system's picture
system HF staff
Update log.txt
385cbb2
Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/log.txt.
Loading nlp dataset rotten_tomatoes, split train.
Loading nlp dataset rotten_tomatoes, split validation.
Loaded dataset. Found: 2 labels: ([0, 1])
Loading transformers AutoModelForSequenceClassification: roberta-base
Tokenizing training data. (len: 8530)
Tokenizing eval data (len: 1066)
Loaded data and tokenized in 10.02656078338623s
Training model across 4 GPUs
***** Running training *****
Num examples = 8530
Batch size = 128
Max sequence length = 128
Num steps = 660
Num epochs = 10
Learning rate = 5e-05
Eval accuracy: 89.11819887429644%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/.
Eval accuracy: 90.0562851782364%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/.
Eval accuracy: 89.9624765478424%
Eval accuracy: 89.77485928705441%
Eval accuracy: 87.99249530956847%
Eval accuracy: 89.02439024390245%
Eval accuracy: 89.21200750469043%
Eval accuracy: 89.8686679174484%
Eval accuracy: 89.58724202626641%
Eval accuracy: 90.33771106941839%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/.
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7fb412aada60> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/.
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/README.md.
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-rotten_tomatoes-2020-06-25-13:09/train_args.json.