Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:wnli-2020-06-29-10:28/log.txt.
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mwnli[0m, split [94mtrain[0m.
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mwnli[0m, split [94mvalidation[0m.
Loaded dataset. Found: 2 labels: ([0, 1])
Loading transformers AutoModelForSequenceClassification: xlnet-base-cased
Tokenizing training data. (len: 635)
Tokenizing eval data (len: 71)
Loaded data and tokenized in 8.763058185577393s
Training model across 1 GPUs
***** Running training *****
	Num examples = 635
	Batch size = 16
	Max sequence length = 256
	Num steps = 195
	Num epochs = 5
	Learning rate = 3e-05
Eval accuracy: 57.74647887323944%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:wnli-2020-06-29-10:28/.
Eval accuracy: 56.33802816901409%
Eval accuracy: 45.07042253521127%
Eval accuracy: 45.07042253521127%
Eval accuracy: 42.25352112676056%
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7f23225c6af0> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:wnli-2020-06-29-10:28/.
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:wnli-2020-06-29-10:28/README.md.
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-glue:wnli-2020-06-29-10:28/train_args.json.