Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:mrpc-2020-06-29-13:11/log.txt.
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mmrpc[0m, split [94mtrain[0m.
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mmrpc[0m, split [94mvalidation[0m.
Loaded dataset. Found: 2 labels: ([0, 1])
Loading transformers AutoModelForSequenceClassification: distilbert-base-uncased
Tokenizing training data. (len: 3668)
Tokenizing eval data (len: 408)
Loaded data and tokenized in 10.926565647125244s
Training model across 4 GPUs
***** Running training *****
	Num examples = 3668
	Batch size = 32
	Max sequence length = 256
	Num steps = 570
	Num epochs = 5
	Learning rate = 2e-05
Eval accuracy: 80.88235294117648%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:mrpc-2020-06-29-13:11/.
Eval accuracy: 85.7843137254902%
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:mrpc-2020-06-29-13:11/.
Eval accuracy: 85.04901960784314%
Eval accuracy: 85.5392156862745%
Eval accuracy: 85.29411764705883%
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7f81f8485160> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:mrpc-2020-06-29-13:11/.
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:mrpc-2020-06-29-13:11/README.md.
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/distilbert-base-uncased-glue:mrpc-2020-06-29-13:11/train_args.json.