Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/log.txt.
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mcola[0m, split [94mtrain[0m.
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mcola[0m, split [94mvalidation[0m.
Loaded dataset. Found: 2 labels: ([0, 1])
Loading transformers AutoModelForSequenceClassification: xlnet-base-cased
Tokenizing training data. (len: 8551)
Tokenizing eval data (len: 1043)
Loaded data and tokenized in 19.534267902374268s
Training model across 4 GPUs
***** Running training *****
	Num examples = 8551
	Batch size = 32
	Max sequence length = 128
	Num steps = 1335
	Num epochs = 5
	Learning rate = 3e-05
Eval accuracy: 69.22339405560882%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/.
Eval accuracy: 79.09875359539788%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/.
Eval accuracy: 79.76989453499522%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/.
Eval accuracy: 79.00287631831256%
Eval accuracy: 79.00287631831256%
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7f15171eebb0> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/.
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/README.md.
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:cola-2020-06-29-17:32/train_args.json.