Commit
·
f2fa818
1
Parent(s):
426d2c2
Upload 5 files
Browse filesclassifier input sentence whether is toxic,yes or no
- config.json +27 -0
- pytorch_model.bin +3 -0
- toxic_classifier.log +152 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "checkpoints/ToxicClassifier/checkpoint-1500",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"problem_type": "single_label_classification",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.27.1",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 28996
|
| 27 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:781342ee2b8d910595f1cf3dda3c26ba2336aa04896b1a3dc17d5260b99f74e5
|
| 3 |
+
size 433315437
|
toxic_classifier.log
ADDED
|
@@ -0,0 +1,152 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Training/evaluation parameters Namespace(data_dir='bias_toxic_truthful/toxic/', model_dir='/data/ryanyip/Text-Classifier', model_name='bert-base-cased', output_dir='checkpoints/ToxicClassifier', do_train=True, do_predict=False, result_output_dir='checkpoints/ToxicClassifier/result', max_length=128, train_batch_size=16, eval_batch_size=16, learning_rate=5e-05, weight_decay=0.01, adam_epsilon=1e-08, max_grad_norm=1.0, epochs=10, warmup_proportion=0.1, earlystop_patience=2, logging_steps=10, save_steps=5000, seed=2021, device=device(type='cuda'), model_type='bert', task_name='qic')
|
| 2 |
+
***** Running training *****
|
| 3 |
+
Num samples 79212
|
| 4 |
+
Num epochs 10
|
| 5 |
+
Num training steps 49510
|
| 6 |
+
Num warmup steps 4951
|
| 7 |
+
***** Running evaluation *****
|
| 8 |
+
Num samples 9903
|
| 9 |
+
qic-bert-base-cased acc: 0.2161971119862668
|
| 10 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-10
|
| 11 |
+
***** Running evaluation *****
|
| 12 |
+
Num samples 9903
|
| 13 |
+
qic-bert-base-cased acc: 0.21791376350600827
|
| 14 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-20
|
| 15 |
+
***** Running evaluation *****
|
| 16 |
+
Num samples 9903
|
| 17 |
+
qic-bert-base-cased acc: 0.22003433303039482
|
| 18 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-30
|
| 19 |
+
***** Running evaluation *****
|
| 20 |
+
Num samples 9903
|
| 21 |
+
qic-bert-base-cased acc: 0.2237705745733616
|
| 22 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-40
|
| 23 |
+
***** Running evaluation *****
|
| 24 |
+
Num samples 9903
|
| 25 |
+
qic-bert-base-cased acc: 0.24012925376148642
|
| 26 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-50
|
| 27 |
+
***** Running evaluation *****
|
| 28 |
+
Num samples 9903
|
| 29 |
+
qic-bert-base-cased acc: 0.27819852569928305
|
| 30 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-60
|
| 31 |
+
***** Running evaluation *****
|
| 32 |
+
Num samples 9903
|
| 33 |
+
qic-bert-base-cased acc: 0.3636271836817126
|
| 34 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-70
|
| 35 |
+
***** Running evaluation *****
|
| 36 |
+
Num samples 9903
|
| 37 |
+
qic-bert-base-cased acc: 0.5175199434514793
|
| 38 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-80
|
| 39 |
+
***** Running evaluation *****
|
| 40 |
+
Num samples 9903
|
| 41 |
+
qic-bert-base-cased acc: 0.6735332727456327
|
| 42 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-90
|
| 43 |
+
***** Running evaluation *****
|
| 44 |
+
Num samples 9903
|
| 45 |
+
qic-bert-base-cased acc: 0.7494698576189034
|
| 46 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-100
|
| 47 |
+
***** Running evaluation *****
|
| 48 |
+
Num samples 9903
|
| 49 |
+
qic-bert-base-cased acc: 0.7759264869231546
|
| 50 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-110
|
| 51 |
+
***** Running evaluation *****
|
| 52 |
+
Num samples 9903
|
| 53 |
+
qic-bert-base-cased acc: 0.7842068060183782
|
| 54 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-120
|
| 55 |
+
***** Running evaluation *****
|
| 56 |
+
Num samples 9903
|
| 57 |
+
qic-bert-base-cased acc: 0.7872361910532162
|
| 58 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-130
|
| 59 |
+
***** Running evaluation *****
|
| 60 |
+
Num samples 9903
|
| 61 |
+
qic-bert-base-cased acc: 0.7874381500555387
|
| 62 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-140
|
| 63 |
+
***** Running evaluation *****
|
| 64 |
+
Num samples 9903
|
| 65 |
+
qic-bert-base-cased acc: 0.7877410885590225
|
| 66 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-150
|
| 67 |
+
***** Running evaluation *****
|
| 68 |
+
Num samples 9903
|
| 69 |
+
qic-bert-base-cased acc: 0.7877410885590225
|
| 70 |
+
Earlystopper counter: 1 out of 2
|
| 71 |
+
***** Running evaluation *****
|
| 72 |
+
Num samples 9903
|
| 73 |
+
qic-bert-base-cased acc: 0.7875391295567
|
| 74 |
+
Earlystopper counter: 2 out of 2
|
| 75 |
+
Training Stop! The best step 150: 0.7877410885590225
|
| 76 |
+
Saving models checkpoint to checkpoints/ToxicClassifier
|
| 77 |
+
Training/evaluation parameters Namespace(data_dir='bias_toxic_truthful/toxic/', model_dir='/data/ryanyip/Text-Classifier', model_name='bert-base-cased', output_dir='checkpoints/ToxicClassifier', do_train=True, do_predict=False, result_output_dir='checkpoints/ToxicClassifier/result', max_length=128, train_batch_size=32, eval_batch_size=16, learning_rate=5e-05, weight_decay=0.01, adam_epsilon=1e-08, max_grad_norm=1.0, epochs=20, warmup_proportion=0.1, earlystop_patience=2, logging_steps=100, save_steps=5000, seed=2021, device=device(type='cuda'), model_type='bert', task_name='qic')
|
| 78 |
+
***** Running training *****
|
| 79 |
+
Num samples 79212
|
| 80 |
+
Num epochs 20
|
| 81 |
+
Num training steps 49520
|
| 82 |
+
Num warmup steps 4952
|
| 83 |
+
***** Running evaluation *****
|
| 84 |
+
Num samples 9903
|
| 85 |
+
qic-bert-base-cased acc: 0.7775421589417348
|
| 86 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-100
|
| 87 |
+
***** Running evaluation *****
|
| 88 |
+
Num samples 9903
|
| 89 |
+
qic-bert-base-cased acc: 0.7878420680601838
|
| 90 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-200
|
| 91 |
+
***** Running evaluation *****
|
| 92 |
+
Num samples 9903
|
| 93 |
+
qic-bert-base-cased acc: 0.8174290619004342
|
| 94 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-300
|
| 95 |
+
***** Running evaluation *****
|
| 96 |
+
Num samples 9903
|
| 97 |
+
qic-bert-base-cased acc: 0.8627688579218419
|
| 98 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-400
|
| 99 |
+
***** Running evaluation *****
|
| 100 |
+
Num samples 9903
|
| 101 |
+
qic-bert-base-cased acc: 0.8869029586993841
|
| 102 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-500
|
| 103 |
+
***** Running evaluation *****
|
| 104 |
+
Num samples 9903
|
| 105 |
+
qic-bert-base-cased acc: 0.8974048268201555
|
| 106 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-600
|
| 107 |
+
***** Running evaluation *****
|
| 108 |
+
Num samples 9903
|
| 109 |
+
qic-bert-base-cased acc: 0.9020498838735737
|
| 110 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-700
|
| 111 |
+
***** Running evaluation *****
|
| 112 |
+
Num samples 9903
|
| 113 |
+
qic-bert-base-cased acc: 0.906392002423508
|
| 114 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-800
|
| 115 |
+
***** Running evaluation *****
|
| 116 |
+
Num samples 9903
|
| 117 |
+
qic-bert-base-cased acc: 0.9112390184792487
|
| 118 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-900
|
| 119 |
+
***** Running evaluation *****
|
| 120 |
+
Num samples 9903
|
| 121 |
+
qic-bert-base-cased acc: 0.9129556699989902
|
| 122 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-1000
|
| 123 |
+
***** Running evaluation *****
|
| 124 |
+
Num samples 9903
|
| 125 |
+
qic-bert-base-cased acc: 0.9129556699989902
|
| 126 |
+
Earlystopper counter: 1 out of 2
|
| 127 |
+
***** Running evaluation *****
|
| 128 |
+
Num samples 9903
|
| 129 |
+
qic-bert-base-cased acc: 0.9181056245582147
|
| 130 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-1200
|
| 131 |
+
***** Running evaluation *****
|
| 132 |
+
Num samples 9903
|
| 133 |
+
qic-bert-base-cased acc: 0.9213369685953752
|
| 134 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-1300
|
| 135 |
+
***** Running evaluation *****
|
| 136 |
+
Num samples 9903
|
| 137 |
+
qic-bert-base-cased acc: 0.9223467636069878
|
| 138 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-1400
|
| 139 |
+
***** Running evaluation *****
|
| 140 |
+
Num samples 9903
|
| 141 |
+
qic-bert-base-cased acc: 0.9253761486418257
|
| 142 |
+
Saving models checkpoint to checkpoints/ToxicClassifier/checkpoint-1500
|
| 143 |
+
***** Running evaluation *****
|
| 144 |
+
Num samples 9903
|
| 145 |
+
qic-bert-base-cased acc: 0.9202261940826012
|
| 146 |
+
Earlystopper counter: 1 out of 2
|
| 147 |
+
***** Running evaluation *****
|
| 148 |
+
Num samples 9903
|
| 149 |
+
qic-bert-base-cased acc: 0.9241643946278906
|
| 150 |
+
Earlystopper counter: 2 out of 2
|
| 151 |
+
Training Stop! The best step 1500: 0.9253761486418257
|
| 152 |
+
Saving models checkpoint to checkpoints/ToxicClassifier
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fec7d82d49b754566299b41a49c31d5fce788d84895912f06781897c0f8e3eb4
|
| 3 |
+
size 1135
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|