Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-glue:wnli-2020-06-29-11:27/log.txt.
Loading nlp dataset glue, subset wnli, split train.
Loading nlp dataset glue, subset wnli, split validation.
Loaded dataset. Found: 2 labels: ([0, 1])
Loading transformers AutoModelForSequenceClassification: bert-base-uncased
Tokenizing training data. (len: 635)
Tokenizing eval data (len: 71)
Loaded data and tokenized in 7.102111577987671s
Training model across 4 GPUs
***** Running training *****
	Num examples = 635
	Batch size = 64
	Max sequence length = 256
	Num steps = 45
	Num epochs = 5
	Learning rate = 5e-05
Eval accuracy: 43.66197183098591%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-glue:wnli-2020-06-29-11:27/.
Eval accuracy: 56.33802816901409%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-glue:wnli-2020-06-29-11:27/.
Eval accuracy: 23.943661971830984%
Eval accuracy: 49.29577464788733%
Eval accuracy: 50.70422535211267%
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7fbf6aea2dc0> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-glue:wnli-2020-06-29-11:27/.
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-glue:wnli-2020-06-29-11:27/README.md.
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-glue:wnli-2020-06-29-11:27/train_args.json.