h1>MNIST Experiment</h1>\n": "<h1>MNIST \u5b9e\u9a8c</h1>\n",
"<h3>Default optimizer configurations</h3>\n": "<h3>\u9ed8\u8ba4\u4f18\u5316\u5668\u914d\u7f6e</h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u521d\u59cb\u5316</h3>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u57f9\u8bad\u6216\u9a8c\u8bc1\u6b65\u9aa4</h3>\n",
"<p> <a id=\"MNISTConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n": "<p><a id=\"MNISTConfigs\"></a></p>\n<h2>\u8bad\u7ec3\u5668\u914d\u7f6e</h2>\n",
"<p>Accuracy function </p>\n": "<p>\u7cbe\u5ea6\u51fd\u6570</p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u5411\u65e5\u5fd7\u6a21\u5757\u8f93\u51fa\u6dfb\u52a0\u94a9\u5b50</p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u589e\u52a0\u4f5c\u4e3a\u72b6\u6001\u6a21\u5757\u7684\u7cbe\u5ea6\u3002\u8fd9\u4e2a\u540d\u5b57\u53ef\u80fd\u4ee4\u4eba\u56f0\u60d1\uff0c\u56e0\u4e3a\u5b83\u65e8\u5728\u5b58\u50a8 RNN \u7684\u8bad\u7ec3\u548c\u9a8c\u8bc1\u4e4b\u95f4\u7684\u72b6\u6001\u3002\u8fd9\u5c06\u4f7f\u7cbe\u5ea6\u6307\u6807\u7edf\u8ba1\u6570\u636e\u5206\u5f00\uff0c\u4ee5\u4fbf\u8fdb\u884c\u8bad\u7ec3\u548c\u9a8c\u8bc1\u3002</p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u8ba1\u7b97\u548c\u8bb0\u5f55\u7cbe\u5ea6</p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u8ba1\u7b97\u5e76\u8bb0\u5f55\u635f\u5931</p>\n",
"<p>Calculate gradients </p>\n": "<p>\u8ba1\u7b97\u68af\u5ea6</p>\n",
"<p>Classification model </p>\n": "<p>\u5206\u7c7b\u6a21\u578b</p>\n",
"<p>Clear the gradients </p>\n": "<p>\u6e05\u9664\u6e10\u53d8</p>\n",
"<p>Get model outputs. </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u8f93\u51fa\u3002</p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u8bb0\u5f55\u6bcf\u4e2a\u7eaa\u5143\u6700\u540e\u4e00\u6279\u7684\u6a21\u578b\u53c2\u6570\u548c\u68af\u5ea6</p>\n",
"<p>Loss function </p>\n": "<p>\u4e8f\u635f\u51fd\u6570</p>\n",
"<p>Move data to the device </p>\n": "<p>\u5c06\u6570\u636e\u79fb\u52a8\u5230\u8bbe\u5907</p>\n",
"<p>Number of epochs to train for </p>\n": "<p>\u8981\u8bad\u7ec3\u7684\u65f6\u4ee3\u6570</p>\n",
"<p>Number of times to switch between training and validation within an epoch </p>\n": "<p>\u4e00\u4e2a\u7eaa\u5143\u5185\u5728\u8bad\u7ec3\u548c\u9a8c\u8bc1\u4e4b\u95f4\u5207\u6362\u7684\u6b21\u6570</p>\n",
"<p>Optimizer </p>\n": "<p>\u4f18\u5316\u5668</p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u4fdd\u5b58\u8ddf\u8e2a\u7684\u6307\u6807</p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u8bbe\u7f6e\u8ddf\u8e2a\u5668\u914d\u7f6e</p>\n",
"<p>Take optimizer step </p>\n": "<p>\u91c7\u53d6\u4f18\u5316\u5668\u6b65\u9aa4</p>\n",
"<p>Train the model </p>\n": "<p>\u8bad\u7ec3\u6a21\u578b</p>\n",
"<p>Training device </p>\n": "<p>\u8bad\u7ec3\u8bbe\u5907</p>\n",
"<p>Training/Evaluation mode </p>\n": "<p>\u8bad\u7ec3/\u8bc4\u4f30\u6a21\u5f0f</p>\n",
"<p>Update global step (number of samples processed) when in training mode </p>\n": "<p>\u5728\u8bad\u7ec3\u6a21\u5f0f\u4e0b\u66f4\u65b0\u5168\u5c40\u6b65\u957f\uff08\u5904\u7406\u7684\u6837\u672c\u6570\uff09</p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u662f\u5426\u6355\u83b7\u6a21\u578b\u8f93\u51fa</p>\n",
"MNIST Experiment": "MNIST \u5b9e\u9a8c",
"This is a reusable trainer for MNIST dataset": "\u8fd9\u662f MNIST \u6570\u636e\u96c6\u7684\u53ef\u91cd\u590d\u4f7f\u7528\u7684\u8bad\u7ec3\u5668"
}
translate_cache/experiments/nlp_autoregression.ja.json
+69
View File
@@ -0,0 +1,69 @@
{
"<h1>Auto-regressive NLP model trainer</h1>\n": "<h1>\u81ea\u5df1\u56de\u5e30 NLP \u30e2\u30c7\u30eb\u30c8\u30ec\u30fc\u30ca\u30fc</h1>\n",
"<h3>Basic english tokenizer</h3>\n<p>We use character level tokenizer in this experiment. You can switch by setting,</p>\n<span translate=no>_^_0_^_</span><p>in the configurations dictionary when starting the experiment.</p>\n": "<h3>\u30d9\u30fc\u30b7\u30c3\u30af\u30fb\u30a4\u30f3\u30b0\u30ea\u30c3\u30b7\u30e5\u30fb\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc</h3>\n<p>\u3053\u306e\u5b9f\u9a13\u3067\u306f\u3001\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u306e\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc\u3092\u4f7f\u7528\u3057\u307e\u3059\u3002\u8a2d\u5b9a\u3067\u5207\u308a\u66ff\u3048\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u304c\u3001</p>\n<span translate=no>_^_0_^_</span><p>\u5b9f\u9a13\u3092\u958b\u59cb\u3059\u308b\u3068\u304d\u306b\u69cb\u6210\u8f9e\u66f8\u306b\u3042\u308a\u307e\u3059\u3002</p>\n",
"<h3>Character level tokenizer configuration</h3>\n": "<h3>\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u306e\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc\u8a2d\u5b9a</h3>\n",
"<h3>Character level tokenizer</h3>\n": "<h3>\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc</h3>\n",
"<h3>Cross entropy loss</h3>\n": "<h3>\u30af\u30ed\u30b9\u30a8\u30f3\u30c8\u30ed\u30d4\u30fc\u640d\u5931</h3>\n",
"<h3>Default <a href=\"../optimizers/configs.html\">optimizer configurations</a></h3>\n": "<h3><a href=\"../optimizers/configs.html\">\u30c7\u30d5\u30a9\u30eb\u30c8\u306e\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc\u69cb\u6210</a></h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u521d\u671f\u5316</h3>\n",
"<h3>Sampling function to generate samples periodically while training</h3>\n": "<h3>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u4e2d\u306b\u5b9a\u671f\u7684\u306b\u30b5\u30f3\u30d7\u30eb\u3092\u751f\u6210\u3059\u308b\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u6a5f\u80fd</h3>\n",
"<h3>Sequential training data loader</h3>\n": "<h3>\u30b7\u30fc\u30b1\u30f3\u30b7\u30e3\u30eb\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</h3>\n",
"<h3>Sequential validation data loader</h3>\n": "<h3>\u30b7\u30fc\u30b1\u30f3\u30b7\u30e3\u30eb\u691c\u8a3c\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</h3>\n",
"<h3>Shuffled training data loader</h3>\n": "<h3>\u30b7\u30e3\u30c3\u30d5\u30eb\u3055\u308c\u305f\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</h3>\n",
"<h3>Shuffled validation data loader</h3>\n": "<h3>\u30b7\u30e3\u30c3\u30d5\u30eb\u3055\u308c\u305f\u691c\u8a3c\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</h3>\n",
"<h3>Tiny Shakespeare dataset</h3>\n<p>It will download from the url if not present</p>\n": "<h3>\u5c0f\u3055\u306a\u30b7\u30a7\u30a4\u30af\u30b9\u30d4\u30a2\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</h3>\n<p>\u5b58\u5728\u3057\u306a\u3044\u5834\u5408\u306f URL \u304b\u3089\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3057\u307e\u3059</p>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u307e\u305f\u306f\u691c\u8a3c\u30b9\u30c6\u30c3\u30d7</h3>\n",
"<h3>Transpose batch</h3>\n<p><span translate=no>_^_0_^_</span> collects the batches on the first dimension. We need to transpose it to be sequence first.</p>\n": "<h3>\u30c8\u30e9\u30f3\u30b9\u30dd\u30fc\u30ba\u30d0\u30c3\u30c1</h3>\n<p><span translate=no>_^_0_^_</span>\u7b2c 1 \u6b21\u5143\u306e\u30d0\u30c3\u30c1\u3092\u53ce\u96c6\u3057\u307e\u3059\u3002\u6700\u521d\u306b\u30b7\u30fc\u30b1\u30f3\u30b9\u306b\u306a\u308b\u3088\u3046\u306b\u30c8\u30e9\u30f3\u30b9\u30dd\u30fc\u30ba\u3059\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u3059</p>\u3002\n",
"<p> <a id=\"NLPAutoRegressionConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n<p>This has the basic configurations for NLP auto-regressive task training. All the properties are configurable.</p>\n": "<p><a id=\"NLPAutoRegressionConfigs\"></a></p>\n<h2>\u30c8\u30ec\u30fc\u30ca\u30fc\u69cb\u6210</h2>\n<p>\u3053\u308c\u306b\u306f\u3001NLP\u81ea\u5df1\u56de\u5e30\u30bf\u30b9\u30af\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306e\u57fa\u672c\u69cb\u6210\u304c\u3042\u308a\u307e\u3059\u3002\u3059\u3079\u3066\u306e\u30d7\u30ed\u30d1\u30c6\u30a3\u306f\u8a2d\u5b9a\u53ef\u80fd\u3067\u3059\u3002</p>\n",
"<p> Get number of tokens</p>\n": "<p>\u30c8\u30fc\u30af\u30f3\u306e\u6570\u3092\u53d6\u5f97</p>\n",
"<p>Accuracy function </p>\n": "<p>\u7cbe\u5ea6\u6a5f\u80fd</p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u30e2\u30b8\u30e5\u30fc\u30eb\u51fa\u529b\u3092\u30ed\u30b0\u306b\u8a18\u9332\u3059\u308b\u30d5\u30c3\u30af\u3092\u8ffd\u52a0</p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u30b9\u30c6\u30fc\u30c8\u30e2\u30b8\u30e5\u30fc\u30eb\u3068\u3057\u3066\u7cbe\u5ea6\u3092\u8ffd\u52a0\u3057\u3066\u304f\u3060\u3055\u3044\u3002\u3053\u306e\u540d\u524d\u306f\u3001RNN \u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3068\u691c\u8a3c\u306e\u9593\u306e\u72b6\u614b\u3092\u4fdd\u5b58\u3059\u308b\u305f\u3081\u306e\u3082\u306e\u306a\u306e\u3067\u3001\u304a\u305d\u3089\u304f\u308f\u304b\u308a\u306b\u304f\u3044\u3067\u3057\u3087\u3046\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u7cbe\u5ea6\u6307\u6a19\u306e\u7d71\u8a08\u60c5\u5831\u304c\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u7528\u3068\u691c\u8a3c\u7528\u306b\u5225\u3005\u306b\u4fdd\u6301\u3055\u308c\u307e\u3059\u3002</p>\n",
"<p>Add the prediction for logging </p>\n": "<p>\u30ed\u30ae\u30f3\u30b0\u7528\u306e\u4e88\u6e2c\u3092\u8ffd\u52a0</p>\n",
"<p>Add the prediction to prompt </p>\n": "<p>\u4e88\u6e2c\u3092\u30d7\u30ed\u30f3\u30d7\u30c8\u306b\u8ffd\u52a0</p>\n",
"<p>Autoregressive model </p>\n": "<p>\u81ea\u5df1\u56de\u5e30\u30e2\u30c7\u30eb</p>\n",
"<p>Batch size </p>\n": "<p>\u30d0\u30c3\u30c1\u30b5\u30a4\u30ba</p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u7cbe\u5ea6\u306e\u8a08\u7b97\u3068\u8a18\u9332</p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u640d\u5931\u306e\u8a08\u7b97\u3068\u8a18\u9332</p>\n",
"<p>Calculate gradients </p>\n": "<p>\u52fe\u914d\u306e\u8a08\u7b97</p>\n",
"<p>Clear the gradients </p>\n": "<p>\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3\u3092\u30af\u30ea\u30a2</p>\n",
"<p>Clip gradients </p>\n": "<p>\u30af\u30ea\u30c3\u30d7\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3</p>\n",
"<p>Collect output for printing </p>\n": "<p>\u5370\u5237\u7528\u306e\u51fa\u529b\u3092\u53ce\u96c6</p>\n",
"<p>Data loaders shuffle with replacement </p>\n": "<p>\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc\u306f\u4ea4\u63db\u6642\u306b\u30b7\u30e3\u30c3\u30d5\u30eb\u3055\u308c\u307e\u3059</p>\n",
"<p>Get model outputs. It&#x27;s returning a tuple for states when using RNNs. This is not implemented yet. \ud83d\ude1c </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u53d6\u5f97\u3057\u307e\u3059\u3002RNN \u3092\u4f7f\u7528\u3059\u308b\u5834\u5408\u306f\u30b9\u30c6\u30fc\u30c8\u306e\u30bf\u30d7\u30eb\u3092\u8fd4\u3057\u307e\u3059\u3002\u3053\u308c\u306f\u307e\u3060\u5b9f\u88c5\u3055\u308c\u3066\u3044\u307e\u305b\u3093\u3002\ud83d\ude1c</p>\n",
"<p>Get the model output </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u53d6\u5f97</p>\n",
"<p>Get the model prediction (greedy) </p>\n": "<p>\u30e2\u30c7\u30eb\u4e88\u6e2c\u3092\u53d6\u5f97 (\u6b32\u5f35\u308a)</p>\n",
"<p>Gradient clipping </p>\n": "<p>\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3\u30af\u30ea\u30c3\u30d4\u30f3\u30b0</p>\n",
"<p>Length of the sequence, or context size </p>\n": "<p>\u30b7\u30fc\u30b1\u30f3\u30b9\u306e\u9577\u3055\u3001\u307e\u305f\u306f\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u30b5\u30a4\u30ba</p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u5404\u30a8\u30dd\u30c3\u30af\u306e\u6700\u5f8c\u306e\u30d0\u30c3\u30c1\u3067\u30e2\u30c7\u30eb\u30d1\u30e9\u30e1\u30fc\u30bf\u3068\u52fe\u914d\u3092\u8a18\u9332\u3057\u307e\u3059</p>\n",
"<p>Loss function </p>\n": "<p>\u640d\u5931\u95a2\u6570</p>\n",
"<p>Model embedding size </p>\n": "<p>\u30e2\u30c7\u30eb\u57cb\u3081\u8fbc\u307f\u30b5\u30a4\u30ba</p>\n",
"<p>Move data to the device </p>\n": "<p>\u30c7\u30fc\u30bf\u3092\u30c7\u30d0\u30a4\u30b9\u306b\u79fb\u52d5</p>\n",
"<p>Number of token in vocabulary </p>\n": "<p>\u30dc\u30ad\u30e3\u30d6\u30e9\u30ea\u30fc\u5185\u306e\u30c8\u30fc\u30af\u30f3\u306e\u6570</p>\n",
"<p>Optimizer </p>\n": "<p>\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc</p>\n",
"<p>Override to calculate and log other metrics </p>\n": "<p>\u30aa\u30fc\u30d0\u30fc\u30e9\u30a4\u30c9\u3057\u3066\u4ed6\u306e\u6307\u6a19\u3092\u8a08\u7b97\u3057\u3066\u8a18\u9332\u3059\u308b</p>\n",
"<p>Print the sampled output </p>\n": "<p>\u30b5\u30f3\u30d7\u30eb\u51fa\u529b\u3092\u5370\u5237\u3059\u308b</p>\n",
"<p>Sample 25 tokens </p>\n": "<p>25\u30c8\u30fc\u30af\u30f3\u306e\u30b5\u30f3\u30d7\u30eb</p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u8ffd\u8de1\u3057\u305f\u30e1\u30c8\u30ea\u30af\u30b9\u3092\u4fdd\u5b58\u3059\u308b</p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u30c8\u30e9\u30c3\u30ab\u30fc\u69cb\u6210\u3092\u8a2d\u5b9a</p>\n",
"<p>Set training/eval mode </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0/\u8a55\u4fa1\u30e2\u30fc\u30c9\u306e\u8a2d\u5b9a</p>\n",
"<p>Stack the batch along the second dimension <span translate=no>_^_0_^_</span> </p>\n": "<p>2 \u756a\u76ee\u306e\u6b21\u5143\u306b\u6cbf\u3063\u3066\u30d0\u30c3\u30c1\u3092\u7a4d\u307f\u91cd\u306d\u308b <span translate=no>_^_0_^_</span></p>\n",
"<p>Starting prompt </p>\n": "<p>\u8d77\u52d5\u30d7\u30ed\u30f3\u30d7\u30c8</p>\n",
"<p>Take optimizer step </p>\n": "<p>\u6700\u9069\u5316\u306e\u4e00\u6b69\u3092\u8e0f\u307f\u51fa\u3059</p>\n",
"<p>Text dataset </p>\n": "<p>\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</p>\n",
"<p>Text prompt to start sampling (for illustration) </p>\n": "<p>\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3092\u958b\u59cb\u3059\u308b\u30c6\u30ad\u30b9\u30c8\u30d7\u30ed\u30f3\u30d7\u30c8 (\u8aac\u660e\u7528)</p>\n",
"<p>The token separator when sampling (blank for character level tokenization) </p>\n": "<p>\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u6642\u306e\u30c8\u30fc\u30af\u30f3\u30bb\u30d1\u30ec\u30fc\u30bf\u30fc (\u6587\u5b57\u30ec\u30d9\u30eb\u306e\u30c8\u30fc\u30af\u30f3\u5316\u306e\u5834\u5408\u306f\u7a7a\u767d)</p>\n",
"<p>Tokenize the prompt </p>\n": "<p>\u30d7\u30ed\u30f3\u30d7\u30c8\u3092\u30c8\u30fc\u30af\u30f3\u5316</p>\n",
"<p>Tokenizer </p>\n": "<p>\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc</p>\n",
"<p>Train the model </p>\n": "<p>\u30e2\u30c7\u30eb\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0</p>\n",
"<p>Training data loader </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
"<p>Training device </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30d0\u30a4\u30b9</p>\n",
"<p>Update global step (number of tokens processed) when in training mode </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30e2\u30fc\u30c9\u6642\u306b\u30b0\u30ed\u30fc\u30d0\u30eb\u30b9\u30c6\u30c3\u30d7 (\u51e6\u7406\u3055\u308c\u305f\u30c8\u30fc\u30af\u30f3\u306e\u6570) \u3092\u66f4\u65b0</p>\n",
"<p>Validation data loader </p>\n": "<p>\u691c\u8a3c\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u30ad\u30e3\u30d7\u30c1\u30e3\u3059\u308b\u304b\u3069\u3046\u304b</p>\n",
"<p>Whether to log model activations (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u30e2\u30c7\u30eb\u306e\u30a2\u30af\u30c6\u30a3\u30d9\u30fc\u30b7\u30e7\u30f3\u3092\u30ed\u30b0\u306b\u8a18\u9332\u3059\u308b\u304b\u3069\u3046\u304b (\u30a8\u30dd\u30c3\u30af\u3054\u3068\u306b 1 \u56de)\u3002\u3053\u308c\u3089\u306f\u30ec\u30a4\u30e4\u30fc\u3054\u3068\u306e\u7d71\u8a08\u60c5\u5831\u3092\u307e\u3068\u3081\u305f\u3082\u306e\u3067\u3059\u304c\u3001\u305d\u308c\u3067\u3082\u975e\u5e38\u306b\u6df1\u3044\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u591a\u304f\u306e\u6307\u6a19\u306b\u3064\u306a\u304c\u308b\u53ef\u80fd\u6027\u304c\u3042\u308a\u307e\u3059</p>\u3002\n",
"<p>Whether to log model parameters and gradients (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u30e2\u30c7\u30eb\u30d1\u30e9\u30e1\u30fc\u30bf\u30fc\u3068\u52fe\u914d\u3092\u8a18\u9332\u3059\u308b\u304b\u3069\u3046\u304b (\u30a8\u30dd\u30c3\u30af\u3054\u3068\u306b 1 \u56de)\u3002\u3053\u308c\u3089\u306f\u30ec\u30a4\u30e4\u30fc\u3054\u3068\u306e\u7d71\u8a08\u60c5\u5831\u3092\u307e\u3068\u3081\u305f\u3082\u306e\u3067\u3059\u304c\u3001\u305d\u308c\u3067\u3082\u975e\u5e38\u306b\u6df1\u3044\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u591a\u304f\u306e\u6307\u6a19\u306b\u3064\u306a\u304c\u308b\u53ef\u80fd\u6027\u304c\u3042\u308a\u307e\u3059</p>\u3002\n",
"<p>Whether to periodically save models </p>\n": "<p>\u30e2\u30c7\u30eb\u3092\u5b9a\u671f\u7684\u306b\u4fdd\u5b58\u3059\u308b\u304b\u3069\u3046\u304b</p>\n",
"NLP auto-regression trainer": "NLP \u81ea\u52d5\u56de\u5e30\u30c8\u30ec\u30fc\u30ca\u30fc",
"This is a reusable trainer for auto-regressive tasks": "\u3053\u308c\u306f\u81ea\u5df1\u56de\u5e30\u30bf\u30b9\u30af\u7528\u306e\u518d\u5229\u7528\u53ef\u80fd\u306a\u30c8\u30ec\u30fc\u30ca\u30fc\u3067\u3059"
}
translate_cache/experiments/nlp_autoregression.si.json
+69
View File
@@ -0,0 +1,69 @@
{
"<h1>Auto-regressive NLP model trainer</h1>\n": "<h1>\u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0d9c\u0dcf\u0db8\u0dd3 \u0d91\u0db1\u0dca\u0d91\u0dbd\u0dca\u0db4\u0dd3 \u0d86\u0d9a\u0dd8\u0dad\u0dd2 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4</h1>\n",
"<h3>Basic english tokenizer</h3>\n<p>We use character level tokenizer in this experiment. You can switch by setting,</p>\n<span translate=no>_^_0_^_</span><p>in the configurations dictionary when starting the experiment.</p>\n": "<h3>\u0db8\u0dd6\u0dbd\u0dd2\u0d9a\u0d89\u0d82\u0d9c\u0dca\u0dbb\u0dd3\u0dc3\u0dd2 \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca</h3>\n<p>\u0db8\u0dd9\u0db8\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8\u0dda\u0daf\u0dd3 \u0d85\u0db4\u0dd2 \u0da0\u0dbb\u0dd2\u0dad \u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db8\u0dd4. \u0dc3\u0dd0\u0d9a\u0dc3\u0dd3\u0db8\u0dd9\u0db1\u0dca \u0d94\u0db6\u0da7 \u0db8\u0dcf\u0dbb\u0dd4 \u0dc0\u0dd2\u0dba \u0dc4\u0dd0\u0d9a\u0dd2\u0dba,</p>\n<span translate=no>_^_0_^_</span><p>\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf\u0db6\u0dd0\u0dbd\u0dd3\u0db8 \u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda\u0daf\u0dd3 \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0dc1\u0db6\u0dca\u0daf\u0d9a\u0ddd\u0dc2\u0dba\u0dda. </p>\n",
"<h3>Character level tokenizer configuration</h3>\n": "<h3>\u0d85\u0d9a\u0dca\u0dc2\u0dbb\u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba</h3>\n",
"<h3>Character level tokenizer</h3>\n": "<h3>\u0d85\u0d9a\u0dca\u0dc2\u0dbb\u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca</h3>\n",
"<h3>Cross entropy loss</h3>\n": "<h3>\u0dc4\u0dbb\u0dc3\u0dca\u0d91\u0db1\u0dca\u0da7\u0dca\u0dbb\u0ddc\u0db4\u0dd2 \u0db1\u0dd0\u0dad\u0dd2\u0dc0\u0dd3\u0db8</h3>\n",
"<h3>Default <a href=\"../optimizers/configs.html\">optimizer configurations</a></h3>\n": "<h3>\u0db4\u0dd9\u0dbb\u0db1\u0dd2\u0db8\u0dd2 <a href=\"../optimizers/configs.html\">\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca</a></h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u0d86\u0dbb\u0db8\u0dca\u0db7\u0d9a\u0d9a\u0dbb\u0dab\u0dba</h3>\n",
"<h3>Sampling function to generate samples periodically while training</h3>\n": "<h3>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0\u0d85\u0dad\u0dbb\u0dad\u0dd4\u0dbb \u0dc0\u0dbb\u0dd2\u0db1\u0dca \u0dc0\u0dbb \u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd \u0da2\u0db1\u0db1\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d9a\u0dcf\u0dbb\u0dca\u0dba\u0dba</h3>\n",
"<h3>Sequential training data loader</h3>\n": "<h3>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0daf\u0dad\u0dca\u0dad \u0d9a\u0dcf\u0dbb\u0d9a\u0dba</h3>\n",
"<h3>Sequential validation data loader</h3>\n": "<h3>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0daf\u0dad\u0dca\u0dad \u0d9a\u0dcf\u0dbb\u0d9a\u0dba</h3>\n",
"<h3>Shuffled training data loader</h3>\n": "<h3>\u0db8\u0dcf\u0dbb\u0dd4\u0d9a\u0dc5 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8</h3>\n",
"<h3>Shuffled validation data loader</h3>\n": "<h3>\u0db8\u0dcf\u0dbb\u0dd4\u0d9a\u0dc5 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8</h3>\n",
"<h3>Tiny Shakespeare dataset</h3>\n<p>It will download from the url if not present</p>\n": "<h3>\u0d9a\u0dd4\u0da9\u0dcf\u0dc2\u0dda\u0d9a\u0dca\u0dc3\u0dca\u0db4\u0dd2\u0dba\u0dbb\u0dca \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h3>\n<p>\u0db1\u0ddc\u0db8\u0dd0\u0dad\u0dd2\u0db1\u0db8\u0dca \u0d91\u0dba url \u0d91\u0d9a\u0dd9\u0db1\u0dca \u0db6\u0dcf\u0d9c\u0dad \u0d9a\u0dbb\u0db1\u0dd4 \u0d87\u0dad</p>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0\u0dc4\u0ddd \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db4\u0dd2\u0dba\u0dc0\u0dbb</h3>\n",
"<h3>Transpose batch</h3>\n<p><span translate=no>_^_0_^_</span> collects the batches on the first dimension. We need to transpose it to be sequence first.</p>\n": "<h3>\u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dc3\u0db8\u0dca\u0db4\u0dca\u0dbb\u0dda\u0dc2\u0dab\u0dba</h3>\n<p><span translate=no>_^_0_^_</span> \u0db4\u0dc5\u0db8\u0dd4 \u0db8\u0dcf\u0db1\u0dba \u0db8\u0dad \u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dca \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0dba\u0dd2. \u0d85\u0db4\u0dd2 \u0d91\u0dba \u0db4\u0dc5\u0db8\u0dd4\u0dc0 \u0d85\u0db1\u0dd4\u0db4\u0dd2\u0dc5\u0dd2\u0dc0\u0dd9\u0dbd\u0da7 \u0dc3\u0db8\u0dca\u0db4\u0dca\u0dbb\u0dda\u0dc2\u0dab\u0dba \u0d9a\u0dc5 \u0dba\u0dd4\u0dad\u0dd4\u0dba. </p>\n",
"<p> <a id=\"NLPAutoRegressionConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n<p>This has the basic configurations for NLP auto-regressive task training. All the properties are configurable.</p>\n": "<p> <a id=\"NLPAutoRegressionConfigs\"></a></p>\n<h2>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4\u0db8\u0dcf\u0db1\u0d9a\u0dbb\u0dab</h2>\n<p>\u0d91\u0db1\u0dca\u0d91\u0dbd\u0dca\u0db4\u0dd3\u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba-\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0d9c\u0dcf\u0db8\u0dd3 \u0d9a\u0dcf\u0dbb\u0dca\u0dba \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0db3\u0dc4\u0dcf \u0db8\u0dd6\u0dbd\u0dd2\u0d9a \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca \u0db8\u0dd9\u0dba\u0da7 \u0d87\u0dad. \u0dc3\u0dd2\u0dba\u0dbd\u0dd4\u0db8 \u0d9c\u0dd4\u0dab\u0dcf\u0d82\u0d9c \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0d9c\u0dad \u0d9a\u0dc5 \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
"<p> Get number of tokens</p>\n": "<p> \u0da7\u0ddd\u0d9a\u0db1\u0d9c\u0dab\u0db1 \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1</p>\n",
"<p>Accuracy function </p>\n": "<p>\u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u0db8\u0ddc\u0da9\u0dd2\u0dba\u0dd4\u0dbd\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0d9a\u0ddc\u0d9a\u0dca\u0d9a\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u0dbb\u0dcf\u0da2\u0dca\u0dba\u0db8\u0ddc\u0da9\u0dd2\u0dba\u0dd4\u0dbd\u0dba\u0d9a\u0dca \u0dbd\u0dd9\u0dc3 \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc0 \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1. RNs \u0dc3\u0db3\u0dc4\u0dcf \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0dad\u0dbb \u0dbb\u0dcf\u0da2\u0dca\u0dba\u0dba\u0db1\u0dca \u0d9c\u0db6\u0da9\u0dcf \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0daf\u0dc4\u0dc3\u0dca \u0d9a\u0dbb\u0db1 \u0db6\u0dd0\u0dc0\u0dd2\u0db1\u0dca \u0db1\u0db8 \u0db6\u0ddc\u0dc4\u0ddd \u0dc0\u0dd2\u0da7 \u0dc0\u0dca\u0dba\u0dcf\u0d9a\u0dd6\u0dbd \u0dc0\u0dda. \u0db8\u0dd9\u0dba \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf \u0db8\u0dd9\u0da7\u0dca\u0dbb\u0dd2\u0d9a\u0dca \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dd9\u0db1\u0db8 \u0dad\u0db6\u0dcf \u0d9c\u0db1\u0dd3. </p>\n",
"<p>Add the prediction for logging </p>\n": "<p>\u0dbd\u0ddc\u0d9c\u0dca\u0dc0\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d85\u0db1\u0dcf\u0dc0\u0dd0\u0d9a\u0dd2\u0dba \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Add the prediction to prompt </p>\n": "<p>\u0dc0\u0dd2\u0db8\u0dc3\u0dd4\u0db8\u0da7\u0d85\u0db1\u0dcf\u0dc0\u0dd0\u0d9a\u0dd2\u0dba \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Autoregressive model </p>\n": "<p>\u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0d9c\u0dcf\u0db8\u0dd3 \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba </p>\n",
"<p>Batch size </p>\n": "<p>\u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dca\u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u0d9c\u0dab\u0db1\u0dba\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc0\u0dba </p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u0d9c\u0dab\u0db1\u0dba\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dbd\u0ddc\u0d9c\u0dca \u0dc0\u0dd3\u0db8 </p>\n",
"<p>Calculate gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d9c\u0dab\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Clear the gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d89\u0dc0\u0dad\u0dca </p>\n",
"<p>Clip gradients </p>\n": "<p>\u0d9a\u0dca\u0dbd\u0dd2\u0db4\u0dca\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a </p>\n",
"<p>Collect output for printing </p>\n": "<p>\u0db8\u0dd4\u0daf\u0dca\u0dbb\u0dab\u0dba\u0dc3\u0db3\u0dc4\u0dcf \u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Data loaders shuffle with replacement </p>\n": "<p>\u0daf\u0dad\u0dca\u0dad\u0d9a\u0dcf\u0dbb\u0d9a\u0dba\u0db1\u0dca \u0d86\u0daf\u0dda\u0dc1 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db8\u0d9f \u0db8\u0dcf\u0dbb\u0dd4 \u0dc0\u0dda </p>\n",
"<p>Get model outputs. It&#x27;s returning a tuple for states when using RNNs. This is not implemented yet. \ud83d\ude1c </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1. \u0d86\u0dbb\u0dca\u0d91\u0db1\u0dca\u0d91\u0dc3\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db1 \u0dc0\u0dd2\u0da7 \u0d91\u0dba \u0db4\u0dca\u0dbb\u0dcf\u0db1\u0dca\u0dad \u0dc3\u0db3\u0dc4\u0dcf \u0da7\u0dd6\u0dbd\u0dca \u0d91\u0d9a\u0d9a\u0dca \u0db1\u0dd0\u0dc0\u0dad \u0dbd\u0db6\u0dcf \u0daf\u0dd9\u0dba\u0dd2. \u0db8\u0dd9\u0dba \u0dad\u0dc0\u0db8 \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dbb \u0db1\u0dd0\u0dad \ud83d\ude1c </p>\n",
"<p>Get the model output </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Get the model prediction (greedy) </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0d85\u0db1\u0dcf\u0dc0\u0dd0\u0d9a\u0dd2\u0dba \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 (\u0d9a\u0dd1\u0daf\u0dbb) </p>\n",
"<p>Gradient clipping </p>\n": "<p>\u0dc1\u0dca\u0dbb\u0dda\u0dab\u0dd2\u0dba\u0dda\u0d9a\u0dca\u0dbd\u0dd2\u0db4\u0dd2\u0db1\u0dca </p>\n",
"<p>Length of the sequence, or context size </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dba\u0dda\u0daf\u0dd2\u0d9c, \u0dc4\u0ddd \u0dc3\u0db1\u0dca\u0daf\u0dbb\u0dca\u0db7\u0dba \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u0dc3\u0dd1\u0db8\u0dba\u0dd4\u0d9c\u0dbd\u0dba\u0d9a\u0db8 \u0d85\u0dc0\u0dc3\u0dcf\u0db1 \u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dda \u0d86\u0daf\u0dbb\u0dca\u0dc1 \u0db4\u0dbb\u0dcf\u0db8\u0dd2\u0dad\u0dd3\u0db1\u0dca \u0dc3\u0dc4 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Loss function </p>\n": "<p>\u0db4\u0dcf\u0da9\u0dd4\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
"<p>Model embedding size </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0d9a\u0dcf\u0dc0\u0dd0\u0daf\u0dca\u0daf\u0dd3\u0db8\u0dda \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
"<p>Move data to the device </p>\n": "<p>\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba\u0dc0\u0dd9\u0dad \u0daf\u0dad\u0dca\u0dad \u0d9c\u0dd9\u0db1\u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Number of token in vocabulary </p>\n": "<p>\u0dc0\u0da0\u0db1\u0db8\u0dcf\u0dbd\u0dcf\u0dc0\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dca \u0d9c\u0dab\u0db1 </p>\n",
"<p>Optimizer </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0d9a\u0dbb\u0dab\u0dba </p>\n",
"<p>Override to calculate and log other metrics </p>\n": "<p>\u0dc0\u0dd9\u0db1\u0dad\u0dca\u0db4\u0dca\u0dbb\u0db8\u0dd2\u0dad\u0dd2\u0d9a \u0d9c\u0dab\u0db1\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0dc4\u0dcf \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0d85\u0db7\u0dd2\u0db6\u0dc0\u0dcf \u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Print the sampled output </p>\n": "<p>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba \u0db8\u0dd4\u0daf\u0dca\u0dbb\u0dab\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Sample 25 tokens </p>\n": "<p>\u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd25 \u0da7\u0ddd\u0d9a\u0db1 </p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u0dbd\u0dd4\u0dc4\u0dd4\u0db6\u0dd0\u0db3\u0d87\u0dad\u0dd2 \u0db4\u0dca\u0dbb\u0db8\u0dd2\u0dad\u0dd2\u0d9a \u0dc3\u0dd4\u0dbb\u0d9a\u0dd2\u0db1\u0dca\u0db1 </p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u0da7\u0dca\u0dbb\u0dd0\u0d9a\u0dbb\u0dca\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca \u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 </p>\n",
"<p>Set training/eval mode </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4/eval\u0db8\u0dcf\u0daf\u0dd2\u0dbd\u0dd2\u0dba \u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 </p>\n",
"<p>Stack the batch along the second dimension <span translate=no>_^_0_^_</span> </p>\n": "<p>\u0daf\u0dd9\u0dc0\u0db1\u0db8\u0dcf\u0db1\u0dba \u0d94\u0dc3\u0dca\u0dc3\u0dda \u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8 \u0d9c\u0ddc\u0da9\u0d9c\u0dc3\u0db1\u0dca\u0db1 <span translate=no>_^_0_^_</span> </p>\n",
"<p>Starting prompt </p>\n": "<p>\u0dc0\u0dd2\u0db8\u0dc3\u0dd4\u0db8\u0d9a\u0dca\u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 </p>\n",
"<p>Take optimizer step </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab\u0db4\u0dd2\u0dba\u0dc0\u0dbb \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Text dataset </p>\n": "<p>\u0db4\u0dd9\u0dc5\u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba </p>\n",
"<p>Text prompt to start sampling (for illustration) </p>\n": "<p>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd3\u0db8\u0dca\u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0db4\u0dd9\u0dc5 \u0dc0\u0dd2\u0db8\u0dc3\u0dd4\u0db8 (\u0db1\u0dd2\u0daf\u0dbb\u0dca\u0dc1\u0db1\u0dba \u0dc3\u0db3\u0dc4\u0dcf) </p>\n",
"<p>The token separator when sampling (blank for character level tokenization) </p>\n": "<p>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd3\u0db8\u0dda\u0daf\u0dd3\u0da7\u0ddd\u0d9a\u0db1\u0dca \u0db6\u0dd9\u0daf\u0dd4\u0db8\u0dca\u0d9a\u0dbb\u0dd4 (\u0d85\u0d9a\u0dca\u0dc2\u0dbb \u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dd3\u0d9a\u0dbb\u0dab\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0dc4\u0dd2\u0dc3\u0dca) </p>\n",
"<p>Tokenize the prompt </p>\n": "<p>\u0dc0\u0dd2\u0db8\u0dc3\u0dd4\u0db8\u0da7\u0ddd\u0d9a\u0dd9\u0db1\u0dca\u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Tokenizer </p>\n": "<p>\u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca </p>\n",
"<p>Train the model </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Training data loader </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 </p>\n",
"<p>Training device </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba </p>\n",
"<p>Update global step (number of tokens processed) when in training mode </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0db4\u0dca\u0dbb\u0d9a\u0dcf\u0dbb\u0dba\u0dda\u0daf\u0dd3 \u0d9c\u0ddd\u0dbd\u0dd3\u0dba \u0db4\u0dd2\u0dba\u0dc0\u0dbb \u0dba\u0dcf\u0dc0\u0dad\u0dca\u0d9a\u0dcf\u0dbd\u0dd3\u0db1 \u0d9a\u0dbb\u0db1\u0dca\u0db1 (\u0dc3\u0dd0\u0d9a\u0dc3\u0dd6 \u0da7\u0ddd\u0d9a\u0db1 \u0d9c\u0dab\u0db1) </p>\n",
"<p>Validation data loader </p>\n": "<p>\u0dc0\u0dbd\u0d82\u0d9c\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 </p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0d9c\u0dca\u0dbb\u0dc4\u0dab\u0dba \u0d9a\u0dbb \u0d9c\u0dad \u0dba\u0dd4\u0dad\u0dd4\u0daf \u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Whether to log model activations (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0dc3\u0d9a\u0dca\u0dbb\u0dd2\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dca \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dc5 \u0dba\u0dd4\u0dad\u0dd4\u0daf \u0dba\u0db1\u0dca\u0db1 (\u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d8a\u0db4\u0ddd\u0da0\u0dca \u0d91\u0d9a\u0d9a\u0da7 \u0dc0\u0dbb\u0d9a\u0dca). \u0db8\u0dda\u0dc0\u0dcf \u0dc3\u0dca\u0dae\u0dbb\u0dba\u0d9a\u0da7 \u0dc3\u0dcf\u0dbb\u0dcf\u0d82\u0dc1\u0d9c\u0dad \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dda, \u0db1\u0db8\u0dd4\u0dad\u0dca \u0d91\u0dba \u0dad\u0dc0\u0db8\u0dad\u0dca \u0d89\u0dad\u0dcf \u0d9c\u0dd0\u0db9\u0dd4\u0dbb\u0dd4 \u0da2\u0dcf\u0dbd \u0dc3\u0db3\u0dc4\u0dcf \u0db6\u0ddc\u0dc4\u0ddd \u0daf\u0dbb\u0dca\u0dc1\u0d9a \u0dc0\u0dbd\u0da7 \u0dc4\u0dda\u0dad\u0dd4 \u0dc0\u0dd2\u0dba \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
"<p>Whether to log model parameters and gradients (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dbb\u0dcf\u0db8\u0dd2\u0dad\u0dd3\u0db1\u0dca \u0dc3\u0dc4 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0dbd\u0d9d\u0dd4-\u0dc3\u0da7\u0dc4\u0db1 \u0dba\u0db1\u0dca\u0db1 (\u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0dc0\u0dbb\u0d9a\u0dca). \u0db8\u0dda\u0dc0\u0dcf \u0dc3\u0dca\u0dae\u0dbb\u0dba\u0d9a\u0da7 \u0dc3\u0dcf\u0dbb\u0dcf\u0d82\u0dc1\u0d9c\u0dad \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dda, \u0db1\u0db8\u0dd4\u0dad\u0dca \u0d91\u0dba \u0dad\u0dc0\u0db8\u0dad\u0dca \u0d89\u0dad\u0dcf \u0d9c\u0dd0\u0db9\u0dd4\u0dbb\u0dd4 \u0da2\u0dcf\u0dbd \u0dc3\u0db3\u0dc4\u0dcf \u0db6\u0ddc\u0dc4\u0ddd \u0daf\u0dbb\u0dca\u0dc1\u0d9a \u0dc0\u0dbd\u0da7 \u0dc4\u0dda\u0dad\u0dd4 \u0dc0\u0dd2\u0dba \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
"<p>Whether to periodically save models </p>\n": "<p>\u0dc0\u0dbb\u0dd2\u0db1\u0dca\u0dc0\u0dbb \u0d86\u0d9a\u0dd8\u0dad\u0dd2 \u0dc3\u0dd4\u0dbb\u0dd0\u0d9a\u0dd3\u0db8\u0da7 \u0dba\u0db1\u0dca\u0db1 </p>\n",
"NLP auto-regression trainer": "NLP \u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba-\u0d85\u0dc0\u0db4\u0dcf\u0dad\u0db1\u0dba \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4",
"This is a reusable trainer for auto-regressive tasks": "\u0db8\u0dd9\u0dba \u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba-\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0d9c\u0dcf\u0db8\u0dd3 \u0d9a\u0dcf\u0dbb\u0dca\u0dba\u0dba\u0db1\u0dca \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd0\u0dc0\u0dad \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbd \u0dc4\u0dd0\u0d9a\u0dd2 \u0dc0\u0db1 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4\u0dc0\u0dd9\u0d9a\u0dd4 \u0dc0\u0dda"
}
translate_cache/experiments/nlp_autoregression.zh.json
+69
View File
@@ -0,0 +1,69 @@
{
"<h1>Auto-regressive NLP model trainer</h1>\n": "<h1>\u81ea\u52a8\u56de\u5f52 NLP \u6a21\u578b\u8bad\u7ec3\u5668</h1>\n",
"<h3>Basic english tokenizer</h3>\n<p>We use character level tokenizer in this experiment. You can switch by setting,</p>\n<span translate=no>_^_0_^_</span><p>in the configurations dictionary when starting the experiment.</p>\n": "<h3>\u57fa\u7840\u82f1\u8bed\u5206\u8bcd\u5668</h3>\n<p>\u6211\u4eec\u5728\u8fd9\u4e2a\u5b9e\u9a8c\u4e2d\u4f7f\u7528\u89d2\u8272\u7b49\u7ea7\u5206\u8bcd\u5668\u3002\u4f60\u53ef\u4ee5\u901a\u8fc7\u8bbe\u7f6e\u8fdb\u884c\u5207\u6362\uff0c</p>\n<span translate=no>_^_0_^_</span><p>\u5f00\u59cb\u5b9e\u9a8c\u65f6\u5728\u914d\u7f6e\u5b57\u5178\u4e2d\u3002</p>\n",
"<h3>Character level tokenizer configuration</h3>\n": "<h3>\u89d2\u8272\u7ea7\u522b\u5206\u8bcd\u5668\u914d\u7f6e</h3>\n",
"<h3>Character level tokenizer</h3>\n": "<h3>\u89d2\u8272\u7b49\u7ea7\u5206\u8bcd\u5668</h3>\n",
"<h3>Cross entropy loss</h3>\n": "<h3>\u4ea4\u53c9\u71b5\u635f\u5931</h3>\n",
"<h3>Default <a href=\"../optimizers/configs.html\">optimizer configurations</a></h3>\n": "<h3>\u9ed8\u8ba4<a href=\"../optimizers/configs.html\">\u4f18\u5316\u5668\u914d\u7f6e</a></h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u521d\u59cb\u5316</h3>\n",
"<h3>Sampling function to generate samples periodically while training</h3>\n": "<h3>\u91c7\u6837\u529f\u80fd\u53ef\u5728\u8bad\u7ec3\u65f6\u5b9a\u671f\u751f\u6210\u6837\u672c</h3>\n",
"<h3>Sequential training data loader</h3>\n": "<h3>\u987a\u5e8f\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</h3>\n",
"<h3>Sequential validation data loader</h3>\n": "<h3>\u987a\u5e8f\u9a8c\u8bc1\u6570\u636e\u52a0\u8f7d\u5668</h3>\n",
"<h3>Shuffled training data loader</h3>\n": "<h3>\u6539\u7ec4\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</h3>\n",
"<h3>Shuffled validation data loader</h3>\n": "<h3>\u6539\u7ec4\u9a8c\u8bc1\u6570\u636e\u52a0\u8f7d\u5668</h3>\n",
"<h3>Tiny Shakespeare dataset</h3>\n<p>It will download from the url if not present</p>\n": "<h3>\u5c0f\u838e\u58eb\u6bd4\u4e9a\u6570\u636e\u96c6</h3>\n<p>\u5982\u679c\u4e0d\u5b58\u5728\uff0c\u5b83\u5c06\u4ece\u7f51\u5740\u4e0b\u8f7d</p>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u57f9\u8bad\u6216\u9a8c\u8bc1\u6b65\u9aa4</h3>\n",
"<h3>Transpose batch</h3>\n<p><span translate=no>_^_0_^_</span> collects the batches on the first dimension. We need to transpose it to be sequence first.</p>\n": "<h3>\u6279\u91cf\u79fb\u8c03</h3>\n<p><span translate=no>_^_0_^_</span>\u6536\u96c6\u7b2c\u4e00\u4e2a\u7ef4\u5ea6\u7684\u6279\u6b21\u3002\u6211\u4eec\u9700\u8981\u5148\u5c06\u5b83\u79fb\u8c03\u4e3a\u987a\u5e8f\u3002</p>\n",
"<p> <a id=\"NLPAutoRegressionConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n<p>This has the basic configurations for NLP auto-regressive task training. All the properties are configurable.</p>\n": "<p><a id=\"NLPAutoRegressionConfigs\"></a></p>\n<h2>\u8bad\u7ec3\u5668\u914d\u7f6e</h2>\n<p>\u5b83\u5177\u6709 NLP \u81ea\u52a8\u56de\u5f52\u4efb\u52a1\u8bad\u7ec3\u7684\u57fa\u672c\u914d\u7f6e\u3002\u6240\u6709\u5c5e\u6027\u90fd\u662f\u53ef\u914d\u7f6e\u7684\u3002</p>\n",
"<p> Get number of tokens</p>\n": "<p>\u83b7\u53d6\u4ee3\u5e01\u6570\u91cf</p>\n",
"<p>Accuracy function </p>\n": "<p>\u7cbe\u5ea6\u51fd\u6570</p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u5411\u65e5\u5fd7\u6a21\u5757\u8f93\u51fa\u6dfb\u52a0\u94a9\u5b50</p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u589e\u52a0\u4f5c\u4e3a\u72b6\u6001\u6a21\u5757\u7684\u7cbe\u5ea6\u3002\u8fd9\u4e2a\u540d\u5b57\u53ef\u80fd\u4ee4\u4eba\u56f0\u60d1\uff0c\u56e0\u4e3a\u5b83\u65e8\u5728\u5b58\u50a8 RNN \u7684\u8bad\u7ec3\u548c\u9a8c\u8bc1\u4e4b\u95f4\u7684\u72b6\u6001\u3002\u8fd9\u5c06\u4f7f\u7cbe\u5ea6\u6307\u6807\u7edf\u8ba1\u6570\u636e\u5206\u5f00\uff0c\u4ee5\u4fbf\u8fdb\u884c\u8bad\u7ec3\u548c\u9a8c\u8bc1\u3002</p>\n",
"<p>Add the prediction for logging </p>\n": "<p>\u6dfb\u52a0\u65e5\u5fd7\u8bb0\u5f55\u7684\u9884\u6d4b</p>\n",
"<p>Add the prediction to prompt </p>\n": "<p>\u5c06\u9884\u6d4b\u6dfb\u52a0\u5230\u63d0\u793a\u7b26\u4e2d</p>\n",
"<p>Autoregressive model </p>\n": "<p>\u81ea\u56de\u5f52\u6a21\u578b</p>\n",
"<p>Batch size </p>\n": "<p>\u6279\u91cf\u5927\u5c0f</p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u8ba1\u7b97\u548c\u8bb0\u5f55\u7cbe\u5ea6</p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u8ba1\u7b97\u5e76\u8bb0\u5f55\u635f\u5931</p>\n",
"<p>Calculate gradients </p>\n": "<p>\u8ba1\u7b97\u68af\u5ea6</p>\n",
"<p>Clear the gradients </p>\n": "<p>\u6e05\u9664\u6e10\u53d8</p>\n",
"<p>Clip gradients </p>\n": "<p>\u526a\u8f91\u6e10\u53d8</p>\n",
"<p>Collect output for printing </p>\n": "<p>\u6536\u96c6\u8f93\u51fa\u4ee5\u8fdb\u884c\u6253\u5370</p>\n",
"<p>Data loaders shuffle with replacement </p>\n": "<p>\u6570\u636e\u52a0\u8f7d\u5668\u968f\u7740\u66ff\u6362\u800c\u968f\u673a\u64ad\u653e</p>\n",
"<p>Get model outputs. It&#x27;s returning a tuple for states when using RNNs. This is not implemented yet. \ud83d\ude1c </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u8f93\u51fa\u3002\u5b83\u5728\u4f7f\u7528 RNN \u65f6\u8fd4\u56de\u72b6\u6001\u7684\u5143\u7ec4\u3002\u8fd9\u8fd8\u6ca1\u6709\u5b9e\u73b0\u3002\ud83d\ude1c</p>\n",
"<p>Get the model output </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u8f93\u51fa</p>\n",
"<p>Get the model prediction (greedy) </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u9884\u6d4b\uff08\u8d2a\u5a6a\uff09</p>\n",
"<p>Gradient clipping </p>\n": "<p>\u6e10\u53d8\u526a\u5207</p>\n",
"<p>Length of the sequence, or context size </p>\n": "<p>\u5e8f\u5217\u7684\u957f\u5ea6\u6216\u4e0a\u4e0b\u6587\u5927\u5c0f</p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u8bb0\u5f55\u6bcf\u4e2a\u7eaa\u5143\u6700\u540e\u4e00\u6279\u7684\u6a21\u578b\u53c2\u6570\u548c\u68af\u5ea6</p>\n",
"<p>Loss function </p>\n": "<p>\u4e8f\u635f\u51fd\u6570</p>\n",
"<p>Model embedding size </p>\n": "<p>\u6a21\u578b\u5d4c\u5165\u5927\u5c0f</p>\n",
"<p>Move data to the device </p>\n": "<p>\u5c06\u6570\u636e\u79fb\u52a8\u5230\u8bbe\u5907</p>\n",
"<p>Number of token in vocabulary </p>\n": "<p>\u8bcd\u6c47\u4e2d\u7684\u4ee3\u5e01\u6570\u91cf</p>\n",
"<p>Optimizer </p>\n": "<p>\u4f18\u5316\u5668</p>\n",
"<p>Override to calculate and log other metrics </p>\n": "<p>\u8986\u76d6\u4ee5\u8ba1\u7b97\u548c\u8bb0\u5f55\u5176\u4ed6\u6307\u6807</p>\n",
"<p>Print the sampled output </p>\n": "<p>\u6253\u5370\u91c7\u6837\u8f93\u51fa</p>\n",
"<p>Sample 25 tokens </p>\n": "<p>\u6837\u672c 25 \u4e2a\u4ee3\u5e01</p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u4fdd\u5b58\u8ddf\u8e2a\u7684\u6307\u6807</p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u8bbe\u7f6e\u8ddf\u8e2a\u5668\u914d\u7f6e</p>\n",
"<p>Set training/eval mode </p>\n": "<p>\u8bbe\u7f6e\u8bad\u7ec3/\u8bc4\u4f30\u6a21\u5f0f</p>\n",
"<p>Stack the batch along the second dimension <span translate=no>_^_0_^_</span> </p>\n": "<p>\u6cbf\u7b2c\u4e8c\u7ef4\u5ea6\u5806\u53e0\u6279\u6b21<span translate=no>_^_0_^_</span></p>\n",
"<p>Starting prompt </p>\n": "<p>\u542f\u52a8\u63d0\u793a</p>\n",
"<p>Take optimizer step </p>\n": "<p>\u91c7\u53d6\u4f18\u5316\u5668\u6b65\u9aa4</p>\n",
"<p>Text dataset </p>\n": "<p>\u6587\u672c\u6570\u636e\u96c6</p>\n",
"<p>Text prompt to start sampling (for illustration) </p>\n": "<p>\u5f00\u59cb\u91c7\u6837\u7684\u6587\u672c\u63d0\u793a\uff08\u7528\u4e8e\u8bf4\u660e\uff09</p>\n",
"<p>The token separator when sampling (blank for character level tokenization) </p>\n": "<p>\u91c7\u6837\u65f6\u7684\u4ee4\u724c\u5206\u9694\u7b26\uff08\u5bf9\u4e8e\u5b57\u7b26\u7ea7\u522b\u6807\u8bb0\u5316\u4e3a\u7a7a\u767d\uff09</p>\n",
"<p>Tokenize the prompt </p>\n": "<p>\u5c06\u63d0\u793a\u7b26\u53f7\u5316</p>\n",
"<p>Tokenizer </p>\n": "<p>\u5206\u8bcd\u5668</p>\n",
"<p>Train the model </p>\n": "<p>\u8bad\u7ec3\u6a21\u578b</p>\n",
"<p>Training data loader </p>\n": "<p>\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p>Training device </p>\n": "<p>\u8bad\u7ec3\u8bbe\u5907</p>\n",
"<p>Update global step (number of tokens processed) when in training mode </p>\n": "<p>\u5728\u8bad\u7ec3\u6a21\u5f0f\u4e0b\u66f4\u65b0\u5168\u5c40\u6b65\u957f\uff08\u5904\u7406\u7684\u4ee4\u724c\u6570\uff09</p>\n",
"<p>Validation data loader </p>\n": "<p>\u9a8c\u8bc1\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u662f\u5426\u6355\u83b7\u6a21\u578b\u8f93\u51fa</p>\n",
"<p>Whether to log model activations (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u662f\u5426\u8bb0\u5f55\u6a21\u578b\u6fc0\u6d3b\uff08\u6bcf\u4e2a\u7eaa\u5143\u4e00\u6b21\uff09\u3002\u8fd9\u4e9b\u662f\u6bcf\u5c42\u7684\u6c47\u603b\u7edf\u8ba1\u6570\u636e\uff0c\u4f46\u5b83\u4ecd\u7136\u53ef\u80fd\u5bfc\u81f4\u975e\u5e38\u6df1\u7684\u7f51\u7edc\u7684\u8bb8\u591a\u6307\u6807\u3002</p>\n",
"<p>Whether to log model parameters and gradients (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u662f\u5426\u8bb0\u5f55\u6a21\u578b\u53c2\u6570\u548c\u68af\u5ea6\uff08\u6bcf\u4e2a\u7eaa\u5143\u4e00\u6b21\uff09\u3002\u8fd9\u4e9b\u662f\u6bcf\u5c42\u7684\u6c47\u603b\u7edf\u8ba1\u6570\u636e\uff0c\u4f46\u5b83\u4ecd\u7136\u53ef\u80fd\u5bfc\u81f4\u975e\u5e38\u6df1\u7684\u7f51\u7edc\u7684\u8bb8\u591a\u6307\u6807\u3002</p>\n",
"<p>Whether to periodically save models </p>\n": "<p>\u662f\u5426\u5b9a\u671f\u4fdd\u5b58\u6a21\u578b</p>\n",
"NLP auto-regression trainer": "NLP \u81ea\u52a8\u56de\u5f52\u8bad\u7ec3\u5668",
"This is a reusable trainer for auto-regressive tasks": "\u8fd9\u662f\u4e00\u6b3e\u7528\u4e8e\u81ea\u52a8\u56de\u5f52\u4efb\u52a1\u7684\u53ef\u91cd\u590d\u4f7f\u7528\u7684\u8bad\u7ec3\u5668"
}
translate_cache/experiments/nlp_classification.ja.json
+71
View File
@@ -0,0 +1,71 @@
{
"<h1>NLP model trainer for classification</h1>\n": "<h1>\u5206\u985e\u7528 NLP \u30e2\u30c7\u30eb\u30c8\u30ec\u30fc\u30ca\u30fc</h1>\n",
"<h2>Function to load data into batches</h2>\n": "<h2>\u30c7\u30fc\u30bf\u3092\u30d0\u30c3\u30c1\u306b\u30ed\u30fc\u30c9\u3059\u308b\u6a5f\u80fd</h2>\n",
"<h3>AG News dataset</h3>\n<p>This loads the AG News dataset and the set the values for <span translate=no>_^_0_^_</span>, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, and <span translate=no>_^_3_^_</span>.</p>\n": "<h3>AG \u30cb\u30e5\u30fc\u30b9\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</h3>\n<p>\u3053\u308c\u306b\u3088\u308a AG News \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u304c\u8aad\u307f\u8fbc\u307e\u308c<span translate=no>_^_0_^_</span>\u3001\u3001<span translate=no>_^_1_^_</span><span translate=no>_^_2_^_</span>\u3001<span translate=no>_^_3_^_</span>\u306e\u5024\u304c\u8a2d\u5b9a\u3055\u308c\u307e\u3059\u3002</p>\n",
"<h3>Basic english tokenizer</h3>\n<p>We use character level tokenizer in this experiment. You can switch by setting,</p>\n<span translate=no>_^_0_^_</span><p>in the configurations dictionary when starting the experiment.</p>\n": "<h3>\u30d9\u30fc\u30b7\u30c3\u30af\u30fb\u30a4\u30f3\u30b0\u30ea\u30c3\u30b7\u30e5\u30fb\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc</h3>\n<p>\u3053\u306e\u5b9f\u9a13\u3067\u306f\u3001\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u306e\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc\u3092\u4f7f\u7528\u3057\u307e\u3059\u3002\u8a2d\u5b9a\u3067\u5207\u308a\u66ff\u3048\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u304c\u3001</p>\n<span translate=no>_^_0_^_</span><p>\u5b9f\u9a13\u3092\u958b\u59cb\u3059\u308b\u3068\u304d\u306b\u69cb\u6210\u8f9e\u66f8\u306b\u3042\u308a\u307e\u3059\u3002</p>\n",
"<h3>Character level tokenizer</h3>\n": "<h3>\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc</h3>\n",
"<h3>Default <a href=\"../optimizers/configs.html\">optimizer configurations</a></h3>\n": "<h3><a href=\"../optimizers/configs.html\">\u30c7\u30d5\u30a9\u30eb\u30c8\u306e\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc\u69cb\u6210</a></h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u521d\u671f\u5316</h3>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u307e\u305f\u306f\u691c\u8a3c\u30b9\u30c6\u30c3\u30d7</h3>\n",
"<p> </p>\n": "<p></p>\n",
"<p> <a id=\"NLPClassificationConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n<p>This has the basic configurations for NLP classification task training. All the properties are configurable.</p>\n": "<p><a id=\"NLPClassificationConfigs\"></a></p>\n<h2>\u30c8\u30ec\u30fc\u30ca\u30fc\u69cb\u6210</h2>\n<p>\u3053\u308c\u306b\u306f\u3001NLP\u5206\u985e\u30bf\u30b9\u30af\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306e\u57fa\u672c\u7684\u306a\u69cb\u6210\u304c\u3042\u308a\u307e\u3059\u3002\u3059\u3079\u3066\u306e\u30d7\u30ed\u30d1\u30c6\u30a3\u306f\u8a2d\u5b9a\u53ef\u80fd\u3067\u3059\u3002</p>\n",
"<p> Character level tokenizer configuration</p>\n": "<p>\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30ec\u30d9\u30eb\u306e\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc\u8a2d\u5b9a</p>\n",
"<p> Get number of tokens</p>\n": "<p>\u30c8\u30fc\u30af\u30f3\u306e\u6570\u3092\u53d6\u5f97</p>\n",
"<p>Accuracy function </p>\n": "<p>\u7cbe\u5ea6\u6a5f\u80fd</p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u30e2\u30b8\u30e5\u30fc\u30eb\u51fa\u529b\u3092\u30ed\u30b0\u306b\u8a18\u9332\u3059\u308b\u30d5\u30c3\u30af\u3092\u8ffd\u52a0</p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u30b9\u30c6\u30fc\u30c8\u30e2\u30b8\u30e5\u30fc\u30eb\u3068\u3057\u3066\u7cbe\u5ea6\u3092\u8ffd\u52a0\u3057\u3066\u304f\u3060\u3055\u3044\u3002\u3053\u306e\u540d\u524d\u306f\u3001RNN \u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3068\u691c\u8a3c\u306e\u9593\u306e\u72b6\u614b\u3092\u4fdd\u5b58\u3059\u308b\u305f\u3081\u306e\u3082\u306e\u306a\u306e\u3067\u3001\u304a\u305d\u3089\u304f\u308f\u304b\u308a\u306b\u304f\u3044\u3067\u3057\u3087\u3046\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u7cbe\u5ea6\u6307\u6a19\u306e\u7d71\u8a08\u60c5\u5831\u304c\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u7528\u3068\u691c\u8a3c\u7528\u306b\u5225\u3005\u306b\u4fdd\u6301\u3055\u308c\u307e\u3059\u3002</p>\n",
"<p>Autoregressive model </p>\n": "<p>\u81ea\u5df1\u56de\u5e30\u30e2\u30c7\u30eb</p>\n",
"<p>Batch size </p>\n": "<p>\u30d0\u30c3\u30c1\u30b5\u30a4\u30ba</p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u7cbe\u5ea6\u306e\u8a08\u7b97\u3068\u8a18\u9332</p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u640d\u5931\u306e\u8a08\u7b97\u3068\u8a18\u9332</p>\n",
"<p>Calculate gradients </p>\n": "<p>\u52fe\u914d\u306e\u8a08\u7b97</p>\n",
"<p>Clear the gradients </p>\n": "<p>\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3\u3092\u30af\u30ea\u30a2</p>\n",
"<p>Clip gradients </p>\n": "<p>\u30af\u30ea\u30c3\u30d7\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3</p>\n",
"<p>Collect tokens from training dataset </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u304b\u3089\u30c8\u30fc\u30af\u30f3\u3092\u53ce\u96c6</p>\n",
"<p>Collect tokens from validation dataset </p>\n": "<p>\u691c\u8a3c\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u304b\u3089\u30c8\u30fc\u30af\u30f3\u3092\u53ce\u96c6</p>\n",
"<p>Create <a href=\"../utils.html#map_style_dataset\">map-style datasets</a> </p>\n": "<p><a href=\"../utils.html#map_style_dataset\">\u30de\u30c3\u30d7\u30b9\u30bf\u30a4\u30eb\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u4f5c\u6210</a></p>\n",
"<p>Create a counter </p>\n": "<p>\u30ab\u30a6\u30f3\u30bf\u30fc\u306e\u4f5c\u6210</p>\n",
"<p>Create training data loader </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc\u306e\u4f5c\u6210</p>\n",
"<p>Create validation data loader </p>\n": "<p>\u691c\u8a3c\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc\u306e\u4f5c\u6210</p>\n",
"<p>Create vocabulary </p>\n": "<p>\u30dc\u30ad\u30e3\u30d6\u30e9\u30ea\u30fc\u306e\u4f5c\u6210</p>\n",
"<p>Empty labels tensor </p>\n": "<p>\u7a7a\u30e9\u30d9\u30eb\u30c6\u30f3\u30bd\u30eb</p>\n",
"<p>Get model outputs. It&#x27;s returning a tuple for states when using RNNs. This is not implemented yet. \ud83d\ude1c </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u53d6\u5f97\u3057\u307e\u3059\u3002RNN \u3092\u4f7f\u7528\u3059\u308b\u5834\u5408\u306f\u30b9\u30c6\u30fc\u30c8\u306e\u30bf\u30d7\u30eb\u3092\u8fd4\u3057\u307e\u3059\u3002\u3053\u308c\u306f\u307e\u3060\u5b9f\u88c5\u3055\u308c\u3066\u3044\u307e\u305b\u3093\u3002\ud83d\ude1c</p>\n",
"<p>Get tokenizer </p>\n": "<p>\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc\u3092\u5165\u624b</p>\n",
"<p>Get training and validation datasets </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3068\u691c\u8a3c\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u5165\u624b</p>\n",
"<p>Gradient clipping </p>\n": "<p>\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3\u30af\u30ea\u30c3\u30d4\u30f3\u30b0</p>\n",
"<p>Input data tensor, initialized with <span translate=no>_^_0_^_</span> </p>\n": "<p>\u3067\u521d\u671f\u5316\u3055\u308c\u305f\u5165\u529b\u30c7\u30fc\u30bf\u30c6\u30f3\u30bd\u30eb <span translate=no>_^_0_^_</span></p>\n",
"<p>Length of the sequence, or context size </p>\n": "<p>\u30b7\u30fc\u30b1\u30f3\u30b9\u306e\u9577\u3055\u3001\u307e\u305f\u306f\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u30b5\u30a4\u30ba</p>\n",
"<p>Load data to memory </p>\n": "<p>\u30c7\u30fc\u30bf\u3092\u30e1\u30e2\u30ea\u306b\u8aad\u307f\u8fbc\u3080</p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u5404\u30a8\u30dd\u30c3\u30af\u306e\u6700\u5f8c\u306e\u30d0\u30c3\u30c1\u3067\u30e2\u30c7\u30eb\u30d1\u30e9\u30e1\u30fc\u30bf\u3068\u52fe\u914d\u3092\u8a18\u9332\u3057\u307e\u3059</p>\n",
"<p>Loop through the samples </p>\n": "<p>\u30b5\u30f3\u30d7\u30eb\u3092\u30eb\u30fc\u30d7\u51e6\u7406</p>\n",
"<p>Loss function </p>\n": "<p>\u640d\u5931\u95a2\u6570</p>\n",
"<p>Model embedding size </p>\n": "<p>\u30e2\u30c7\u30eb\u57cb\u3081\u8fbc\u307f\u30b5\u30a4\u30ba</p>\n",
"<p>Move data to the device </p>\n": "<p>\u30c7\u30fc\u30bf\u3092\u30c7\u30d0\u30a4\u30b9\u306b\u79fb\u52d5</p>\n",
"<p>Number of classes </p>\n": "<p>\u30af\u30e9\u30b9\u6570</p>\n",
"<p>Number of token in vocabulary </p>\n": "<p>\u30dc\u30ad\u30e3\u30d6\u30e9\u30ea\u30fc\u5185\u306e\u30c8\u30fc\u30af\u30f3\u306e\u6570</p>\n",
"<p>Optimizer </p>\n": "<p>\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc</p>\n",
"<p>Return <span translate=no>_^_0_^_</span>, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, and <span translate=no>_^_3_^_</span> </p>\n": "<p>\u30ea\u30bf\u30fc\u30f3<span translate=no>_^_0_^_</span><span translate=no>_^_1_^_</span>\u3001<span translate=no>_^_2_^_</span>\u3001\u3001<span translate=no>_^_3_^_</span></p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u8ffd\u8de1\u3057\u305f\u30e1\u30c8\u30ea\u30af\u30b9\u3092\u4fdd\u5b58\u3059\u308b</p>\n",
"<p>Set the final token in the sequence to <span translate=no>_^_0_^_</span> </p>\n": "<p>\u30b7\u30fc\u30b1\u30f3\u30b9\u306e\u6700\u5f8c\u306e\u30c8\u30fc\u30af\u30f3\u3092\u6b21\u306e\u3088\u3046\u306b\u8a2d\u5b9a\u3057\u307e\u3059\u3002<span translate=no>_^_0_^_</span></p>\n",
"<p>Set the label </p>\n": "<p>\u30e9\u30d9\u30eb\u3092\u8a2d\u5b9a</p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u30c8\u30e9\u30c3\u30ab\u30fc\u69cb\u6210\u3092\u8a2d\u5b9a</p>\n",
"<p>Take optimizer step </p>\n": "<p>\u6700\u9069\u5316\u306e\u4e00\u6b69\u3092\u8e0f\u307f\u51fa\u3059</p>\n",
"<p>Tokenize the input text </p>\n": "<p>\u5165\u529b\u30c6\u30ad\u30b9\u30c8\u3092\u30c8\u30fc\u30af\u30f3\u5316</p>\n",
"<p>Tokenizer </p>\n": "<p>\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc</p>\n",
"<p>Train the model </p>\n": "<p>\u30e2\u30c7\u30eb\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0</p>\n",
"<p>Training data loader </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
"<p>Training device </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30d0\u30a4\u30b9</p>\n",
"<p>Transpose and add to data </p>\n": "<p>\u8ee2\u7f6e\u3057\u3066\u30c7\u30fc\u30bf\u306b\u8ffd\u52a0</p>\n",
"<p>Truncate upto <span translate=no>_^_0_^_</span> </p>\n": "<p>\u6700\u5927\u307e\u3067\u5207\u308a\u6368\u3066 <span translate=no>_^_0_^_</span></p>\n",
"<p>Update global step (number of tokens processed) when in training mode </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30e2\u30fc\u30c9\u6642\u306b\u30b0\u30ed\u30fc\u30d0\u30eb\u30b9\u30c6\u30c3\u30d7 (\u51e6\u7406\u3055\u308c\u305f\u30c8\u30fc\u30af\u30f3\u306e\u6570) \u3092\u66f4\u65b0</p>\n",
"<p>Validation data loader </p>\n": "<p>\u691c\u8a3c\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
"<p>Vocabulary </p>\n": "<p>\u8a9e\u5f59</p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u30e2\u30c7\u30eb\u51fa\u529b\u3092\u30ad\u30e3\u30d7\u30c1\u30e3\u3059\u308b\u304b\u3069\u3046\u304b</p>\n",
"<p>Whether to log model activations (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u30e2\u30c7\u30eb\u306e\u30a2\u30af\u30c6\u30a3\u30d9\u30fc\u30b7\u30e7\u30f3\u3092\u30ed\u30b0\u306b\u8a18\u9332\u3059\u308b\u304b\u3069\u3046\u304b (\u30a8\u30dd\u30c3\u30af\u3054\u3068\u306b 1 \u56de)\u3002\u3053\u308c\u3089\u306f\u30ec\u30a4\u30e4\u30fc\u3054\u3068\u306e\u7d71\u8a08\u60c5\u5831\u3092\u307e\u3068\u3081\u305f\u3082\u306e\u3067\u3059\u304c\u3001\u305d\u308c\u3067\u3082\u975e\u5e38\u306b\u6df1\u3044\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u591a\u304f\u306e\u6307\u6a19\u306b\u3064\u306a\u304c\u308b\u53ef\u80fd\u6027\u304c\u3042\u308a\u307e\u3059</p>\u3002\n",
"<p>Whether to log model parameters and gradients (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u30e2\u30c7\u30eb\u30d1\u30e9\u30e1\u30fc\u30bf\u30fc\u3068\u52fe\u914d\u3092\u8a18\u9332\u3059\u308b\u304b\u3069\u3046\u304b (\u30a8\u30dd\u30c3\u30af\u3054\u3068\u306b 1 \u56de)\u3002\u3053\u308c\u3089\u306f\u30ec\u30a4\u30e4\u30fc\u3054\u3068\u306e\u7d71\u8a08\u60c5\u5831\u3092\u307e\u3068\u3081\u305f\u3082\u306e\u3067\u3059\u304c\u3001\u305d\u308c\u3067\u3082\u975e\u5e38\u306b\u6df1\u3044\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u591a\u304f\u306e\u6307\u6a19\u306b\u3064\u306a\u304c\u308b\u53ef\u80fd\u6027\u304c\u3042\u308a\u307e\u3059</p>\u3002\n",
"<p>Whether to periodically save models </p>\n": "<p>\u30e2\u30c7\u30eb\u3092\u5b9a\u671f\u7684\u306b\u4fdd\u5b58\u3059\u308b\u304b\u3069\u3046\u304b</p>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the batch of data collected by the <span translate=no>_^_1_^_</span></li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u304c\u53ce\u96c6\u3057\u305f\u30c7\u30fc\u30bf\u306e\u30d0\u30c3\u30c1\u3067\u3059 <span translate=no>_^_1_^_</span></li></ul>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the tokenizer function </li>\n<li><span translate=no>_^_1_^_</span> is the vocabulary </li>\n<li><span translate=no>_^_2_^_</span> is the length of the sequence </li>\n<li><span translate=no>_^_3_^_</span> is the token used for padding when the <span translate=no>_^_4_^_</span> is larger than the text length </li>\n<li><span translate=no>_^_5_^_</span> is the <span translate=no>_^_6_^_</span> token which we set at end of the input</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u30c8\u30fc\u30af\u30ca\u30a4\u30b6\u30fc\u95a2\u6570\u3067\u3059</li>\n<li><span translate=no>_^_1_^_</span>\u306f\u30dc\u30ad\u30e3\u30d6\u30e9\u30ea\u30fc</li>\n<li><span translate=no>_^_2_^_</span>\u30b7\u30fc\u30b1\u30f3\u30b9\u306e\u9577\u3055\u3067\u3059</li>\n<li><span translate=no>_^_3_^_</span><span translate=no>_^_4_^_</span>\u304c\u30c6\u30ad\u30b9\u30c8\u306e\u9577\u3055\u3088\u308a\u5927\u304d\u3044\u5834\u5408\u306b\u30d1\u30c7\u30a3\u30f3\u30b0\u306b\u4f7f\u7528\u3055\u308c\u308b\u30c8\u30fc\u30af\u30f3\u3067\u3059</li>\n<li><span translate=no>_^_5_^_</span><span translate=no>_^_6_^_</span>\u5165\u529b\u306e\u6700\u5f8c\u306b\u8a2d\u5b9a\u3057\u305f\u30c8\u30fc\u30af\u30f3\u3067\u3059</li></ul>\n",
"NLP classification trainer": "NLP \u5206\u985e\u30c8\u30ec\u30fc\u30ca\u30fc",
"This is a reusable trainer for classification tasks": "\u3053\u308c\u306f\u5206\u985e\u4f5c\u696d\u7528\u306e\u518d\u5229\u7528\u53ef\u80fd\u306a\u30c8\u30ec\u30fc\u30ca\u30fc\u3067\u3059"
}
translate_cache/experiments/nlp_classification.si.json
+71
View File
@@ -0,0 +1,71 @@
{
"<h1>NLP model trainer for classification</h1>\n": "<h1>\u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab\u0dba\u0dc3\u0db3\u0dc4\u0dcf NLP \u0d86\u0d9a\u0dd8\u0dad\u0dd2 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4</h1>\n",
"<h2>Function to load data into batches</h2>\n": "<h2>\u0daf\u0dad\u0dca\u0dad\u0d9a\u0dcf\u0dab\u0dca\u0da9\u0dc0\u0dbd\u0da7 \u0db4\u0dd0\u0da7\u0dc0\u0dd3\u0db8\u0dda \u0d9a\u0dcf\u0dbb\u0dca\u0dba\u0dba</h2>\n",
"<h3>AG News dataset</h3>\n<p>This loads the AG News dataset and the set the values for <span translate=no>_^_0_^_</span>, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, and <span translate=no>_^_3_^_</span>.</p>\n": "<h3>AG\u0db4\u0dd4\u0dc0\u0dad\u0dca \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h3>\n<p>\u0db8\u0dd9\u0dbaAG \u0db4\u0dd4\u0dc0\u0dad\u0dca \u0daf\u0dad\u0dca\u0dad \u0dc3\u0db8\u0dd4\u0daf\u0dcf\u0dba \u0db4\u0da7\u0dc0\u0db1 \u0d85\u0dad\u0dbb <span translate=no>_^_0_^_</span>, \u0dc3\u0db3\u0dc4\u0dcf \u0d85\u0d9c\u0dba\u0db1\u0dca \u0dc3\u0d9a\u0dc3\u0dcf \u0d87\u0dad, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, \u0dc3\u0dc4 <span translate=no>_^_3_^_</span>. </p>\n",
"<h3>Basic english tokenizer</h3>\n<p>We use character level tokenizer in this experiment. You can switch by setting,</p>\n<span translate=no>_^_0_^_</span><p>in the configurations dictionary when starting the experiment.</p>\n": "<h3>\u0db8\u0dd6\u0dbd\u0dd2\u0d9a\u0d89\u0d82\u0d9c\u0dca\u0dbb\u0dd3\u0dc3\u0dd2 \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca</h3>\n<p>\u0db8\u0dd9\u0db8\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8\u0dda\u0daf\u0dd3 \u0d85\u0db4\u0dd2 \u0da0\u0dbb\u0dd2\u0dad \u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db8\u0dd4. \u0dc3\u0dd0\u0d9a\u0dc3\u0dd3\u0db8\u0dd9\u0db1\u0dca \u0d94\u0db6\u0da7 \u0db8\u0dcf\u0dbb\u0dd4 \u0dc0\u0dd2\u0dba \u0dc4\u0dd0\u0d9a\u0dd2\u0dba,</p>\n<span translate=no>_^_0_^_</span><p>\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf\u0db6\u0dd0\u0dbd\u0dd3\u0db8 \u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda\u0daf\u0dd3 \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0dc1\u0db6\u0dca\u0daf\u0d9a\u0ddd\u0dc2\u0dba\u0dda. </p>\n",
"<h3>Character level tokenizer</h3>\n": "<h3>\u0d85\u0d9a\u0dca\u0dc2\u0dbb\u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca</h3>\n",
"<h3>Default <a href=\"../optimizers/configs.html\">optimizer configurations</a></h3>\n": "<h3>\u0db4\u0dd9\u0dbb\u0db1\u0dd2\u0db8\u0dd2 <a href=\"../optimizers/configs.html\">\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca</a></h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u0d86\u0dbb\u0db8\u0dca\u0db7\u0d9a\u0d9a\u0dbb\u0dab\u0dba</h3>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0\u0dc4\u0ddd \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db4\u0dd2\u0dba\u0dc0\u0dbb</h3>\n",
"<p> </p>\n": "<p> </p>\n",
"<p> <a id=\"NLPClassificationConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n<p>This has the basic configurations for NLP classification task training. All the properties are configurable.</p>\n": "<p> <a id=\"NLPClassificationConfigs\"></a></p>\n<h2>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4\u0db8\u0dcf\u0db1\u0d9a\u0dbb\u0dab</h2>\n<p>\u0d91\u0db1\u0dca\u0d91\u0dbd\u0dca\u0db4\u0dd3\u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab \u0d9a\u0dcf\u0dbb\u0dca\u0dba \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0db3\u0dc4\u0dcf \u0db8\u0dd6\u0dbd\u0dd2\u0d9a \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca \u0db8\u0dd9\u0dba\u0da7 \u0d87\u0dad. \u0dc3\u0dd2\u0dba\u0dbd\u0dd4\u0db8 \u0d9c\u0dd4\u0dab\u0dcf\u0d82\u0d9c \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0d9c\u0dad \u0d9a\u0dc5 \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
"<p> Character level tokenizer configuration</p>\n": "<p> \u0d85\u0d9a\u0dca\u0dc2\u0dbb\u0db8\u0da7\u0dca\u0da7\u0db8\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca \u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba</p>\n",
"<p> Get number of tokens</p>\n": "<p> \u0da7\u0ddd\u0d9a\u0db1\u0d9c\u0dab\u0db1 \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1</p>\n",
"<p>Accuracy function </p>\n": "<p>\u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u0db8\u0ddc\u0da9\u0dd2\u0dba\u0dd4\u0dbd\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0d9a\u0ddc\u0d9a\u0dca\u0d9a\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u0dbb\u0dcf\u0da2\u0dca\u0dba\u0db8\u0ddc\u0da9\u0dd2\u0dba\u0dd4\u0dbd\u0dba\u0d9a\u0dca \u0dbd\u0dd9\u0dc3 \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc0 \u0d91\u0d9a\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1. RNs \u0dc3\u0db3\u0dc4\u0dcf \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0dad\u0dbb \u0dbb\u0dcf\u0da2\u0dca\u0dba\u0dba\u0db1\u0dca \u0d9c\u0db6\u0da9\u0dcf \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0d85\u0daf\u0dc4\u0dc3\u0dca \u0d9a\u0dbb\u0db1 \u0db6\u0dd0\u0dc0\u0dd2\u0db1\u0dca \u0db1\u0db8 \u0db6\u0ddc\u0dc4\u0ddd \u0dc0\u0dd2\u0da7 \u0dc0\u0dca\u0dba\u0dcf\u0d9a\u0dd6\u0dbd \u0dc0\u0dda. \u0db8\u0dd9\u0dba \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0 \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf \u0db8\u0dd9\u0da7\u0dca\u0dbb\u0dd2\u0d9a\u0dca \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dd9\u0db1\u0db8 \u0dad\u0db6\u0dcf \u0d9c\u0db1\u0dd3. </p>\n",
"<p>Autoregressive model </p>\n": "<p>\u0dc3\u0dca\u0dc0\u0dba\u0d82\u0d9a\u0dca\u0dbb\u0dd3\u0dba\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0d9c\u0dcf\u0db8\u0dd3 \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba </p>\n",
"<p>Batch size </p>\n": "<p>\u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dca\u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u0d9c\u0dab\u0db1\u0dba\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db1\u0dd2\u0dbb\u0dc0\u0daf\u0dca\u0dba\u0dad\u0dcf\u0dc0\u0dba </p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u0d9c\u0dab\u0db1\u0dba\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dbd\u0ddc\u0d9c\u0dca \u0dc0\u0dd3\u0db8 </p>\n",
"<p>Calculate gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d9c\u0dab\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Clear the gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d89\u0dc0\u0dad\u0dca </p>\n",
"<p>Clip gradients </p>\n": "<p>\u0d9a\u0dca\u0dbd\u0dd2\u0db4\u0dca\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a </p>\n",
"<p>Collect tokens from training dataset </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba\u0dd9\u0db1\u0dca \u0da7\u0ddd\u0d9a\u0db1 \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Collect tokens from validation dataset </p>\n": "<p>\u0dc0\u0dbd\u0d82\u0d9c\u0dd4\u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba\u0dd9\u0db1\u0dca \u0da7\u0ddd\u0d9a\u0db1 \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Create <a href=\"../utils.html#map_style_dataset\">map-style datasets</a> </p>\n": "<p><a href=\"../utils.html#map_style_dataset\">\u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0dca \u0d86\u0d9a\u0dcf\u0dbb\u0dba\u0dda \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd</a> \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Create a counter </p>\n": "<p>\u0d9a\u0dc0\u0dd4\u0db1\u0dca\u0da7\u0dbb\u0dba\u0d9a\u0dca\u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Create training data loader </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Create validation data loader </p>\n": "<p>\u0dc0\u0dbd\u0d82\u0d9c\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Create vocabulary </p>\n": "<p>\u0dc0\u0da0\u0db1\u0db8\u0dcf\u0dbd\u0dcf\u0dc0 \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
"<p>Empty labels tensor </p>\n": "<p>\u0dc4\u0dd2\u0dc3\u0dca\u0dbd\u0dda\u0db6\u0dbd\u0dca \u0da7\u0dd9\u0db1\u0dca\u0dc3\u0dbb\u0dca </p>\n",
"<p>Get model outputs. It&#x27;s returning a tuple for states when using RNNs. This is not implemented yet. \ud83d\ude1c </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1. \u0d86\u0dbb\u0dca\u0d91\u0db1\u0dca\u0d91\u0dc3\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db1 \u0dc0\u0dd2\u0da7 \u0d91\u0dba \u0db4\u0dca\u0dbb\u0dcf\u0db1\u0dca\u0dad \u0dc3\u0db3\u0dc4\u0dcf \u0da7\u0dd6\u0dbd\u0dca \u0d91\u0d9a\u0d9a\u0dca \u0db1\u0dd0\u0dc0\u0dad \u0dbd\u0db6\u0dcf \u0daf\u0dd9\u0dba\u0dd2. \u0db8\u0dd9\u0dba \u0dad\u0dc0\u0db8 \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dbb \u0db1\u0dd0\u0dad \ud83d\ude1c </p>\n",
"<p>Get tokenizer </p>\n": "<p>\u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca\u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Get training and validation datasets </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0\u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Gradient clipping </p>\n": "<p>\u0dc1\u0dca\u0dbb\u0dda\u0dab\u0dd2\u0dba\u0dda\u0d9a\u0dca\u0dbd\u0dd2\u0db4\u0dd2\u0db1\u0dca </p>\n",
"<p>Input data tensor, initialized with <span translate=no>_^_0_^_</span> </p>\n": "<p>\u0d86\u0daf\u0dcf\u0db1\u0daf\u0dad\u0dca\u0dad \u0da7\u0dd9\u0db1\u0dca\u0dc3\u0dbb\u0dba, \u0dc3\u0db8\u0d9f \u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dbb \u0d87\u0dad <span translate=no>_^_0_^_</span> </p>\n",
"<p>Length of the sequence, or context size </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dba\u0dda\u0daf\u0dd2\u0d9c, \u0dc4\u0ddd \u0dc3\u0db1\u0dca\u0daf\u0dbb\u0dca\u0db7\u0dba \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
"<p>Load data to memory </p>\n": "<p>\u0db8\u0dad\u0d9a\u0dba\u0dc0\u0dd9\u0dad \u0daf\u0dad\u0dca\u0dad \u0db4\u0dd6\u0dbb\u0dab\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u0dc3\u0dd1\u0db8\u0dba\u0dd4\u0d9c\u0dbd\u0dba\u0d9a\u0db8 \u0d85\u0dc0\u0dc3\u0dcf\u0db1 \u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dda \u0d86\u0daf\u0dbb\u0dca\u0dc1 \u0db4\u0dbb\u0dcf\u0db8\u0dd2\u0dad\u0dd3\u0db1\u0dca \u0dc3\u0dc4 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Loop through the samples </p>\n": "<p>\u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd\u0dc4\u0dbb\u0dc4\u0dcf \u0dbd\u0dd6\u0db4 </p>\n",
"<p>Loss function </p>\n": "<p>\u0db4\u0dcf\u0da9\u0dd4\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
"<p>Model embedding size </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0d9a\u0dcf\u0dc0\u0dd0\u0daf\u0dca\u0daf\u0dd3\u0db8\u0dda \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
"<p>Move data to the device </p>\n": "<p>\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba\u0dc0\u0dd9\u0dad \u0daf\u0dad\u0dca\u0dad \u0d9c\u0dd9\u0db1\u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Number of classes </p>\n": "<p>\u0db4\u0db1\u0dca\u0dad\u0dd2\u0d9c\u0dab\u0db1 </p>\n",
"<p>Number of token in vocabulary </p>\n": "<p>\u0dc0\u0da0\u0db1\u0db8\u0dcf\u0dbd\u0dcf\u0dc0\u0dda \u0da7\u0ddd\u0d9a\u0db1\u0dca \u0d9c\u0dab\u0db1 </p>\n",
"<p>Optimizer </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0d9a\u0dbb\u0dab\u0dba </p>\n",
"<p>Return <span translate=no>_^_0_^_</span>, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, and <span translate=no>_^_3_^_</span> </p>\n": "<p>\u0d86\u0db4\u0dc3\u0dd4 <span translate=no>_^_0_^_</span><span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, \u0dc3\u0dc4 <span translate=no>_^_3_^_</span> </p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u0dbd\u0dd4\u0dc4\u0dd4\u0db6\u0dd0\u0db3\u0d87\u0dad\u0dd2 \u0db4\u0dca\u0dbb\u0db8\u0dd2\u0dad\u0dd2\u0d9a \u0dc3\u0dd4\u0dbb\u0d9a\u0dd2\u0db1\u0dca\u0db1 </p>\n",
"<p>Set the final token in the sequence to <span translate=no>_^_0_^_</span> </p>\n": "<p>\u0d85\u0db1\u0dd4\u0db4\u0dd2\u0dc5\u0dd2\u0dc0\u0dd9\u0dbd\u0dd9\u0dc4\u0dd2\u0d85\u0dc0\u0dc3\u0dcf\u0db1 \u0da7\u0ddd\u0d9a\u0db1\u0dba \u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 <span translate=no>_^_0_^_</span> </p>\n",
"<p>Set the label </p>\n": "<p>\u0dbd\u0dda\u0db6\u0dbd\u0dba\u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 </p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u0da7\u0dca\u0dbb\u0dd0\u0d9a\u0dbb\u0dca\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca \u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 </p>\n",
"<p>Take optimizer step </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab\u0db4\u0dd2\u0dba\u0dc0\u0dbb \u0d9c\u0db1\u0dca\u0db1 </p>\n",
"<p>Tokenize the input text </p>\n": "<p>\u0d86\u0daf\u0dcf\u0db1\u0db4\u0dd9\u0dc5 \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dca \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Tokenizer </p>\n": "<p>\u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca </p>\n",
"<p>Train the model </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Training data loader </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 </p>\n",
"<p>Training device </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba </p>\n",
"<p>Transpose and add to data </p>\n": "<p>\u0daf\u0dad\u0dca\u0dad\u0dc3\u0db8\u0dca\u0db4\u0dca\u0dbb\u0dda\u0dc2\u0dab\u0dba \u0d9a\u0dbb \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
"<p>Truncate upto <span translate=no>_^_0_^_</span> </p>\n": "<p>\u0daf\u0d9a\u0dca\u0dc0\u0dcf\u0da7\u0db1\u0dca\u0d9a \u0d9a\u0dbb\u0db1\u0dca\u0db1 <span translate=no>_^_0_^_</span> </p>\n",
"<p>Update global step (number of tokens processed) when in training mode </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0db4\u0dca\u0dbb\u0d9a\u0dcf\u0dbb\u0dba\u0dda\u0daf\u0dd3 \u0d9c\u0ddd\u0dbd\u0dd3\u0dba \u0db4\u0dd2\u0dba\u0dc0\u0dbb \u0dba\u0dcf\u0dc0\u0dad\u0dca\u0d9a\u0dcf\u0dbd\u0dd3\u0db1 \u0d9a\u0dbb\u0db1\u0dca\u0db1 (\u0dc3\u0dd0\u0d9a\u0dc3\u0dd6 \u0da7\u0ddd\u0d9a\u0db1 \u0d9c\u0dab\u0db1) </p>\n",
"<p>Validation data loader </p>\n": "<p>\u0dc0\u0dbd\u0d82\u0d9c\u0dd4\u0daf\u0dad\u0dca\u0dad \u0db4\u0dd0\u0da7\u0dc0\u0dd4\u0db8 </p>\n",
"<p>Vocabulary </p>\n": "<p>\u0dc0\u0da0\u0db1\u0db8\u0dcf\u0dbd\u0dcf\u0dc0 </p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0d9c\u0dca\u0dbb\u0dc4\u0dab\u0dba \u0d9a\u0dbb \u0d9c\u0dad \u0dba\u0dd4\u0dad\u0dd4\u0daf \u0dba\u0db1\u0dca\u0db1 </p>\n",
"<p>Whether to log model activations (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0dc3\u0d9a\u0dca\u0dbb\u0dd2\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dca \u0dbd\u0ddc\u0d9c\u0dca \u0d9a\u0dc5 \u0dba\u0dd4\u0dad\u0dd4\u0daf \u0dba\u0db1\u0dca\u0db1 (\u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d8a\u0db4\u0ddd\u0da0\u0dca \u0d91\u0d9a\u0d9a\u0da7 \u0dc0\u0dbb\u0d9a\u0dca). \u0db8\u0dda\u0dc0\u0dcf \u0dc3\u0dca\u0dae\u0dbb\u0dba\u0d9a\u0da7 \u0dc3\u0dcf\u0dbb\u0dcf\u0d82\u0dc1\u0d9c\u0dad \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dda, \u0db1\u0db8\u0dd4\u0dad\u0dca \u0d91\u0dba \u0dad\u0dc0\u0db8\u0dad\u0dca \u0d89\u0dad\u0dcf \u0d9c\u0dd0\u0db9\u0dd4\u0dbb\u0dd4 \u0da2\u0dcf\u0dbd \u0dc3\u0db3\u0dc4\u0dcf \u0db6\u0ddc\u0dc4\u0ddd \u0daf\u0dbb\u0dca\u0dc1\u0d9a \u0dc0\u0dbd\u0da7 \u0dc4\u0dda\u0dad\u0dd4 \u0dc0\u0dd2\u0dba \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
"<p>Whether to log model parameters and gradients (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u0d86\u0daf\u0dbb\u0dca\u0dc1\u0db4\u0dbb\u0dcf\u0db8\u0dd2\u0dad\u0dd3\u0db1\u0dca \u0dc3\u0dc4 \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a \u0dbd\u0d9d\u0dd4-\u0dc3\u0da7\u0dc4\u0db1 \u0dba\u0db1\u0dca\u0db1 (\u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0dc0\u0dbb\u0d9a\u0dca). \u0db8\u0dda\u0dc0\u0dcf \u0dc3\u0dca\u0dae\u0dbb\u0dba\u0d9a\u0da7 \u0dc3\u0dcf\u0dbb\u0dcf\u0d82\u0dc1\u0d9c\u0dad \u0dc3\u0d82\u0d9b\u0dca\u0dba\u0dcf\u0db1 \u0dc0\u0dda, \u0db1\u0db8\u0dd4\u0dad\u0dca \u0d91\u0dba \u0dad\u0dc0\u0db8\u0dad\u0dca \u0d89\u0dad\u0dcf \u0d9c\u0dd0\u0db9\u0dd4\u0dbb\u0dd4 \u0da2\u0dcf\u0dbd \u0dc3\u0db3\u0dc4\u0dcf \u0db6\u0ddc\u0dc4\u0ddd \u0daf\u0dbb\u0dca\u0dc1\u0d9a \u0dc0\u0dbd\u0da7 \u0dc4\u0dda\u0dad\u0dd4 \u0dc0\u0dd2\u0dba \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
"<p>Whether to periodically save models </p>\n": "<p>\u0dc0\u0dbb\u0dd2\u0db1\u0dca\u0dc0\u0dbb \u0d86\u0d9a\u0dd8\u0dad\u0dd2 \u0dc3\u0dd4\u0dbb\u0dd0\u0d9a\u0dd3\u0db8\u0da7 \u0dba\u0db1\u0dca\u0db1 </p>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the batch of data collected by the <span translate=no>_^_1_^_</span></li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0dc0\u0dd2\u0dc3\u0dd2\u0db1\u0dca \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0dcf\u0dab\u0dca\u0da9\u0dba\u0dba\u0dd2 <span translate=no>_^_1_^_</span></li></ul>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the tokenizer function </li>\n<li><span translate=no>_^_1_^_</span> is the vocabulary </li>\n<li><span translate=no>_^_2_^_</span> is the length of the sequence </li>\n<li><span translate=no>_^_3_^_</span> is the token used for padding when the <span translate=no>_^_4_^_</span> is larger than the text length </li>\n<li><span translate=no>_^_5_^_</span> is the <span translate=no>_^_6_^_</span> token which we set at end of the input</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dd2\u0dc3\u0dbb\u0dca \u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba \u0dc0\u0dda </li>\n<li><span translate=no>_^_1_^_</span> \u0dba\u0db1\u0dd4 \u0dc0\u0da0\u0db1 \u0db8\u0dcf\u0dbd\u0dcf\u0dc0\u0dba\u0dd2 </li>\n<li><span translate=no>_^_2_^_</span> \u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dba\u0dda \u0daf\u0dd2\u0d9c \u0dc0\u0dda </li>\n<li><span translate=no>_^_3_^_</span> \u0dba\u0db1\u0dd4 \u0db4\u0dd9\u0dc5 \u0daf\u0dd2\u0d9c\u0da7 \u0dc0\u0da9\u0dcf \u0dc0\u0dd2\u0dc1\u0dcf\u0dbd <span translate=no>_^_4_^_</span> \u0dc0\u0db1 \u0dc0\u0dd2\u0da7 \u0db4\u0dd1\u0da9\u0dd2\u0db1\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db1 \u0da7\u0ddd\u0d9a\u0db1\u0dba \u0dc0\u0dda </li>\n<li><span translate=no>_^_5_^_</span> \u0dba\u0db1\u0dd4 \u0d86\u0daf\u0dcf\u0db1\u0dba \u0d85\u0dc0\u0dc3\u0dcf\u0db1\u0dba\u0dda \u0d85\u0db4 \u0dc0\u0dd2\u0dc3\u0dd2\u0db1\u0dca \u0dc3\u0d9a\u0dc3\u0dca \u0d9a\u0dbb\u0db1 \u0dbd\u0daf <span translate=no>_^_6_^_</span> \u0da7\u0ddd\u0d9a\u0db1\u0dba\u0dba\u0dd2</li></ul>\n",
"NLP classification trainer": "NLP \u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4",
"This is a reusable trainer for classification tasks": "\u0db8\u0dd9\u0dba \u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab \u0d9a\u0dcf\u0dbb\u0dca\u0dba\u0dba\u0db1\u0dca \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd0\u0dc0\u0dad \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dc5 \u0dc4\u0dd0\u0d9a\u0dd2 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d9a\u0dbb\u0dd4\u0dc0\u0d9a\u0dd2"
}
translate_cache/experiments/nlp_classification.zh.json
+71
View File
@@ -0,0 +1,71 @@
{
"<h1>NLP model trainer for classification</h1>\n": "<h1>\u7528\u4e8e\u5206\u7c7b\u7684 NLP \u6a21\u578b\u8bad\u7ec3\u5668</h1>\n",
"<h2>Function to load data into batches</h2>\n": "<h2>\u5c06\u6570\u636e\u52a0\u8f7d\u5230\u6279\u5904\u7406\u4e2d\u7684\u51fd\u6570</h2>\n",
"<h3>AG News dataset</h3>\n<p>This loads the AG News dataset and the set the values for <span translate=no>_^_0_^_</span>, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, and <span translate=no>_^_3_^_</span>.</p>\n": "<h3>AG \u65b0\u95fb\u6570\u636e\u96c6</h3>\n<p>\u8fd9\u5c06\u52a0\u8f7d AG News \u6570\u636e\u96c6\u5e76\u8bbe\u7f6e<span translate=no>_^_0_^_</span>\u3001<span translate=no>_^_1_^_</span><span translate=no>_^_2_^_</span>\u3001\u548c\u7684\u503c<span translate=no>_^_3_^_</span>\u3002</p>\n",
"<h3>Basic english tokenizer</h3>\n<p>We use character level tokenizer in this experiment. You can switch by setting,</p>\n<span translate=no>_^_0_^_</span><p>in the configurations dictionary when starting the experiment.</p>\n": "<h3>\u57fa\u7840\u82f1\u8bed\u5206\u8bcd\u5668</h3>\n<p>\u6211\u4eec\u5728\u8fd9\u4e2a\u5b9e\u9a8c\u4e2d\u4f7f\u7528\u89d2\u8272\u7b49\u7ea7\u5206\u8bcd\u5668\u3002\u4f60\u53ef\u4ee5\u901a\u8fc7\u8bbe\u7f6e\u8fdb\u884c\u5207\u6362\uff0c</p>\n<span translate=no>_^_0_^_</span><p>\u5f00\u59cb\u5b9e\u9a8c\u65f6\u5728\u914d\u7f6e\u5b57\u5178\u4e2d\u3002</p>\n",
"<h3>Character level tokenizer</h3>\n": "<h3>\u89d2\u8272\u7b49\u7ea7\u5206\u8bcd\u5668</h3>\n",
"<h3>Default <a href=\"../optimizers/configs.html\">optimizer configurations</a></h3>\n": "<h3>\u9ed8\u8ba4<a href=\"../optimizers/configs.html\">\u4f18\u5316\u5668\u914d\u7f6e</a></h3>\n",
"<h3>Initialization</h3>\n": "<h3>\u521d\u59cb\u5316</h3>\n",
"<h3>Training or validation step</h3>\n": "<h3>\u57f9\u8bad\u6216\u9a8c\u8bc1\u6b65\u9aa4</h3>\n",
"<p> </p>\n": "<p></p>\n",
"<p> <a id=\"NLPClassificationConfigs\"></a></p>\n<h2>Trainer configurations</h2>\n<p>This has the basic configurations for NLP classification task training. All the properties are configurable.</p>\n": "<p><a id=\"NLPClassificationConfigs\"></a></p>\n<h2>\u8bad\u7ec3\u5668\u914d\u7f6e</h2>\n<p>\u5b83\u5177\u6709 NLP \u5206\u7c7b\u4efb\u52a1\u57f9\u8bad\u7684\u57fa\u672c\u914d\u7f6e\u3002\u6240\u6709\u5c5e\u6027\u90fd\u662f\u53ef\u914d\u7f6e\u7684\u3002</p>\n",
"<p> Character level tokenizer configuration</p>\n": "<p>\u89d2\u8272\u7ea7\u522b\u5206\u8bcd\u5668\u914d\u7f6e</p>\n",
"<p> Get number of tokens</p>\n": "<p>\u83b7\u53d6\u4ee3\u5e01\u6570\u91cf</p>\n",
"<p>Accuracy function </p>\n": "<p>\u7cbe\u5ea6\u51fd\u6570</p>\n",
"<p>Add a hook to log module outputs </p>\n": "<p>\u5411\u65e5\u5fd7\u6a21\u5757\u8f93\u51fa\u6dfb\u52a0\u94a9\u5b50</p>\n",
"<p>Add accuracy as a state module. The name is probably confusing, since it&#x27;s meant to store states between training and validation for RNNs. This will keep the accuracy metric stats separate for training and validation. </p>\n": "<p>\u589e\u52a0\u4f5c\u4e3a\u72b6\u6001\u6a21\u5757\u7684\u7cbe\u5ea6\u3002\u8fd9\u4e2a\u540d\u5b57\u53ef\u80fd\u4ee4\u4eba\u56f0\u60d1\uff0c\u56e0\u4e3a\u5b83\u65e8\u5728\u5b58\u50a8 RNN \u7684\u8bad\u7ec3\u548c\u9a8c\u8bc1\u4e4b\u95f4\u7684\u72b6\u6001\u3002\u8fd9\u5c06\u4f7f\u7cbe\u5ea6\u6307\u6807\u7edf\u8ba1\u6570\u636e\u5206\u5f00\uff0c\u4ee5\u4fbf\u8fdb\u884c\u8bad\u7ec3\u548c\u9a8c\u8bc1\u3002</p>\n",
"<p>Autoregressive model </p>\n": "<p>\u81ea\u56de\u5f52\u6a21\u578b</p>\n",
"<p>Batch size </p>\n": "<p>\u6279\u91cf\u5927\u5c0f</p>\n",
"<p>Calculate and log accuracy </p>\n": "<p>\u8ba1\u7b97\u548c\u8bb0\u5f55\u7cbe\u5ea6</p>\n",
"<p>Calculate and log loss </p>\n": "<p>\u8ba1\u7b97\u5e76\u8bb0\u5f55\u635f\u5931</p>\n",
"<p>Calculate gradients </p>\n": "<p>\u8ba1\u7b97\u68af\u5ea6</p>\n",
"<p>Clear the gradients </p>\n": "<p>\u6e05\u9664\u6e10\u53d8</p>\n",
"<p>Clip gradients </p>\n": "<p>\u526a\u8f91\u6e10\u53d8</p>\n",
"<p>Collect tokens from training dataset </p>\n": "<p>\u4ece\u8bad\u7ec3\u6570\u636e\u96c6\u4e2d\u6536\u96c6\u4ee4\u724c</p>\n",
"<p>Collect tokens from validation dataset </p>\n": "<p>\u4ece\u9a8c\u8bc1\u6570\u636e\u96c6\u4e2d\u6536\u96c6\u4ee4\u724c</p>\n",
"<p>Create <a href=\"../utils.html#map_style_dataset\">map-style datasets</a> </p>\n": "<p>\u521b\u5efa<a href=\"../utils.html#map_style_dataset\">\u5730\u56fe\u6837\u5f0f\u6570\u636e\u96c6</a></p>\n",
"<p>Create a counter </p>\n": "<p>\u521b\u5efa\u8ba1\u6570\u5668</p>\n",
"<p>Create training data loader </p>\n": "<p>\u521b\u5efa\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p>Create validation data loader </p>\n": "<p>\u521b\u5efa\u9a8c\u8bc1\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p>Create vocabulary </p>\n": "<p>\u521b\u5efa\u8bcd\u6c47</p>\n",
"<p>Empty labels tensor </p>\n": "<p>\u7a7a\u6807\u7b7e\u5f20\u91cf</p>\n",
"<p>Get model outputs. It&#x27;s returning a tuple for states when using RNNs. This is not implemented yet. \ud83d\ude1c </p>\n": "<p>\u83b7\u53d6\u6a21\u578b\u8f93\u51fa\u3002\u5b83\u5728\u4f7f\u7528 RNN \u65f6\u8fd4\u56de\u72b6\u6001\u7684\u5143\u7ec4\u3002\u8fd9\u8fd8\u6ca1\u6709\u5b9e\u73b0\u3002\ud83d\ude1c</p>\n",
"<p>Get tokenizer </p>\n": "<p>\u83b7\u53d6\u5206\u8bcd\u5668</p>\n",
"<p>Get training and validation datasets </p>\n": "<p>\u83b7\u53d6\u8bad\u7ec3\u548c\u9a8c\u8bc1\u6570\u636e\u96c6</p>\n",
"<p>Gradient clipping </p>\n": "<p>\u6e10\u53d8\u526a\u5207</p>\n",
"<p>Input data tensor, initialized with <span translate=no>_^_0_^_</span> </p>\n": "<p>\u8f93\u5165\u6570\u636e\u5f20\u91cf\uff0c\u521d\u59cb\u5316\u4e3a<span translate=no>_^_0_^_</span></p>\n",
"<p>Length of the sequence, or context size </p>\n": "<p>\u5e8f\u5217\u7684\u957f\u5ea6\u6216\u4e0a\u4e0b\u6587\u5927\u5c0f</p>\n",
"<p>Load data to memory </p>\n": "<p>\u5c06\u6570\u636e\u52a0\u8f7d\u5230\u5185\u5b58</p>\n",
"<p>Log the model parameters and gradients on last batch of every epoch </p>\n": "<p>\u8bb0\u5f55\u6bcf\u4e2a\u7eaa\u5143\u6700\u540e\u4e00\u6279\u7684\u6a21\u578b\u53c2\u6570\u548c\u68af\u5ea6</p>\n",
"<p>Loop through the samples </p>\n": "<p>\u5faa\u73af\u6d4f\u89c8\u6837\u672c</p>\n",
"<p>Loss function </p>\n": "<p>\u4e8f\u635f\u51fd\u6570</p>\n",
"<p>Model embedding size </p>\n": "<p>\u6a21\u578b\u5d4c\u5165\u5927\u5c0f</p>\n",
"<p>Move data to the device </p>\n": "<p>\u5c06\u6570\u636e\u79fb\u52a8\u5230\u8bbe\u5907</p>\n",
"<p>Number of classes </p>\n": "<p>\u73ed\u7ea7\u6570</p>\n",
"<p>Number of token in vocabulary </p>\n": "<p>\u8bcd\u6c47\u4e2d\u7684\u4ee3\u5e01\u6570\u91cf</p>\n",
"<p>Optimizer </p>\n": "<p>\u4f18\u5316\u5668</p>\n",
"<p>Return <span translate=no>_^_0_^_</span>, <span translate=no>_^_1_^_</span>, <span translate=no>_^_2_^_</span>, and <span translate=no>_^_3_^_</span> </p>\n": "<p>\u8fd4\u56de<span translate=no>_^_0_^_</span><span translate=no>_^_1_^_</span>\u3001<span translate=no>_^_2_^_</span>\u3001\u548c<span translate=no>_^_3_^_</span></p>\n",
"<p>Save the tracked metrics </p>\n": "<p>\u4fdd\u5b58\u8ddf\u8e2a\u7684\u6307\u6807</p>\n",
"<p>Set the final token in the sequence to <span translate=no>_^_0_^_</span> </p>\n": "<p>\u5c06\u5e8f\u5217\u4e2d\u7684\u6700\u540e\u4e00\u4e2a\u4ee4\u724c\u8bbe\u7f6e\u4e3a<span translate=no>_^_0_^_</span></p>\n",
"<p>Set the label </p>\n": "<p>\u8bbe\u7f6e\u6807\u7b7e</p>\n",
"<p>Set tracker configurations </p>\n": "<p>\u8bbe\u7f6e\u8ddf\u8e2a\u5668\u914d\u7f6e</p>\n",
"<p>Take optimizer step </p>\n": "<p>\u91c7\u53d6\u4f18\u5316\u5668\u6b65\u9aa4</p>\n",
"<p>Tokenize the input text </p>\n": "<p>\u6807\u8bb0\u8f93\u5165\u6587\u672c</p>\n",
"<p>Tokenizer </p>\n": "<p>\u5206\u8bcd\u5668</p>\n",
"<p>Train the model </p>\n": "<p>\u8bad\u7ec3\u6a21\u578b</p>\n",
"<p>Training data loader </p>\n": "<p>\u8bad\u7ec3\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p>Training device </p>\n": "<p>\u8bad\u7ec3\u8bbe\u5907</p>\n",
"<p>Transpose and add to data </p>\n": "<p>\u8f6c\u7f6e\u5e76\u6dfb\u52a0\u5230\u6570\u636e</p>\n",
"<p>Truncate upto <span translate=no>_^_0_^_</span> </p>\n": "<p>\u622a\u65ad\u6700\u591a<span translate=no>_^_0_^_</span></p>\n",
"<p>Update global step (number of tokens processed) when in training mode </p>\n": "<p>\u5728\u8bad\u7ec3\u6a21\u5f0f\u4e0b\u66f4\u65b0\u5168\u5c40\u6b65\u957f\uff08\u5904\u7406\u7684\u4ee4\u724c\u6570\uff09</p>\n",
"<p>Validation data loader </p>\n": "<p>\u9a8c\u8bc1\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
"<p>Vocabulary </p>\n": "<p>\u8bcd\u6c47</p>\n",
"<p>Whether to capture model outputs </p>\n": "<p>\u662f\u5426\u6355\u83b7\u6a21\u578b\u8f93\u51fa</p>\n",
"<p>Whether to log model activations (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u662f\u5426\u8bb0\u5f55\u6a21\u578b\u6fc0\u6d3b\uff08\u6bcf\u4e2a\u7eaa\u5143\u4e00\u6b21\uff09\u3002\u8fd9\u4e9b\u662f\u6bcf\u5c42\u7684\u6c47\u603b\u7edf\u8ba1\u6570\u636e\uff0c\u4f46\u5b83\u4ecd\u7136\u53ef\u80fd\u5bfc\u81f4\u975e\u5e38\u6df1\u7684\u7f51\u7edc\u7684\u8bb8\u591a\u6307\u6807\u3002</p>\n",
"<p>Whether to log model parameters and gradients (once per epoch). These are summarized stats per layer, but it could still lead to many indicators for very deep networks. </p>\n": "<p>\u662f\u5426\u8bb0\u5f55\u6a21\u578b\u53c2\u6570\u548c\u68af\u5ea6\uff08\u6bcf\u4e2a\u7eaa\u5143\u4e00\u6b21\uff09\u3002\u8fd9\u4e9b\u662f\u6bcf\u5c42\u7684\u6c47\u603b\u7edf\u8ba1\u6570\u636e\uff0c\u4f46\u5b83\u4ecd\u7136\u53ef\u80fd\u5bfc\u81f4\u975e\u5e38\u6df1\u7684\u7f51\u7edc\u7684\u8bb8\u591a\u6307\u6807\u3002</p>\n",
"<p>Whether to periodically save models </p>\n": "<p>\u662f\u5426\u5b9a\u671f\u4fdd\u5b58\u6a21\u578b</p>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the batch of data collected by the <span translate=no>_^_1_^_</span></li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u662f\u7531<span translate=no>_^_1_^_</span></li></ul>\n",
"<ul><li><span translate=no>_^_0_^_</span> is the tokenizer function </li>\n<li><span translate=no>_^_1_^_</span> is the vocabulary </li>\n<li><span translate=no>_^_2_^_</span> is the length of the sequence </li>\n<li><span translate=no>_^_3_^_</span> is the token used for padding when the <span translate=no>_^_4_^_</span> is larger than the text length </li>\n<li><span translate=no>_^_5_^_</span> is the <span translate=no>_^_6_^_</span> token which we set at end of the input</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u662f\u5206\u8bcd\u5668\u51fd\u6570</li>\n<li><span translate=no>_^_1_^_</span>\u662f\u8bcd\u6c47</li>\n<li><span translate=no>_^_2_^_</span>\u662f\u5e8f\u5217\u7684\u957f\u5ea6</li>\n<li><span translate=no>_^_3_^_</span>\u662f\u5927\u4e8e\u6587\u672c\u957f\u5ea6\u65f6<span translate=no>_^_4_^_</span>\u7528\u4e8e\u586b\u5145\u7684\u6807\u8bb0</li>\n<li><span translate=no>_^_5_^_</span>\u662f\u6211\u4eec\u5728\u8f93\u5165\u672b\u5c3e\u8bbe\u7f6e\u7684<span translate=no>_^_6_^_</span>\u4ee4\u724c</li></ul>\n",
"NLP classification trainer": "NLP \u5206\u7c7b\u57f9\u8bad\u5e08",
"This is a reusable trainer for classification tasks": "\u8fd9\u662f\u4e00\u6b3e\u53ef\u91cd\u590d\u4f7f\u7528\u7684\u5206\u7c7b\u4efb\u52a1\u8bad\u7ec3\u5668"
}