Update log.txt
Browse files
log.txt
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/log.txt.
|
| 2 |
+
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mrte[0m, split [94mtrain[0m.
|
| 3 |
+
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mrte[0m, split [94mvalidation[0m.
|
| 4 |
+
Loaded dataset. Found: 2 labels: [0, 1]
|
| 5 |
+
Loading transformers AutoModelForSequenceClassification: facebook/bart-base
|
| 6 |
+
Training model across 1 GPUs
|
| 7 |
+
Wrote original training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/train_args.json.
|
| 8 |
+
***** Running training *****
|
| 9 |
+
Num examples = 2490
|
| 10 |
+
Batch size = 64
|
| 11 |
+
Max sequence length = 128
|
| 12 |
+
Num steps = 190
|
| 13 |
+
Num epochs = 5
|
| 14 |
+
Learning rate = 2e-05
|
| 15 |
+
Train accuracy: 50.6425702811245%
|
| 16 |
+
Eval accuracy: 58.48375451263538%
|
| 17 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/.
|
| 18 |
+
Saved updated args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/train_args.json
|
| 19 |
+
Train accuracy: 65.3012048192771%
|
| 20 |
+
Eval accuracy: 70.3971119133574%
|
| 21 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/.
|
| 22 |
+
Saved updated args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/train_args.json
|
| 23 |
+
Train accuracy: 77.75100401606426%
|
| 24 |
+
Eval accuracy: 70.3971119133574%
|
| 25 |
+
Train accuracy: 84.17670682730923%
|
| 26 |
+
Eval accuracy: 71.84115523465704%
|
| 27 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/.
|
| 28 |
+
Saved updated args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/train_args.json
|
| 29 |
+
Train accuracy: 89.87951807228916%
|
| 30 |
+
Eval accuracy: 72.56317689530685%
|
| 31 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/.
|
| 32 |
+
Saved updated args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/train_args.json
|
| 33 |
+
Finished training. Re-loading and evaluating model from disk.
|
| 34 |
+
Loading transformers AutoModelForSequenceClassification: facebook/bart-base
|
| 35 |
+
Saved model accuracy: 72.56317689530685%
|
| 36 |
+
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7f056ac36910> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/.
|
| 37 |
+
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/README.md.
|
| 38 |
+
Wrote final training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/facebook/bart-base-glue^rte-2020-08-20-11-41-21-898210/train_args.json.
|