Rodrigo1771
commited on
Commit
•
1538eb6
1
Parent(s):
8e76363
End of training
Browse files- README.md +14 -13
- all_results.json +23 -16
- eval_results.json +9 -8
- predict_results.json +8 -8
- predictions.txt +0 -0
- tb/events.out.tfevents.1715612339.c331905616cf.3060.1 +3 -0
- train.log +50 -0
- train_results.json +9 -0
- trainer_state.json +388 -0
README.md
CHANGED
@@ -2,9 +2,10 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
-
- multi-train-drugtemist-dev-ner
|
8 |
metrics:
|
9 |
- precision
|
10 |
- recall
|
@@ -17,24 +18,24 @@ model-index:
|
|
17 |
name: Token Classification
|
18 |
type: token-classification
|
19 |
dataset:
|
20 |
-
name: multi-train-drugtemist-dev-ner
|
21 |
-
type: multi-train-drugtemist-dev-ner
|
22 |
config: MultiTrainDrugTEMISTDevNER
|
23 |
split: validation
|
24 |
args: MultiTrainDrugTEMISTDevNER
|
25 |
metrics:
|
26 |
- name: Precision
|
27 |
type: precision
|
28 |
-
value: 0.
|
29 |
- name: Recall
|
30 |
type: recall
|
31 |
-
value: 0.
|
32 |
- name: F1
|
33 |
type: f1
|
34 |
-
value: 0.
|
35 |
- name: Accuracy
|
36 |
type: accuracy
|
37 |
-
value: 0.
|
38 |
---
|
39 |
|
40 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -42,13 +43,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
42 |
|
43 |
# output
|
44 |
|
45 |
-
This model is a fine-tuned version of [PlanTL-GOB-ES/bsc-bio-ehr-es](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es) on the multi-train-drugtemist-dev-ner dataset.
|
46 |
It achieves the following results on the evaluation set:
|
47 |
-
- Loss:
|
48 |
-
- Precision: 0.
|
49 |
-
- Recall: 0.
|
50 |
-
- F1: 0.
|
51 |
-
- Accuracy: 0.
|
52 |
|
53 |
## Model description
|
54 |
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
|
4 |
tags:
|
5 |
+
- token-classification
|
6 |
- generated_from_trainer
|
7 |
datasets:
|
8 |
+
- Rodrigo1771/multi-train-drugtemist-dev-ner
|
9 |
metrics:
|
10 |
- precision
|
11 |
- recall
|
|
|
18 |
name: Token Classification
|
19 |
type: token-classification
|
20 |
dataset:
|
21 |
+
name: Rodrigo1771/multi-train-drugtemist-dev-ner
|
22 |
+
type: Rodrigo1771/multi-train-drugtemist-dev-ner
|
23 |
config: MultiTrainDrugTEMISTDevNER
|
24 |
split: validation
|
25 |
args: MultiTrainDrugTEMISTDevNER
|
26 |
metrics:
|
27 |
- name: Precision
|
28 |
type: precision
|
29 |
+
value: 0.09691960931630353
|
30 |
- name: Recall
|
31 |
type: recall
|
32 |
+
value: 0.9485294117647058
|
33 |
- name: F1
|
34 |
type: f1
|
35 |
+
value: 0.17586912065439672
|
36 |
- name: Accuracy
|
37 |
type: accuracy
|
38 |
+
value: 0.8099635429897495
|
39 |
---
|
40 |
|
41 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
43 |
|
44 |
# output
|
45 |
|
46 |
+
This model is a fine-tuned version of [PlanTL-GOB-ES/bsc-bio-ehr-es](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es) on the Rodrigo1771/multi-train-drugtemist-dev-ner dataset.
|
47 |
It achieves the following results on the evaluation set:
|
48 |
+
- Loss: 0.6631
|
49 |
+
- Precision: 0.0969
|
50 |
+
- Recall: 0.9485
|
51 |
+
- F1: 0.1759
|
52 |
+
- Accuracy: 0.8100
|
53 |
|
54 |
## Model description
|
55 |
|
all_results.json
CHANGED
@@ -1,19 +1,26 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"
|
7 |
-
"
|
|
|
8 |
"eval_samples": 6807,
|
9 |
-
"eval_samples_per_second":
|
10 |
-
"eval_steps_per_second":
|
11 |
-
"predict_accuracy": 0.
|
12 |
-
"predict_f1": 0.
|
13 |
-
"predict_loss":
|
14 |
-
"predict_precision": 0.
|
15 |
-
"predict_recall": 0.
|
16 |
-
"predict_runtime":
|
17 |
-
"predict_samples_per_second":
|
18 |
-
"predict_steps_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 9.997061416397296,
|
3 |
+
"eval_accuracy": 0.8099635429897495,
|
4 |
+
"eval_f1": 0.17586912065439672,
|
5 |
+
"eval_loss": 0.6631014347076416,
|
6 |
+
"eval_precision": 0.09691960931630353,
|
7 |
+
"eval_recall": 0.9485294117647058,
|
8 |
+
"eval_runtime": 15.8157,
|
9 |
"eval_samples": 6807,
|
10 |
+
"eval_samples_per_second": 430.395,
|
11 |
+
"eval_steps_per_second": 53.807,
|
12 |
+
"predict_accuracy": 0.8099635429897495,
|
13 |
+
"predict_f1": 0.17586912065439672,
|
14 |
+
"predict_loss": 0.6631014347076416,
|
15 |
+
"predict_precision": 0.09691960931630353,
|
16 |
+
"predict_recall": 0.9485294117647058,
|
17 |
+
"predict_runtime": 15.8984,
|
18 |
+
"predict_samples_per_second": 428.156,
|
19 |
+
"predict_steps_per_second": 53.527,
|
20 |
+
"total_flos": 6700722040732752.0,
|
21 |
+
"train_loss": 0.08913132946046923,
|
22 |
+
"train_runtime": 3337.712,
|
23 |
+
"train_samples": 27224,
|
24 |
+
"train_samples_per_second": 81.565,
|
25 |
+
"train_steps_per_second": 5.096
|
26 |
}
|
eval_results.json
CHANGED
@@ -1,11 +1,12 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"
|
7 |
-
"
|
|
|
8 |
"eval_samples": 6807,
|
9 |
-
"eval_samples_per_second":
|
10 |
-
"eval_steps_per_second":
|
11 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 9.997061416397296,
|
3 |
+
"eval_accuracy": 0.8099635429897495,
|
4 |
+
"eval_f1": 0.17586912065439672,
|
5 |
+
"eval_loss": 0.6631014347076416,
|
6 |
+
"eval_precision": 0.09691960931630353,
|
7 |
+
"eval_recall": 0.9485294117647058,
|
8 |
+
"eval_runtime": 15.8157,
|
9 |
"eval_samples": 6807,
|
10 |
+
"eval_samples_per_second": 430.395,
|
11 |
+
"eval_steps_per_second": 53.807
|
12 |
}
|
predict_results.json
CHANGED
@@ -1,10 +1,10 @@
|
|
1 |
{
|
2 |
-
"predict_accuracy": 0.
|
3 |
-
"predict_f1": 0.
|
4 |
-
"predict_loss":
|
5 |
-
"predict_precision": 0.
|
6 |
-
"predict_recall": 0.
|
7 |
-
"predict_runtime":
|
8 |
-
"predict_samples_per_second":
|
9 |
-
"predict_steps_per_second":
|
10 |
}
|
|
|
1 |
{
|
2 |
+
"predict_accuracy": 0.8099635429897495,
|
3 |
+
"predict_f1": 0.17586912065439672,
|
4 |
+
"predict_loss": 0.6631014347076416,
|
5 |
+
"predict_precision": 0.09691960931630353,
|
6 |
+
"predict_recall": 0.9485294117647058,
|
7 |
+
"predict_runtime": 15.8984,
|
8 |
+
"predict_samples_per_second": 428.156,
|
9 |
+
"predict_steps_per_second": 53.527
|
10 |
}
|
predictions.txt
CHANGED
The diff for this file is too large to render.
See raw diff
|
|
tb/events.out.tfevents.1715612339.c331905616cf.3060.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fd631ac54bf6215fcba3db4eb9803c4beb0d2d937cbe6eea52ab0ff263ca8fae
|
3 |
+
size 569
|
train.log
CHANGED
@@ -1613,3 +1613,53 @@ Training completed. Do not forget to share your model on huggingface.co/models =
|
|
1613 |
[INFO|modeling_utils.py:2590] 2024-05-13 14:58:40,302 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
|
1614 |
[INFO|tokenization_utils_base.py:2488] 2024-05-13 14:58:40,303 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
|
1615 |
[INFO|tokenization_utils_base.py:2497] 2024-05-13 14:58:40,303 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1616 |
0%| | 0/851 [00:00<?, ?it/s]
|
1617 |
1%| | 10/851 [00:00<00:09, 92.39it/s]
|
1618 |
2%|▏ | 20/851 [00:00<00:10, 78.68it/s]
|
1619 |
3%|▎ | 28/851 [00:00<00:10, 75.69it/s]
|
1620 |
4%|▍ | 36/851 [00:00<00:10, 74.36it/s]
|
1621 |
5%|▌ | 44/851 [00:00<00:10, 75.32it/s]
|
1622 |
6%|▌ | 52/851 [00:00<00:10, 75.93it/s]
|
1623 |
7%|▋ | 60/851 [00:00<00:10, 76.39it/s]
|
1624 |
8%|▊ | 68/851 [00:00<00:11, 70.86it/s]
|
1625 |
9%|▉ | 76/851 [00:01<00:11, 70.00it/s]
|
1626 |
10%|▉ | 84/851 [00:01<00:10, 70.49it/s]
|
1627 |
11%|█ | 92/851 [00:01<00:10, 70.55it/s]
|
1628 |
12%|█▏ | 100/851 [00:01<00:10, 70.25it/s]
|
1629 |
13%|█▎ | 108/851 [00:01<00:10, 69.63it/s]
|
1630 |
14%|█▎ | 115/851 [00:01<00:10, 69.25it/s]
|
1631 |
15%|█▍ | 124/851 [00:01<00:09, 73.89it/s]
|
1632 |
16%|█▌ | 132/851 [00:01<00:10, 66.17it/s]
|
1633 |
16%|█▋ | 139/851 [00:01<00:10, 66.18it/s]
|
1634 |
17%|█▋ | 147/851 [00:02<00:10, 67.87it/s]
|
1635 |
18%|█▊ | 154/851 [00:02<00:10, 67.52it/s]
|
1636 |
19%|█▉ | 162/851 [00:02<00:09, 69.68it/s]
|
1637 |
20%|█▉ | 170/851 [00:02<00:09, 71.19it/s]
|
1638 |
21%|██ | 178/851 [00:02<00:09, 73.27it/s]
|
1639 |
22%|██▏ | 186/851 [00:02<00:09, 73.02it/s]
|
1640 |
23%|██▎ | 194/851 [00:02<00:08, 74.19it/s]
|
1641 |
24%|██▍ | 203/851 [00:02<00:08, 76.36it/s]
|
1642 |
25%|██▍ | 211/851 [00:02<00:08, 71.75it/s]
|
1643 |
26%|██▌ | 219/851 [00:03<00:09, 69.67it/s]
|
1644 |
27%|██▋ | 227/851 [00:03<00:08, 71.38it/s]
|
1645 |
28%|██▊ | 235/851 [00:03<00:08, 72.05it/s]
|
1646 |
29%|██▊ | 243/851 [00:03<00:09, 67.00it/s]
|
1647 |
29%|██▉ | 251/851 [00:03<00:08, 69.80it/s]
|
1648 |
30%|███ | 259/851 [00:03<00:08, 72.22it/s]
|
1649 |
31%|███▏ | 267/851 [00:03<00:08, 72.74it/s]
|
1650 |
32%|███▏ | 275/851 [00:03<00:07, 73.62it/s]
|
1651 |
33%|███▎ | 283/851 [00:03<00:07, 74.61it/s]
|
1652 |
34%|███▍ | 291/851 [00:04<00:07, 72.54it/s]
|
1653 |
35%|███▌ | 299/851 [00:04<00:07, 74.26it/s]
|
1654 |
36%|███▌ | 307/851 [00:04<00:07, 75.74it/s]
|
1655 |
37%|███▋ | 315/851 [00:04<00:07, 70.27it/s]
|
1656 |
38%|███▊ | 323/851 [00:04<00:07, 72.85it/s]
|
1657 |
39%|███▉ | 331/851 [00:04<00:07, 72.36it/s]
|
1658 |
40%|███▉ | 339/851 [00:04<00:07, 71.89it/s]
|
1659 |
41%|████ | 347/851 [00:04<00:06, 72.58it/s]
|
1660 |
42%|████▏ | 355/851 [00:04<00:07, 69.70it/s]
|
1661 |
43%|████▎ | 363/851 [00:05<00:06, 69.83it/s]
|
1662 |
44%|████▎ | 371/851 [00:05<00:06, 69.06it/s]
|
1663 |
44%|████▍ | 378/851 [00:05<00:06, 69.30it/s]
|
1664 |
45%|████▌ | 385/851 [00:05<00:06, 68.06it/s]
|
1665 |
46%|████▌ | 393/851 [00:05<00:06, 69.52it/s]
|
1666 |
47%|████▋ | 401/851 [00:05<00:06, 69.90it/s]
|
1667 |
48%|████▊ | 408/851 [00:05<00:06, 64.81it/s]
|
1668 |
49%|████▉ | 416/851 [00:05<00:06, 67.34it/s]
|
1669 |
50%|████▉ | 424/851 [00:05<00:06, 69.53it/s]
|
1670 |
51%|█████ | 431/851 [00:06<00:06, 68.95it/s]
|
1671 |
52%|█████▏ | 439/851 [00:06<00:05, 71.11it/s]
|
1672 |
53%|█████▎ | 447/851 [00:06<00:05, 70.55it/s]
|
1673 |
53%|█████▎ | 455/851 [00:06<00:05, 71.50it/s]
|
1674 |
54%|█████▍ | 463/851 [00:06<00:05, 70.54it/s]
|
1675 |
55%|█████▌ | 471/851 [00:06<00:06, 61.54it/s]
|
1676 |
56%|█████▌ | 478/851 [00:06<00:05, 63.35it/s]
|
1677 |
57%|█████▋ | 485/851 [00:06<00:05, 63.12it/s]
|
1678 |
58%|█████▊ | 493/851 [00:06<00:05, 66.52it/s]
|
1679 |
59%|█████▉ | 502/851 [00:07<00:04, 70.10it/s]
|
1680 |
60%|█████▉ | 510/851 [00:07<00:04, 69.87it/s]
|
1681 |
61%|██████ | 518/851 [00:07<00:04, 71.30it/s]
|
1682 |
62%|██████▏ | 526/851 [00:07<00:04, 66.37it/s]
|
1683 |
63%|██████▎ | 534/851 [00:07<00:04, 68.15it/s]
|
1684 |
64%|██████▎ | 542/851 [00:07<00:04, 69.51it/s]
|
1685 |
65%|██████▍ | 550/851 [00:07<00:04, 67.52it/s]
|
1686 |
66%|██████▌ | 558/851 [00:07<00:04, 69.41it/s]
|
1687 |
67%|██████▋ | 566/851 [00:08<00:03, 72.24it/s]
|
1688 |
67%|██████▋ | 574/851 [00:08<00:03, 72.87it/s]
|
1689 |
68%|██████▊ | 582/851 [00:08<00:03, 69.95it/s]
|
1690 |
69%|██████▉ | 590/851 [00:08<00:03, 67.24it/s]
|
1691 |
70%|███████ | 597/851 [00:08<00:03, 67.77it/s]
|
1692 |
71%|███████ | 605/851 [00:08<00:03, 67.61it/s]
|
1693 |
72%|███████▏ | 612/851 [00:08<00:03, 66.37it/s]
|
1694 |
73%|███████▎ | 619/851 [00:08<00:03, 64.78it/s]
|
1695 |
74%|███████▎ | 626/851 [00:08<00:03, 64.41it/s]
|
1696 |
74%|███████▍ | 633/851 [00:09<00:03, 64.56it/s]
|
1697 |
75%|███████▌ | 640/851 [00:09<00:03, 64.37it/s]
|
1698 |
76%|███████▌ | 647/851 [00:09<00:03, 62.56it/s]
|
1699 |
77%|███████▋ | 655/851 [00:09<00:02, 65.73it/s]
|
1700 |
78%|███████▊ | 662/851 [00:09<00:02, 66.42it/s]
|
1701 |
79%|███████▊ | 670/851 [00:09<00:02, 67.98it/s]
|
1702 |
80%|███████▉ | 678/851 [00:09<00:02, 69.04it/s]
|
1703 |
81%|████████ | 686/851 [00:09<00:02, 69.89it/s]
|
1704 |
82%|████████▏ | 694/851 [00:09<00:02, 71.78it/s]
|
1705 |
82%|████████▏ | 702/851 [00:10<00:02, 72.47it/s]
|
1706 |
83%|████████▎ | 710/851 [00:10<00:01, 73.89it/s]
|
1707 |
84%|████████▍ | 718/851 [00:10<00:01, 70.91it/s]
|
1708 |
85%|████████▌ | 726/851 [00:10<00:01, 70.61it/s]
|
1709 |
86%|████████▋ | 734/851 [00:10<00:01, 72.17it/s]
|
1710 |
87%|████████▋ | 742/851 [00:10<00:01, 73.05it/s]
|
1711 |
88%|████████▊ | 750/851 [00:10<00:01, 72.59it/s]
|
1712 |
89%|████████▉ | 758/851 [00:10<00:01, 73.09it/s]
|
1713 |
90%|█████████ | 766/851 [00:10<00:01, 69.09it/s]
|
1714 |
91%|█████████ | 774/851 [00:11<00:01, 68.91it/s]
|
1715 |
92%|█████████▏| 781/851 [00:11<00:01, 66.57it/s]
|
1716 |
93%|█████████▎| 788/851 [00:11<00:00, 67.34it/s]
|
1717 |
94%|█████████▎| 796/851 [00:11<00:00, 69.12it/s]
|
1718 |
94%|█████████▍| 804/851 [00:11<00:00, 71.53it/s]
|
1719 |
95%|█████████▌| 812/851 [00:11<00:00, 69.03it/s]
|
1720 |
96%|█████████▋| 820/851 [00:11<00:00, 69.99it/s]
|
1721 |
97%|█████████▋| 828/851 [00:11<00:00, 71.40it/s]
|
1722 |
98%|█████████▊| 836/851 [00:11<00:00, 71.06it/s]
|
1723 |
99%|█████████▉| 844/851 [00:12<00:00, 67.57it/s]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1724 |
0%| | 0/851 [00:00<?, ?it/s]
|
1725 |
1%| | 10/851 [00:00<00:08, 94.43it/s]
|
1726 |
2%|▏ | 20/851 [00:00<00:10, 79.64it/s]
|
1727 |
3%|▎ | 29/851 [00:00<00:10, 75.61it/s]
|
1728 |
4%|▍ | 37/851 [00:00<00:11, 73.82it/s]
|
1729 |
5%|▌ | 45/851 [00:00<00:10, 73.71it/s]
|
1730 |
6%|▌ | 53/851 [00:00<00:10, 74.31it/s]
|
1731 |
7%|▋ | 62/851 [00:00<00:10, 75.34it/s]
|
1732 |
8%|▊ | 70/851 [00:00<00:11, 70.02it/s]
|
1733 |
9%|▉ | 78/851 [00:01<00:10, 71.64it/s]
|
1734 |
10%|█ | 86/851 [00:01<00:10, 71.26it/s]
|
1735 |
11%|█ | 94/851 [00:01<00:10, 69.42it/s]
|
1736 |
12%|█▏ | 102/851 [00:01<00:10, 70.77it/s]
|
1737 |
13%|█▎ | 110/851 [00:01<00:10, 69.03it/s]
|
1738 |
14%|█▍ | 118/851 [00:01<00:10, 70.01it/s]
|
1739 |
15%|█▍ | 126/851 [00:01<00:10, 67.17it/s]
|
1740 |
16%|█▌ | 133/851 [00:01<00:10, 67.65it/s]
|
1741 |
16%|█▋ | 140/851 [00:01<00:10, 67.39it/s]
|
1742 |
17%|█▋ | 148/851 [00:02<00:10, 66.64it/s]
|
1743 |
18%|█▊ | 156/851 [00:02<00:10, 68.76it/s]
|
1744 |
19%|█▉ | 164/851 [00:02<00:09, 69.74it/s]
|
1745 |
20%|██ | 172/851 [00:02<00:09, 71.66it/s]
|
1746 |
21%|██ | 180/851 [00:02<00:09, 72.19it/s]
|
1747 |
22%|██▏ | 188/851 [00:02<00:09, 72.67it/s]
|
1748 |
23%|██▎ | 196/851 [00:02<00:08, 73.59it/s]
|
1749 |
24%|██▍ | 204/851 [00:02<00:08, 75.18it/s]
|
1750 |
25%|██▍ | 212/851 [00:02<00:09, 70.71it/s]
|
1751 |
26%|██▌ | 220/851 [00:03<00:09, 68.53it/s]
|
1752 |
27%|██▋ | 228/851 [00:03<00:08, 69.85it/s]
|
1753 |
28%|██▊ | 236/851 [00:03<00:08, 69.98it/s]
|
1754 |
29%|██▊ | 244/851 [00:03<00:09, 66.03it/s]
|
1755 |
30%|██▉ | 252/851 [00:03<00:08, 69.13it/s]
|
1756 |
31%|███ | 261/851 [00:03<00:08, 72.34it/s]
|
1757 |
32%|███▏ | 269/851 [00:03<00:08, 71.00it/s]
|
1758 |
33%|███▎ | 277/851 [00:03<00:07, 73.30it/s]
|
1759 |
33%|███▎ | 285/851 [00:03<00:07, 74.56it/s]
|
1760 |
34%|███▍ | 293/851 [00:04<00:07, 72.43it/s]
|
1761 |
35%|███▌ | 301/851 [00:04<00:07, 73.31it/s]
|
1762 |
36%|███▋ | 310/851 [00:04<00:07, 75.84it/s]
|
1763 |
37%|███▋ | 318/851 [00:04<00:07, 71.54it/s]
|
1764 |
38%|███▊ | 326/851 [00:04<00:07, 71.08it/s]
|
1765 |
39%|███▉ | 334/851 [00:04<00:07, 71.61it/s]
|
1766 |
40%|████ | 342/851 [00:04<00:07, 72.39it/s]
|
1767 |
41%|████ | 350/851 [00:04<00:06, 71.99it/s]
|
1768 |
42%|████▏ | 358/851 [00:05<00:07, 67.28it/s]
|
1769 |
43%|████▎ | 366/851 [00:05<00:07, 67.95it/s]
|
1770 |
44%|████▍ | 373/851 [00:05<00:06, 68.31it/s]
|
1771 |
45%|████▍ | 380/851 [00:05<00:07, 66.41it/s]
|
1772 |
46%|████▌ | 388/851 [00:05<00:06, 69.17it/s]
|
1773 |
46%|████▋ | 395/851 [00:05<00:06, 69.34it/s]
|
1774 |
47%|████▋ | 402/851 [00:05<00:06, 69.06it/s]
|
1775 |
48%|████▊ | 409/851 [00:05<00:06, 65.01it/s]
|
1776 |
49%|████▉ | 417/851 [00:05<00:06, 67.26it/s]
|
1777 |
50%|████▉ | 425/851 [00:06<00:06, 66.34it/s]
|
1778 |
51%|█████ | 433/851 [00:06<00:06, 68.38it/s]
|
1779 |
52%|█████▏ | 441/851 [00:06<00:05, 70.44it/s]
|
1780 |
53%|█████▎ | 449/851 [00:06<00:05, 70.15it/s]
|
1781 |
54%|█████▎ | 457/851 [00:06<00:05, 71.36it/s]
|
1782 |
55%|█████▍ | 465/851 [00:06<00:05, 68.14it/s]
|
1783 |
55%|█████▌ | 472/851 [00:06<00:06, 61.77it/s]
|
1784 |
56%|█████▋ | 479/851 [00:06<00:05, 63.68it/s]
|
1785 |
57%|█████▋ | 486/851 [00:06<00:05, 62.95it/s]
|
1786 |
58%|█████▊ | 494/851 [00:07<00:05, 66.26it/s]
|
1787 |
59%|█████▉ | 502/851 [00:07<00:05, 69.52it/s]
|
1788 |
60%|█████▉ | 510/851 [00:07<00:04, 69.09it/s]
|
1789 |
61%|██████ | 518/851 [00:07<00:04, 70.28it/s]
|
1790 |
62%|██████▏ | 526/851 [00:07<00:04, 65.43it/s]
|
1791 |
63%|██████▎ | 534/851 [00:07<00:04, 67.57it/s]
|
1792 |
64%|██████▎ | 542/851 [00:07<00:04, 69.07it/s]
|
1793 |
65%|██████▍ | 549/851 [00:07<00:04, 66.40it/s]
|
1794 |
65%|██████▌ | 557/851 [00:07<00:04, 68.83it/s]
|
1795 |
67%|██████▋ | 566/851 [00:08<00:03, 72.25it/s]
|
1796 |
67%|██████▋ | 574/851 [00:08<00:03, 71.57it/s]
|
1797 |
68%|██████▊ | 582/851 [00:08<00:03, 69.26it/s]
|
1798 |
69%|██████▉ | 589/851 [00:08<00:03, 66.06it/s]
|
1799 |
70%|███████ | 596/851 [00:08<00:03, 67.01it/s]
|
1800 |
71%|███████ | 604/851 [00:08<00:03, 68.21it/s]
|
1801 |
72%|███████▏ | 611/851 [00:08<00:03, 67.43it/s]
|
1802 |
73%|███████▎ | 618/851 [00:08<00:03, 63.33it/s]
|
1803 |
74%|███████▎ | 626/851 [00:09<00:03, 64.04it/s]
|
1804 |
74%|███████▍ | 633/851 [00:09<00:03, 63.85it/s]
|
1805 |
75%|███████▌ | 640/851 [00:09<00:03, 64.00it/s]
|
1806 |
76%|███████▌ | 647/851 [00:09<00:03, 62.36it/s]
|
1807 |
77%|███████▋ | 655/851 [00:09<00:02, 65.58it/s]
|
1808 |
78%|███████▊ | 662/851 [00:09<00:02, 66.21it/s]
|
1809 |
79%|███████▊ | 669/851 [00:09<00:02, 67.09it/s]
|
1810 |
79%|███████▉ | 676/851 [00:09<00:02, 66.93it/s]
|
1811 |
80%|████████ | 683/851 [00:09<00:02, 67.45it/s]
|
1812 |
81%|████████ | 691/851 [00:09<00:02, 70.83it/s]
|
1813 |
82%|████████▏ | 699/851 [00:10<00:02, 70.91it/s]
|
1814 |
83%|████████▎ | 708/851 [00:10<00:01, 74.14it/s]
|
1815 |
84%|████████▍ | 716/851 [00:10<00:01, 71.95it/s]
|
1816 |
85%|████████▌ | 724/851 [00:10<00:01, 72.30it/s]
|
1817 |
86%|████████▌ | 732/851 [00:10<00:01, 73.78it/s]
|
1818 |
87%|████████▋ | 740/851 [00:10<00:01, 73.29it/s]
|
1819 |
88%|████████▊ | 748/851 [00:10<00:01, 72.54it/s]
|
1820 |
89%|████████▉ | 756/851 [00:10<00:01, 72.39it/s]
|
1821 |
90%|████████▉ | 764/851 [00:10<00:01, 73.41it/s]
|
1822 |
91%|█████████ | 772/851 [00:11<00:01, 68.50it/s]
|
1823 |
92%|█████████▏| 779/851 [00:11<00:01, 66.70it/s]
|
1824 |
92%|█████████▏| 786/851 [00:11<00:00, 67.37it/s]
|
1825 |
93%|█████████▎| 794/851 [00:11<00:00, 68.93it/s]
|
1826 |
94%|█████████▍| 802/851 [00:11<00:00, 71.12it/s]
|
1827 |
95%|█████████▌| 810/851 [00:11<00:00, 68.26it/s]
|
1828 |
96%|█████████▌| 818/851 [00:11<00:00, 69.10it/s]
|
1829 |
97%|█████████▋| 826/851 [00:11<00:00, 70.24it/s]
|
1830 |
98%|█████████▊| 834/851 [00:11<00:00, 69.55it/s]
|
1831 |
99%|█████████▉| 841/851 [00:12<00:00, 69.21it/s]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1613 |
[INFO|modeling_utils.py:2590] 2024-05-13 14:58:40,302 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
|
1614 |
[INFO|tokenization_utils_base.py:2488] 2024-05-13 14:58:40,303 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
|
1615 |
[INFO|tokenization_utils_base.py:2497] 2024-05-13 14:58:40,303 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
|
1616 |
+
{'eval_loss': 1.7860842943191528, 'eval_precision': 0.09270693512304251, 'eval_recall': 0.9522058823529411, 'eval_f1': 0.16896354888689555, 'eval_accuracy': 0.7845534874460183, 'eval_runtime': 15.9319, 'eval_samples_per_second': 427.256, 'eval_steps_per_second': 53.415, 'epoch': 10.0}
|
1617 |
+
{'train_runtime': 3337.712, 'train_samples_per_second': 81.565, 'train_steps_per_second': 5.096, 'train_loss': 0.08913132946046923, 'epoch': 10.0}
|
1618 |
+
***** train metrics *****
|
1619 |
+
epoch = 9.9971
|
1620 |
+
total_flos = 6240533GF
|
1621 |
+
train_loss = 0.0891
|
1622 |
+
train_runtime = 0:55:37.71
|
1623 |
+
train_samples = 27224
|
1624 |
+
train_samples_per_second = 81.565
|
1625 |
+
train_steps_per_second = 5.096
|
1626 |
+
05/13/2024 14:58:43 - INFO - __main__ - *** Evaluate ***
|
1627 |
+
[INFO|trainer.py:786] 2024-05-13 14:58:43,309 >> The following columns in the evaluation set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: ner_tags, tokens, id. If ner_tags, tokens, id are not expected by `RobertaForTokenClassification.forward`, you can safely ignore this message.
|
1628 |
+
[INFO|trainer.py:3614] 2024-05-13 14:58:43,312 >> ***** Running Evaluation *****
|
1629 |
+
[INFO|trainer.py:3616] 2024-05-13 14:58:43,312 >> Num examples = 6807
|
1630 |
+
[INFO|trainer.py:3619] 2024-05-13 14:58:43,312 >> Batch size = 8
|
1631 |
+
|
1632 |
0%| | 0/851 [00:00<?, ?it/s]
|
1633 |
1%| | 10/851 [00:00<00:09, 92.39it/s]
|
1634 |
2%|▏ | 20/851 [00:00<00:10, 78.68it/s]
|
1635 |
3%|▎ | 28/851 [00:00<00:10, 75.69it/s]
|
1636 |
4%|▍ | 36/851 [00:00<00:10, 74.36it/s]
|
1637 |
5%|▌ | 44/851 [00:00<00:10, 75.32it/s]
|
1638 |
6%|▌ | 52/851 [00:00<00:10, 75.93it/s]
|
1639 |
7%|▋ | 60/851 [00:00<00:10, 76.39it/s]
|
1640 |
8%|▊ | 68/851 [00:00<00:11, 70.86it/s]
|
1641 |
9%|▉ | 76/851 [00:01<00:11, 70.00it/s]
|
1642 |
10%|▉ | 84/851 [00:01<00:10, 70.49it/s]
|
1643 |
11%|█ | 92/851 [00:01<00:10, 70.55it/s]
|
1644 |
12%|█▏ | 100/851 [00:01<00:10, 70.25it/s]
|
1645 |
13%|█▎ | 108/851 [00:01<00:10, 69.63it/s]
|
1646 |
14%|█▎ | 115/851 [00:01<00:10, 69.25it/s]
|
1647 |
15%|█▍ | 124/851 [00:01<00:09, 73.89it/s]
|
1648 |
16%|█▌ | 132/851 [00:01<00:10, 66.17it/s]
|
1649 |
16%|█▋ | 139/851 [00:01<00:10, 66.18it/s]
|
1650 |
17%|█▋ | 147/851 [00:02<00:10, 67.87it/s]
|
1651 |
18%|█▊ | 154/851 [00:02<00:10, 67.52it/s]
|
1652 |
19%|█▉ | 162/851 [00:02<00:09, 69.68it/s]
|
1653 |
20%|█▉ | 170/851 [00:02<00:09, 71.19it/s]
|
1654 |
21%|██ | 178/851 [00:02<00:09, 73.27it/s]
|
1655 |
22%|██▏ | 186/851 [00:02<00:09, 73.02it/s]
|
1656 |
23%|██▎ | 194/851 [00:02<00:08, 74.19it/s]
|
1657 |
24%|██▍ | 203/851 [00:02<00:08, 76.36it/s]
|
1658 |
25%|██▍ | 211/851 [00:02<00:08, 71.75it/s]
|
1659 |
26%|██▌ | 219/851 [00:03<00:09, 69.67it/s]
|
1660 |
27%|██▋ | 227/851 [00:03<00:08, 71.38it/s]
|
1661 |
28%|██▊ | 235/851 [00:03<00:08, 72.05it/s]
|
1662 |
29%|██▊ | 243/851 [00:03<00:09, 67.00it/s]
|
1663 |
29%|██▉ | 251/851 [00:03<00:08, 69.80it/s]
|
1664 |
30%|███ | 259/851 [00:03<00:08, 72.22it/s]
|
1665 |
31%|███▏ | 267/851 [00:03<00:08, 72.74it/s]
|
1666 |
32%|███▏ | 275/851 [00:03<00:07, 73.62it/s]
|
1667 |
33%|███▎ | 283/851 [00:03<00:07, 74.61it/s]
|
1668 |
34%|███▍ | 291/851 [00:04<00:07, 72.54it/s]
|
1669 |
35%|███▌ | 299/851 [00:04<00:07, 74.26it/s]
|
1670 |
36%|███▌ | 307/851 [00:04<00:07, 75.74it/s]
|
1671 |
37%|███▋ | 315/851 [00:04<00:07, 70.27it/s]
|
1672 |
38%|███▊ | 323/851 [00:04<00:07, 72.85it/s]
|
1673 |
39%|███▉ | 331/851 [00:04<00:07, 72.36it/s]
|
1674 |
40%|███▉ | 339/851 [00:04<00:07, 71.89it/s]
|
1675 |
41%|████ | 347/851 [00:04<00:06, 72.58it/s]
|
1676 |
42%|████▏ | 355/851 [00:04<00:07, 69.70it/s]
|
1677 |
43%|████▎ | 363/851 [00:05<00:06, 69.83it/s]
|
1678 |
44%|████▎ | 371/851 [00:05<00:06, 69.06it/s]
|
1679 |
44%|████▍ | 378/851 [00:05<00:06, 69.30it/s]
|
1680 |
45%|████▌ | 385/851 [00:05<00:06, 68.06it/s]
|
1681 |
46%|████▌ | 393/851 [00:05<00:06, 69.52it/s]
|
1682 |
47%|████▋ | 401/851 [00:05<00:06, 69.90it/s]
|
1683 |
48%|████▊ | 408/851 [00:05<00:06, 64.81it/s]
|
1684 |
49%|████▉ | 416/851 [00:05<00:06, 67.34it/s]
|
1685 |
50%|████▉ | 424/851 [00:05<00:06, 69.53it/s]
|
1686 |
51%|█████ | 431/851 [00:06<00:06, 68.95it/s]
|
1687 |
52%|█████▏ | 439/851 [00:06<00:05, 71.11it/s]
|
1688 |
53%|█████▎ | 447/851 [00:06<00:05, 70.55it/s]
|
1689 |
53%|█████▎ | 455/851 [00:06<00:05, 71.50it/s]
|
1690 |
54%|█████▍ | 463/851 [00:06<00:05, 70.54it/s]
|
1691 |
55%|█████▌ | 471/851 [00:06<00:06, 61.54it/s]
|
1692 |
56%|█████▌ | 478/851 [00:06<00:05, 63.35it/s]
|
1693 |
57%|█████▋ | 485/851 [00:06<00:05, 63.12it/s]
|
1694 |
58%|█████▊ | 493/851 [00:06<00:05, 66.52it/s]
|
1695 |
59%|█████▉ | 502/851 [00:07<00:04, 70.10it/s]
|
1696 |
60%|█████▉ | 510/851 [00:07<00:04, 69.87it/s]
|
1697 |
61%|██████ | 518/851 [00:07<00:04, 71.30it/s]
|
1698 |
62%|██████▏ | 526/851 [00:07<00:04, 66.37it/s]
|
1699 |
63%|██████▎ | 534/851 [00:07<00:04, 68.15it/s]
|
1700 |
64%|██████▎ | 542/851 [00:07<00:04, 69.51it/s]
|
1701 |
65%|██████▍ | 550/851 [00:07<00:04, 67.52it/s]
|
1702 |
66%|██████▌ | 558/851 [00:07<00:04, 69.41it/s]
|
1703 |
67%|██████▋ | 566/851 [00:08<00:03, 72.24it/s]
|
1704 |
67%|██████▋ | 574/851 [00:08<00:03, 72.87it/s]
|
1705 |
68%|██████▊ | 582/851 [00:08<00:03, 69.95it/s]
|
1706 |
69%|██████▉ | 590/851 [00:08<00:03, 67.24it/s]
|
1707 |
70%|███████ | 597/851 [00:08<00:03, 67.77it/s]
|
1708 |
71%|███████ | 605/851 [00:08<00:03, 67.61it/s]
|
1709 |
72%|███████▏ | 612/851 [00:08<00:03, 66.37it/s]
|
1710 |
73%|███████▎ | 619/851 [00:08<00:03, 64.78it/s]
|
1711 |
74%|███████▎ | 626/851 [00:08<00:03, 64.41it/s]
|
1712 |
74%|███████▍ | 633/851 [00:09<00:03, 64.56it/s]
|
1713 |
75%|███████▌ | 640/851 [00:09<00:03, 64.37it/s]
|
1714 |
76%|███████▌ | 647/851 [00:09<00:03, 62.56it/s]
|
1715 |
77%|███████▋ | 655/851 [00:09<00:02, 65.73it/s]
|
1716 |
78%|███████▊ | 662/851 [00:09<00:02, 66.42it/s]
|
1717 |
79%|███████▊ | 670/851 [00:09<00:02, 67.98it/s]
|
1718 |
80%|███████▉ | 678/851 [00:09<00:02, 69.04it/s]
|
1719 |
81%|████████ | 686/851 [00:09<00:02, 69.89it/s]
|
1720 |
82%|████████▏ | 694/851 [00:09<00:02, 71.78it/s]
|
1721 |
82%|████████▏ | 702/851 [00:10<00:02, 72.47it/s]
|
1722 |
83%|████████▎ | 710/851 [00:10<00:01, 73.89it/s]
|
1723 |
84%|████████▍ | 718/851 [00:10<00:01, 70.91it/s]
|
1724 |
85%|████████▌ | 726/851 [00:10<00:01, 70.61it/s]
|
1725 |
86%|████████▋ | 734/851 [00:10<00:01, 72.17it/s]
|
1726 |
87%|████████▋ | 742/851 [00:10<00:01, 73.05it/s]
|
1727 |
88%|████████▊ | 750/851 [00:10<00:01, 72.59it/s]
|
1728 |
89%|████████▉ | 758/851 [00:10<00:01, 73.09it/s]
|
1729 |
90%|█████████ | 766/851 [00:10<00:01, 69.09it/s]
|
1730 |
91%|█████████ | 774/851 [00:11<00:01, 68.91it/s]
|
1731 |
92%|█████████▏| 781/851 [00:11<00:01, 66.57it/s]
|
1732 |
93%|█████████▎| 788/851 [00:11<00:00, 67.34it/s]
|
1733 |
94%|█████████▎| 796/851 [00:11<00:00, 69.12it/s]
|
1734 |
94%|█████████▍| 804/851 [00:11<00:00, 71.53it/s]
|
1735 |
95%|█████████▌| 812/851 [00:11<00:00, 69.03it/s]
|
1736 |
96%|█████████▋| 820/851 [00:11<00:00, 69.99it/s]
|
1737 |
97%|█████████▋| 828/851 [00:11<00:00, 71.40it/s]
|
1738 |
98%|█████████▊| 836/851 [00:11<00:00, 71.06it/s]
|
1739 |
99%|█████████▉| 844/851 [00:12<00:00, 67.57it/s]
|
1740 |
+
_warn_prf(average, modifier, msg_start, len(result))
|
1741 |
+
|
1742 |
+
***** eval metrics *****
|
1743 |
+
epoch = 9.9971
|
1744 |
+
eval_accuracy = 0.81
|
1745 |
+
eval_f1 = 0.1759
|
1746 |
+
eval_loss = 0.6631
|
1747 |
+
eval_precision = 0.0969
|
1748 |
+
eval_recall = 0.9485
|
1749 |
+
eval_runtime = 0:00:15.81
|
1750 |
+
eval_samples = 6807
|
1751 |
+
eval_samples_per_second = 430.395
|
1752 |
+
eval_steps_per_second = 53.807
|
1753 |
+
05/13/2024 14:58:59 - INFO - __main__ - *** Predict ***
|
1754 |
+
[INFO|trainer.py:786] 2024-05-13 14:58:59,135 >> The following columns in the test set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: ner_tags, tokens, id. If ner_tags, tokens, id are not expected by `RobertaForTokenClassification.forward`, you can safely ignore this message.
|
1755 |
+
[INFO|trainer.py:3614] 2024-05-13 14:58:59,137 >> ***** Running Prediction *****
|
1756 |
+
[INFO|trainer.py:3616] 2024-05-13 14:58:59,137 >> Num examples = 6807
|
1757 |
+
[INFO|trainer.py:3619] 2024-05-13 14:58:59,138 >> Batch size = 8
|
1758 |
+
|
1759 |
0%| | 0/851 [00:00<?, ?it/s]
|
1760 |
1%| | 10/851 [00:00<00:08, 94.43it/s]
|
1761 |
2%|▏ | 20/851 [00:00<00:10, 79.64it/s]
|
1762 |
3%|▎ | 29/851 [00:00<00:10, 75.61it/s]
|
1763 |
4%|▍ | 37/851 [00:00<00:11, 73.82it/s]
|
1764 |
5%|▌ | 45/851 [00:00<00:10, 73.71it/s]
|
1765 |
6%|▌ | 53/851 [00:00<00:10, 74.31it/s]
|
1766 |
7%|▋ | 62/851 [00:00<00:10, 75.34it/s]
|
1767 |
8%|▊ | 70/851 [00:00<00:11, 70.02it/s]
|
1768 |
9%|▉ | 78/851 [00:01<00:10, 71.64it/s]
|
1769 |
10%|█ | 86/851 [00:01<00:10, 71.26it/s]
|
1770 |
11%|█ | 94/851 [00:01<00:10, 69.42it/s]
|
1771 |
12%|█▏ | 102/851 [00:01<00:10, 70.77it/s]
|
1772 |
13%|█▎ | 110/851 [00:01<00:10, 69.03it/s]
|
1773 |
14%|█▍ | 118/851 [00:01<00:10, 70.01it/s]
|
1774 |
15%|█▍ | 126/851 [00:01<00:10, 67.17it/s]
|
1775 |
16%|█▌ | 133/851 [00:01<00:10, 67.65it/s]
|
1776 |
16%|█▋ | 140/851 [00:01<00:10, 67.39it/s]
|
1777 |
17%|█▋ | 148/851 [00:02<00:10, 66.64it/s]
|
1778 |
18%|█▊ | 156/851 [00:02<00:10, 68.76it/s]
|
1779 |
19%|█▉ | 164/851 [00:02<00:09, 69.74it/s]
|
1780 |
20%|██ | 172/851 [00:02<00:09, 71.66it/s]
|
1781 |
21%|██ | 180/851 [00:02<00:09, 72.19it/s]
|
1782 |
22%|██▏ | 188/851 [00:02<00:09, 72.67it/s]
|
1783 |
23%|██▎ | 196/851 [00:02<00:08, 73.59it/s]
|
1784 |
24%|██▍ | 204/851 [00:02<00:08, 75.18it/s]
|
1785 |
25%|██▍ | 212/851 [00:02<00:09, 70.71it/s]
|
1786 |
26%|██▌ | 220/851 [00:03<00:09, 68.53it/s]
|
1787 |
27%|██▋ | 228/851 [00:03<00:08, 69.85it/s]
|
1788 |
28%|██▊ | 236/851 [00:03<00:08, 69.98it/s]
|
1789 |
29%|██▊ | 244/851 [00:03<00:09, 66.03it/s]
|
1790 |
30%|██▉ | 252/851 [00:03<00:08, 69.13it/s]
|
1791 |
31%|███ | 261/851 [00:03<00:08, 72.34it/s]
|
1792 |
32%|███▏ | 269/851 [00:03<00:08, 71.00it/s]
|
1793 |
33%|███▎ | 277/851 [00:03<00:07, 73.30it/s]
|
1794 |
33%|███▎ | 285/851 [00:03<00:07, 74.56it/s]
|
1795 |
34%|███▍ | 293/851 [00:04<00:07, 72.43it/s]
|
1796 |
35%|███▌ | 301/851 [00:04<00:07, 73.31it/s]
|
1797 |
36%|███▋ | 310/851 [00:04<00:07, 75.84it/s]
|
1798 |
37%|███▋ | 318/851 [00:04<00:07, 71.54it/s]
|
1799 |
38%|███▊ | 326/851 [00:04<00:07, 71.08it/s]
|
1800 |
39%|███▉ | 334/851 [00:04<00:07, 71.61it/s]
|
1801 |
40%|████ | 342/851 [00:04<00:07, 72.39it/s]
|
1802 |
41%|████ | 350/851 [00:04<00:06, 71.99it/s]
|
1803 |
42%|████▏ | 358/851 [00:05<00:07, 67.28it/s]
|
1804 |
43%|████▎ | 366/851 [00:05<00:07, 67.95it/s]
|
1805 |
44%|████▍ | 373/851 [00:05<00:06, 68.31it/s]
|
1806 |
45%|████▍ | 380/851 [00:05<00:07, 66.41it/s]
|
1807 |
46%|████▌ | 388/851 [00:05<00:06, 69.17it/s]
|
1808 |
46%|████▋ | 395/851 [00:05<00:06, 69.34it/s]
|
1809 |
47%|████▋ | 402/851 [00:05<00:06, 69.06it/s]
|
1810 |
48%|████▊ | 409/851 [00:05<00:06, 65.01it/s]
|
1811 |
49%|████▉ | 417/851 [00:05<00:06, 67.26it/s]
|
1812 |
50%|████▉ | 425/851 [00:06<00:06, 66.34it/s]
|
1813 |
51%|█████ | 433/851 [00:06<00:06, 68.38it/s]
|
1814 |
52%|█████▏ | 441/851 [00:06<00:05, 70.44it/s]
|
1815 |
53%|█████▎ | 449/851 [00:06<00:05, 70.15it/s]
|
1816 |
54%|█████▎ | 457/851 [00:06<00:05, 71.36it/s]
|
1817 |
55%|█████▍ | 465/851 [00:06<00:05, 68.14it/s]
|
1818 |
55%|█████▌ | 472/851 [00:06<00:06, 61.77it/s]
|
1819 |
56%|█████▋ | 479/851 [00:06<00:05, 63.68it/s]
|
1820 |
57%|█████▋ | 486/851 [00:06<00:05, 62.95it/s]
|
1821 |
58%|█████▊ | 494/851 [00:07<00:05, 66.26it/s]
|
1822 |
59%|█████▉ | 502/851 [00:07<00:05, 69.52it/s]
|
1823 |
60%|█████▉ | 510/851 [00:07<00:04, 69.09it/s]
|
1824 |
61%|██████ | 518/851 [00:07<00:04, 70.28it/s]
|
1825 |
62%|██████▏ | 526/851 [00:07<00:04, 65.43it/s]
|
1826 |
63%|██████▎ | 534/851 [00:07<00:04, 67.57it/s]
|
1827 |
64%|██████▎ | 542/851 [00:07<00:04, 69.07it/s]
|
1828 |
65%|██████▍ | 549/851 [00:07<00:04, 66.40it/s]
|
1829 |
65%|██████▌ | 557/851 [00:07<00:04, 68.83it/s]
|
1830 |
67%|██████▋ | 566/851 [00:08<00:03, 72.25it/s]
|
1831 |
67%|██████▋ | 574/851 [00:08<00:03, 71.57it/s]
|
1832 |
68%|██████▊ | 582/851 [00:08<00:03, 69.26it/s]
|
1833 |
69%|██████▉ | 589/851 [00:08<00:03, 66.06it/s]
|
1834 |
70%|███████ | 596/851 [00:08<00:03, 67.01it/s]
|
1835 |
71%|███████ | 604/851 [00:08<00:03, 68.21it/s]
|
1836 |
72%|███████▏ | 611/851 [00:08<00:03, 67.43it/s]
|
1837 |
73%|███████▎ | 618/851 [00:08<00:03, 63.33it/s]
|
1838 |
74%|███████▎ | 626/851 [00:09<00:03, 64.04it/s]
|
1839 |
74%|███████▍ | 633/851 [00:09<00:03, 63.85it/s]
|
1840 |
75%|███████▌ | 640/851 [00:09<00:03, 64.00it/s]
|
1841 |
76%|███████▌ | 647/851 [00:09<00:03, 62.36it/s]
|
1842 |
77%|███████▋ | 655/851 [00:09<00:02, 65.58it/s]
|
1843 |
78%|███████▊ | 662/851 [00:09<00:02, 66.21it/s]
|
1844 |
79%|███████▊ | 669/851 [00:09<00:02, 67.09it/s]
|
1845 |
79%|███████▉ | 676/851 [00:09<00:02, 66.93it/s]
|
1846 |
80%|████████ | 683/851 [00:09<00:02, 67.45it/s]
|
1847 |
81%|████████ | 691/851 [00:09<00:02, 70.83it/s]
|
1848 |
82%|████████▏ | 699/851 [00:10<00:02, 70.91it/s]
|
1849 |
83%|████████▎ | 708/851 [00:10<00:01, 74.14it/s]
|
1850 |
84%|████████▍ | 716/851 [00:10<00:01, 71.95it/s]
|
1851 |
85%|████████▌ | 724/851 [00:10<00:01, 72.30it/s]
|
1852 |
86%|████████▌ | 732/851 [00:10<00:01, 73.78it/s]
|
1853 |
87%|████████▋ | 740/851 [00:10<00:01, 73.29it/s]
|
1854 |
88%|████████▊ | 748/851 [00:10<00:01, 72.54it/s]
|
1855 |
89%|████████▉ | 756/851 [00:10<00:01, 72.39it/s]
|
1856 |
90%|████████▉ | 764/851 [00:10<00:01, 73.41it/s]
|
1857 |
91%|█████████ | 772/851 [00:11<00:01, 68.50it/s]
|
1858 |
92%|█████████▏| 779/851 [00:11<00:01, 66.70it/s]
|
1859 |
92%|█████████▏| 786/851 [00:11<00:00, 67.37it/s]
|
1860 |
93%|█████████▎| 794/851 [00:11<00:00, 68.93it/s]
|
1861 |
94%|█████████▍| 802/851 [00:11<00:00, 71.12it/s]
|
1862 |
95%|█████████▌| 810/851 [00:11<00:00, 68.26it/s]
|
1863 |
96%|█████████▌| 818/851 [00:11<00:00, 69.10it/s]
|
1864 |
97%|█████████▋| 826/851 [00:11<00:00, 70.24it/s]
|
1865 |
98%|█████████▊| 834/851 [00:11<00:00, 69.55it/s]
|
1866 |
99%|█████████▉| 841/851 [00:12<00:00, 69.21it/s]
|
1867 |
+
[INFO|trainer.py:3305] 2024-05-13 14:59:15,354 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
|
1868 |
+
[INFO|configuration_utils.py:471] 2024-05-13 14:59:15,355 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
|
1869 |
+
[INFO|modeling_utils.py:2590] 2024-05-13 14:59:16,582 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
|
1870 |
+
[INFO|tokenization_utils_base.py:2488] 2024-05-13 14:59:16,583 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
|
1871 |
+
[INFO|tokenization_utils_base.py:2497] 2024-05-13 14:59:16,584 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
|
1872 |
+
***** predict metrics *****
|
1873 |
+
predict_accuracy = 0.81
|
1874 |
+
predict_f1 = 0.1759
|
1875 |
+
predict_loss = 0.6631
|
1876 |
+
predict_precision = 0.0969
|
1877 |
+
predict_recall = 0.9485
|
1878 |
+
predict_runtime = 0:00:15.89
|
1879 |
+
predict_samples_per_second = 428.156
|
1880 |
+
predict_steps_per_second = 53.527
|
1881 |
+
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.997061416397296,
|
3 |
+
"total_flos": 6700722040732752.0,
|
4 |
+
"train_loss": 0.08913132946046923,
|
5 |
+
"train_runtime": 3337.712,
|
6 |
+
"train_samples": 27224,
|
7 |
+
"train_samples_per_second": 81.565,
|
8 |
+
"train_steps_per_second": 5.096
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,388 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.17586912065439672,
|
3 |
+
"best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-3403",
|
4 |
+
"epoch": 9.997061416397296,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 17010,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.29385836027034967,
|
13 |
+
"grad_norm": 1.9668159484863281,
|
14 |
+
"learning_rate": 4.853027630805409e-05,
|
15 |
+
"loss": 0.4174,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.5877167205406993,
|
20 |
+
"grad_norm": 3.246731758117676,
|
21 |
+
"learning_rate": 4.7060552616108174e-05,
|
22 |
+
"loss": 0.2765,
|
23 |
+
"step": 1000
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.8815750808110491,
|
27 |
+
"grad_norm": 2.936720609664917,
|
28 |
+
"learning_rate": 4.559082892416226e-05,
|
29 |
+
"loss": 0.2596,
|
30 |
+
"step": 1500
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.9997061416397296,
|
34 |
+
"eval_accuracy": 0.7671626010120082,
|
35 |
+
"eval_f1": 0.14622946114658822,
|
36 |
+
"eval_loss": 0.7913026213645935,
|
37 |
+
"eval_precision": 0.0793057825511713,
|
38 |
+
"eval_recall": 0.9365808823529411,
|
39 |
+
"eval_runtime": 16.052,
|
40 |
+
"eval_samples_per_second": 424.06,
|
41 |
+
"eval_steps_per_second": 53.015,
|
42 |
+
"step": 1701
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 1.1754334410813987,
|
46 |
+
"grad_norm": 2.302980661392212,
|
47 |
+
"learning_rate": 4.4121105232216346e-05,
|
48 |
+
"loss": 0.2135,
|
49 |
+
"step": 2000
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 1.4692918013517484,
|
53 |
+
"grad_norm": 3.7150967121124268,
|
54 |
+
"learning_rate": 4.265138154027043e-05,
|
55 |
+
"loss": 0.1839,
|
56 |
+
"step": 2500
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 1.7631501616220981,
|
60 |
+
"grad_norm": 1.2019191980361938,
|
61 |
+
"learning_rate": 4.118165784832452e-05,
|
62 |
+
"loss": 0.1853,
|
63 |
+
"step": 3000
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 2.0,
|
67 |
+
"eval_accuracy": 0.8099635429897495,
|
68 |
+
"eval_f1": 0.17586912065439672,
|
69 |
+
"eval_loss": 0.6631014347076416,
|
70 |
+
"eval_precision": 0.09691960931630353,
|
71 |
+
"eval_recall": 0.9485294117647058,
|
72 |
+
"eval_runtime": 15.875,
|
73 |
+
"eval_samples_per_second": 428.788,
|
74 |
+
"eval_steps_per_second": 53.606,
|
75 |
+
"step": 3403
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 2.0570085218924477,
|
79 |
+
"grad_norm": 2.090308904647827,
|
80 |
+
"learning_rate": 3.971193415637861e-05,
|
81 |
+
"loss": 0.1684,
|
82 |
+
"step": 3500
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 2.3508668821627974,
|
86 |
+
"grad_norm": 0.9995286464691162,
|
87 |
+
"learning_rate": 3.824221046443269e-05,
|
88 |
+
"loss": 0.1222,
|
89 |
+
"step": 4000
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 2.644725242433147,
|
93 |
+
"grad_norm": 1.3526027202606201,
|
94 |
+
"learning_rate": 3.677248677248677e-05,
|
95 |
+
"loss": 0.1277,
|
96 |
+
"step": 4500
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 2.938583602703497,
|
100 |
+
"grad_norm": 2.2816500663757324,
|
101 |
+
"learning_rate": 3.530276308054086e-05,
|
102 |
+
"loss": 0.1254,
|
103 |
+
"step": 5000
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 2.9997061416397295,
|
107 |
+
"eval_accuracy": 0.7754975935626944,
|
108 |
+
"eval_f1": 0.16528259292106748,
|
109 |
+
"eval_loss": 1.072888731956482,
|
110 |
+
"eval_precision": 0.0905877154220062,
|
111 |
+
"eval_recall": 0.9420955882352942,
|
112 |
+
"eval_runtime": 15.8836,
|
113 |
+
"eval_samples_per_second": 428.556,
|
114 |
+
"eval_steps_per_second": 53.577,
|
115 |
+
"step": 5104
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 3.2324419629738466,
|
119 |
+
"grad_norm": 0.6440290808677673,
|
120 |
+
"learning_rate": 3.3833039388594945e-05,
|
121 |
+
"loss": 0.0929,
|
122 |
+
"step": 5500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 3.5263003232441963,
|
126 |
+
"grad_norm": 3.7906153202056885,
|
127 |
+
"learning_rate": 3.2363315696649034e-05,
|
128 |
+
"loss": 0.0896,
|
129 |
+
"step": 6000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 3.820158683514546,
|
133 |
+
"grad_norm": 1.0953032970428467,
|
134 |
+
"learning_rate": 3.0893592004703116e-05,
|
135 |
+
"loss": 0.0823,
|
136 |
+
"step": 6500
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 4.0,
|
140 |
+
"eval_accuracy": 0.7719411469883488,
|
141 |
+
"eval_f1": 0.16243814311523055,
|
142 |
+
"eval_loss": 1.2567578554153442,
|
143 |
+
"eval_precision": 0.08880872627329726,
|
144 |
+
"eval_recall": 0.9503676470588235,
|
145 |
+
"eval_runtime": 15.8976,
|
146 |
+
"eval_samples_per_second": 428.179,
|
147 |
+
"eval_steps_per_second": 53.53,
|
148 |
+
"step": 6806
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 4.114017043784895,
|
152 |
+
"grad_norm": 1.7823286056518555,
|
153 |
+
"learning_rate": 2.9423868312757202e-05,
|
154 |
+
"loss": 0.0761,
|
155 |
+
"step": 7000
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 4.407875404055245,
|
159 |
+
"grad_norm": 1.165720820426941,
|
160 |
+
"learning_rate": 2.795414462081129e-05,
|
161 |
+
"loss": 0.0603,
|
162 |
+
"step": 7500
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 4.701733764325595,
|
166 |
+
"grad_norm": 4.1130452156066895,
|
167 |
+
"learning_rate": 2.648442092886537e-05,
|
168 |
+
"loss": 0.0589,
|
169 |
+
"step": 8000
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 4.9955921245959445,
|
173 |
+
"grad_norm": 0.3957385718822479,
|
174 |
+
"learning_rate": 2.501469723691946e-05,
|
175 |
+
"loss": 0.0597,
|
176 |
+
"step": 8500
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 4.99970614163973,
|
180 |
+
"eval_accuracy": 0.7836540772119656,
|
181 |
+
"eval_f1": 0.17101181993461315,
|
182 |
+
"eval_loss": 1.1907650232315063,
|
183 |
+
"eval_precision": 0.0940872613227562,
|
184 |
+
"eval_recall": 0.9375,
|
185 |
+
"eval_runtime": 16.019,
|
186 |
+
"eval_samples_per_second": 424.932,
|
187 |
+
"eval_steps_per_second": 53.124,
|
188 |
+
"step": 8507
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 5.289450484866294,
|
192 |
+
"grad_norm": 1.0087623596191406,
|
193 |
+
"learning_rate": 2.3544973544973546e-05,
|
194 |
+
"loss": 0.0423,
|
195 |
+
"step": 9000
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 5.583308845136644,
|
199 |
+
"grad_norm": 2.162200450897217,
|
200 |
+
"learning_rate": 2.2075249853027632e-05,
|
201 |
+
"loss": 0.043,
|
202 |
+
"step": 9500
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 5.877167205406994,
|
206 |
+
"grad_norm": 1.1820895671844482,
|
207 |
+
"learning_rate": 2.0605526161081718e-05,
|
208 |
+
"loss": 0.0446,
|
209 |
+
"step": 10000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 6.0,
|
213 |
+
"eval_accuracy": 0.7811686840461102,
|
214 |
+
"eval_f1": 0.1718036055495555,
|
215 |
+
"eval_loss": 1.384422779083252,
|
216 |
+
"eval_precision": 0.09443784820531555,
|
217 |
+
"eval_recall": 0.9503676470588235,
|
218 |
+
"eval_runtime": 15.9006,
|
219 |
+
"eval_samples_per_second": 428.098,
|
220 |
+
"eval_steps_per_second": 53.52,
|
221 |
+
"step": 10209
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 6.171025565677343,
|
225 |
+
"grad_norm": 2.190476894378662,
|
226 |
+
"learning_rate": 1.91358024691358e-05,
|
227 |
+
"loss": 0.0333,
|
228 |
+
"step": 10500
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 6.464883925947693,
|
232 |
+
"grad_norm": 1.5180469751358032,
|
233 |
+
"learning_rate": 1.766607877718989e-05,
|
234 |
+
"loss": 0.0333,
|
235 |
+
"step": 11000
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 6.758742286218043,
|
239 |
+
"grad_norm": 1.4279770851135254,
|
240 |
+
"learning_rate": 1.6196355085243976e-05,
|
241 |
+
"loss": 0.0325,
|
242 |
+
"step": 11500
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 6.99970614163973,
|
246 |
+
"eval_accuracy": 0.7866406684471785,
|
247 |
+
"eval_f1": 0.1704836709384043,
|
248 |
+
"eval_loss": 1.5515447854995728,
|
249 |
+
"eval_precision": 0.09366766603070772,
|
250 |
+
"eval_recall": 0.9476102941176471,
|
251 |
+
"eval_runtime": 15.8267,
|
252 |
+
"eval_samples_per_second": 430.095,
|
253 |
+
"eval_steps_per_second": 53.77,
|
254 |
+
"step": 11910
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 7.052600646488393,
|
258 |
+
"grad_norm": 0.6558970212936401,
|
259 |
+
"learning_rate": 1.472663139329806e-05,
|
260 |
+
"loss": 0.0289,
|
261 |
+
"step": 12000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 7.346459006758742,
|
265 |
+
"grad_norm": 0.26189878582954407,
|
266 |
+
"learning_rate": 1.3256907701352148e-05,
|
267 |
+
"loss": 0.0224,
|
268 |
+
"step": 12500
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 7.640317367029092,
|
272 |
+
"grad_norm": 1.370686650276184,
|
273 |
+
"learning_rate": 1.1787184009406232e-05,
|
274 |
+
"loss": 0.0231,
|
275 |
+
"step": 13000
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 7.934175727299442,
|
279 |
+
"grad_norm": 0.36619672179222107,
|
280 |
+
"learning_rate": 1.0317460317460318e-05,
|
281 |
+
"loss": 0.022,
|
282 |
+
"step": 13500
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 8.0,
|
286 |
+
"eval_accuracy": 0.7842582611859856,
|
287 |
+
"eval_f1": 0.1688722903304376,
|
288 |
+
"eval_loss": 1.6299601793289185,
|
289 |
+
"eval_precision": 0.09261733012734882,
|
290 |
+
"eval_recall": 0.9558823529411765,
|
291 |
+
"eval_runtime": 15.9057,
|
292 |
+
"eval_samples_per_second": 427.961,
|
293 |
+
"eval_steps_per_second": 53.503,
|
294 |
+
"step": 13612
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 8.22803408756979,
|
298 |
+
"grad_norm": 1.0557399988174438,
|
299 |
+
"learning_rate": 8.847736625514404e-06,
|
300 |
+
"loss": 0.0164,
|
301 |
+
"step": 14000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 8.521892447840141,
|
305 |
+
"grad_norm": 4.32920503616333,
|
306 |
+
"learning_rate": 7.37801293356849e-06,
|
307 |
+
"loss": 0.0164,
|
308 |
+
"step": 14500
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 8.81575080811049,
|
312 |
+
"grad_norm": 0.1674884408712387,
|
313 |
+
"learning_rate": 5.908289241622575e-06,
|
314 |
+
"loss": 0.017,
|
315 |
+
"step": 15000
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 8.999706141639729,
|
319 |
+
"eval_accuracy": 0.7844848301762433,
|
320 |
+
"eval_f1": 0.16934759532946844,
|
321 |
+
"eval_loss": 1.7459304332733154,
|
322 |
+
"eval_precision": 0.09292947396720136,
|
323 |
+
"eval_recall": 0.953125,
|
324 |
+
"eval_runtime": 16.0915,
|
325 |
+
"eval_samples_per_second": 423.018,
|
326 |
+
"eval_steps_per_second": 52.885,
|
327 |
+
"step": 15313
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 9.10960916838084,
|
331 |
+
"grad_norm": 3.3734261989593506,
|
332 |
+
"learning_rate": 4.438565549676661e-06,
|
333 |
+
"loss": 0.0171,
|
334 |
+
"step": 15500
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 9.40346752865119,
|
338 |
+
"grad_norm": 0.19828377664089203,
|
339 |
+
"learning_rate": 2.9688418577307467e-06,
|
340 |
+
"loss": 0.013,
|
341 |
+
"step": 16000
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 9.69732588892154,
|
345 |
+
"grad_norm": 1.491190791130066,
|
346 |
+
"learning_rate": 1.4991181657848325e-06,
|
347 |
+
"loss": 0.0133,
|
348 |
+
"step": 16500
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 9.991184249191889,
|
352 |
+
"grad_norm": 2.5641109943389893,
|
353 |
+
"learning_rate": 2.9394473838918286e-08,
|
354 |
+
"loss": 0.0135,
|
355 |
+
"step": 17000
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 9.997061416397296,
|
359 |
+
"eval_accuracy": 0.7845534874460183,
|
360 |
+
"eval_f1": 0.16896354888689555,
|
361 |
+
"eval_loss": 1.7860842943191528,
|
362 |
+
"eval_precision": 0.09270693512304251,
|
363 |
+
"eval_recall": 0.9522058823529411,
|
364 |
+
"eval_runtime": 15.9319,
|
365 |
+
"eval_samples_per_second": 427.256,
|
366 |
+
"eval_steps_per_second": 53.415,
|
367 |
+
"step": 17010
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 9.997061416397296,
|
371 |
+
"step": 17010,
|
372 |
+
"total_flos": 6700722040732752.0,
|
373 |
+
"train_loss": 0.08913132946046923,
|
374 |
+
"train_runtime": 3337.712,
|
375 |
+
"train_samples_per_second": 81.565,
|
376 |
+
"train_steps_per_second": 5.096
|
377 |
+
}
|
378 |
+
],
|
379 |
+
"logging_steps": 500,
|
380 |
+
"max_steps": 17010,
|
381 |
+
"num_input_tokens_seen": 0,
|
382 |
+
"num_train_epochs": 10,
|
383 |
+
"save_steps": 500,
|
384 |
+
"total_flos": 6700722040732752.0,
|
385 |
+
"train_batch_size": 4,
|
386 |
+
"trial_name": null,
|
387 |
+
"trial_params": null
|
388 |
+
}
|