Update log.txt
Browse files
log.txt
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/log.txt.
|
| 2 |
+
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mcola[0m, split [94mtrain[0m.
|
| 3 |
+
Loading [94mnlp[0m dataset [94mglue[0m, subset [94mcola[0m, split [94mvalidation[0m.
|
| 4 |
+
Loaded dataset. Found: 2 labels: ([0, 1])
|
| 5 |
+
Loading transformers AutoModelForSequenceClassification: albert-base-v2
|
| 6 |
+
Tokenizing training data. (len: 8551)
|
| 7 |
+
Tokenizing eval data (len: 1043)
|
| 8 |
+
Loaded data and tokenized in 21.12540578842163s
|
| 9 |
+
Training model across 1 GPUs
|
| 10 |
+
***** Running training *****
|
| 11 |
+
Num examples = 8551
|
| 12 |
+
Batch size = 32
|
| 13 |
+
Max sequence length = 128
|
| 14 |
+
Num steps = 1335
|
| 15 |
+
Num epochs = 5
|
| 16 |
+
Learning rate = 3e-05
|
| 17 |
+
Eval accuracy: 75.26366251198466%
|
| 18 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/.
|
| 19 |
+
Eval accuracy: 80.72866730584852%
|
| 20 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/.
|
| 21 |
+
Eval accuracy: 82.45445829338448%
|
| 22 |
+
Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/.
|
| 23 |
+
Eval accuracy: 81.39980824544583%
|
| 24 |
+
Eval accuracy: 81.01629913710451%
|
| 25 |
+
Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7f52294d2ca0> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/.
|
| 26 |
+
Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/README.md.
|
| 27 |
+
Wrote training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/albert-base-v2-glue:cola-2020-06-29-01:49/train_args.json.
|