Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/log.txt. Loading nlp dataset imdb, split train. Loading nlp dataset imdb, split test. Loaded dataset. Found: 2 labels: ([0, 1]) Loading transformers AutoModelForSequenceClassification: xlnet-base-cased Tokenizing training data. (len: 25000) Tokenizing eval data (len: 25000) Loaded data and tokenized in 76.48014068603516s Training model across 4 GPUs ***** Running training ***** Num examples = 25000 Batch size = 32 Max sequence length = 512 Num steps = 3905 Num epochs = 5 Learning rate = 2e-05 Eval accuracy: 94.98400000000001% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/. Eval accuracy: 95.024% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/. Eval accuracy: 95.352% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/. Eval accuracy: 95.00800000000001% Eval accuracy: 95.268% Saved tokenizer to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/. Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/README.md. Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/xlnet-base-cased-imdb-2020-06-28-20:07/train_args.json.