File size: 1,921 Bytes
e928596 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 |
Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/log.txt. Loading [94mnlp[0m dataset [94mglue[0m, subset [94mrte[0m, split [94mtrain[0m. Loading [94mnlp[0m dataset [94mglue[0m, subset [94mrte[0m, split [94mvalidation[0m. Loaded dataset. Found: 2 labels: ([0, 1]) Loading transformers AutoModelForSequenceClassification: roberta-base Tokenizing training data. (len: 2490) Tokenizing eval data (len: 277) Loaded data and tokenized in 18.287664651870728s Training model across 1 GPUs ***** Running training ***** Num examples = 2490 Batch size = 16 Max sequence length = 128 Num steps = 775 Num epochs = 5 Learning rate = 2e-05 Eval accuracy: 59.92779783393502% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/. Eval accuracy: 71.48014440433214% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/. Eval accuracy: 75.45126353790613% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/. Eval accuracy: 79.42238267148014% Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/. Eval accuracy: 78.33935018050542% Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7f386dbd2e80> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/. Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/README.md. Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/roberta-base-glue:rte-2020-06-29-12:22/train_args.json. |