cocoons

Ariannapera's group workspace

Group: analyse-tal

1-11

of 11

Timestamps visible

2023-03-30 20:27:47

{'eval_loss': 0.47326168751822284, 'eval_f1': 0.7627308101951935, 'eval_roc_auc': 0.8372369638689949, 'eval_accuracy': 0.84, 'eval_runtime': 7.2457, 'eval_samples_per_second': 103.51, 'eval_steps_per_second': 3.312, 'epoch': 8.0}

2023-03-30 20:34:03

100%|██████████| 1720/1720 [30:47<00:00, 1.07s/it]

2023-03-30 20:30:47

{'loss': 0.0506, 'learning_rate': 2.0000000000000003e-06, 'epoch': 9.0}

2023-03-30 20:30:53

{'eval_loss': 0.4799283541848895, 'eval_f1': 0.757713748161855, 'eval_roc_auc': 0.8357356964673694, 'eval_accuracy': 0.8333333333333334, 'eval_runtime': 7.248, 'eval_samples_per_second': 103.477, 'eval_steps_per_second': 3.311, 'epoch': 9.0}

2023-03-30 20:33:49

{'loss': 0.0439, 'learning_rate': 0.0, 'epoch': 10.0}

2023-03-30 20:33:51

29%|██▉ | 7/24 [00:01<00:04, 3.43it/s]

2023-03-30 20:33:57

{'eval_loss': 0.47726168751822284, 'eval_f1': 0.7594969379854055, 'eval_roc_auc': 0.8325822805182264, 'eval_accuracy': 0.836, 'eval_runtime': 7.2459, 'eval_samples_per_second': 103.507, 'eval_steps_per_second': 3.312, 'epoch': 10.0}

2023-03-30 20:34:11

100%|██████████| 24/24 [00:06<00:00, 3.47it/s]

2023-03-30 20:34:03

{'train_runtime': 1847.3406, 'train_samples_per_second': 29.773, 'train_steps_per_second': 0.931, 'train_loss': 0.19254385288371595, 'epoch': 10.0}

2023-03-30 20:35:15

100%|██████████| 212/212 [01:04<00:00, 3.27it/s]

2023-03-30 20:34:11

{'eval_loss': 0.46126168751822283, 'eval_f1': 0.7669966497896672, 'eval_roc_auc': 0.8664651565140421, 'eval_accuracy': 0.852, 'eval_runtime': 7.2464, 'eval_samples_per_second': 103.499, 'eval_steps_per_second': 3.312, 'epoch': 10.0}

2023-03-30 20:35:15

/home/agmo/worker_vs_gpt/src/worker_vs_gpt/classification/trainers.py:192: UserWarning: Implicit dimension choice for softmax has been deprecated. Change the call to include dim=X as an argument.

2023-03-30 20:35:15

probs = softmax(torch.Tensor(y_pred_logits)).numpy()