Commit
•
25c63a6
1
Parent(s):
a0e178e
🍻 cheers
Browse files- README.md +10 -6
- all_results.json +16 -0
- config.json +1 -1
- eval_results.json +12 -0
- train_results.json +8 -0
- trainer_state.json +1775 -0
README.md
CHANGED
@@ -1,7 +1,11 @@
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: facebook/dinov2-base
|
4 |
tags:
|
|
|
|
|
5 |
- generated_from_trainer
|
6 |
metrics:
|
7 |
- accuracy
|
@@ -15,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# dino-base-2023_11_27-with_custom_head
|
17 |
|
18 |
-
This model is a fine-tuned version of [facebook/dinov2-base](https://huggingface.co/facebook/dinov2-base) on the
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
-
- F1 Micro: 0.
|
22 |
-
- F1 Macro: 0.
|
23 |
-
- Roc Auc: 0.
|
24 |
-
- Accuracy: 0.
|
25 |
- Learning Rate: 0.0000
|
26 |
|
27 |
## Model description
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- eng
|
4 |
license: apache-2.0
|
5 |
base_model: facebook/dinov2-base
|
6 |
tags:
|
7 |
+
- multilabel-image-classification
|
8 |
+
- multilabel
|
9 |
- generated_from_trainer
|
10 |
metrics:
|
11 |
- accuracy
|
|
|
19 |
|
20 |
# dino-base-2023_11_27-with_custom_head
|
21 |
|
22 |
+
This model is a fine-tuned version of [facebook/dinov2-base](https://huggingface.co/facebook/dinov2-base) on the multilabel_complete_dataset dataset.
|
23 |
It achieves the following results on the evaluation set:
|
24 |
+
- Loss: 0.1028
|
25 |
+
- F1 Micro: 0.8515
|
26 |
+
- F1 Macro: 0.8179
|
27 |
+
- Roc Auc: 0.9057
|
28 |
+
- Accuracy: 0.5716
|
29 |
- Learning Rate: 0.0000
|
30 |
|
31 |
## Model description
|
all_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 90.0,
|
3 |
+
"eval_accuracy": 0.5716297078493489,
|
4 |
+
"eval_f1_macro": 0.8179000572616444,
|
5 |
+
"eval_f1_micro": 0.8515016354445436,
|
6 |
+
"eval_loss": 0.1028159111738205,
|
7 |
+
"eval_roc_auc": 0.9056864620607311,
|
8 |
+
"eval_runtime": 666.1885,
|
9 |
+
"eval_samples_per_second": 4.265,
|
10 |
+
"eval_steps_per_second": 0.267,
|
11 |
+
"learning_rate": 1.0000000000000002e-06,
|
12 |
+
"train_loss": 0.13765035268679188,
|
13 |
+
"train_runtime": 238196.7751,
|
14 |
+
"train_samples_per_second": 3.237,
|
15 |
+
"train_steps_per_second": 0.203
|
16 |
+
}
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "facebook/dinov2-
|
3 |
"apply_layernorm": true,
|
4 |
"architectures": [
|
5 |
"NewheadDinov2ForImageClassification"
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "facebook/dinov2-base2023_11_27",
|
3 |
"apply_layernorm": true,
|
4 |
"architectures": [
|
5 |
"NewheadDinov2ForImageClassification"
|
eval_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 90.0,
|
3 |
+
"eval_accuracy": 0.5716297078493489,
|
4 |
+
"eval_f1_macro": 0.8179000572616444,
|
5 |
+
"eval_f1_micro": 0.8515016354445436,
|
6 |
+
"eval_loss": 0.1028159111738205,
|
7 |
+
"eval_roc_auc": 0.9056864620607311,
|
8 |
+
"eval_runtime": 666.1885,
|
9 |
+
"eval_samples_per_second": 4.265,
|
10 |
+
"eval_steps_per_second": 0.267,
|
11 |
+
"learning_rate": 1.0000000000000002e-06
|
12 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 90.0,
|
3 |
+
"learning_rate": 1.0000000000000002e-06,
|
4 |
+
"train_loss": 0.13765035268679188,
|
5 |
+
"train_runtime": 238196.7751,
|
6 |
+
"train_samples_per_second": 3.237,
|
7 |
+
"train_steps_per_second": 0.203
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1775 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.1039666160941124,
|
3 |
+
"best_model_checkpoint": "./dino-base-2023_11_27-with_custom_head/checkpoint-43952",
|
4 |
+
"epoch": 90.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 48240,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.93,
|
13 |
+
"learning_rate": 0.01,
|
14 |
+
"loss": 0.2471,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 1.0,
|
19 |
+
"eval_accuracy": 0.39442658092175775,
|
20 |
+
"eval_f1_macro": 0.40064106475011957,
|
21 |
+
"eval_f1_micro": 0.5684482898035094,
|
22 |
+
"eval_loss": 0.2136440873146057,
|
23 |
+
"eval_roc_auc": 0.7020299868465498,
|
24 |
+
"eval_runtime": 648.3221,
|
25 |
+
"eval_samples_per_second": 4.317,
|
26 |
+
"eval_steps_per_second": 0.27,
|
27 |
+
"learning_rate": 0.01,
|
28 |
+
"step": 536
|
29 |
+
},
|
30 |
+
{
|
31 |
+
"epoch": 1.87,
|
32 |
+
"learning_rate": 0.01,
|
33 |
+
"loss": 0.2208,
|
34 |
+
"step": 1000
|
35 |
+
},
|
36 |
+
{
|
37 |
+
"epoch": 2.0,
|
38 |
+
"eval_accuracy": 0.4090746695248303,
|
39 |
+
"eval_f1_macro": 0.646218183693934,
|
40 |
+
"eval_f1_micro": 0.6908665105386417,
|
41 |
+
"eval_loss": 0.21994435787200928,
|
42 |
+
"eval_roc_auc": 0.7944947312416977,
|
43 |
+
"eval_runtime": 649.3168,
|
44 |
+
"eval_samples_per_second": 4.311,
|
45 |
+
"eval_steps_per_second": 0.27,
|
46 |
+
"learning_rate": 0.01,
|
47 |
+
"step": 1072
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 2.8,
|
51 |
+
"learning_rate": 0.01,
|
52 |
+
"loss": 0.2181,
|
53 |
+
"step": 1500
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 3.0,
|
57 |
+
"eval_accuracy": 0.43122543765630583,
|
58 |
+
"eval_f1_macro": 0.5770940809584216,
|
59 |
+
"eval_f1_micro": 0.6712328767123288,
|
60 |
+
"eval_loss": 0.19793672859668732,
|
61 |
+
"eval_roc_auc": 0.7666801441240528,
|
62 |
+
"eval_runtime": 645.3042,
|
63 |
+
"eval_samples_per_second": 4.337,
|
64 |
+
"eval_steps_per_second": 0.271,
|
65 |
+
"learning_rate": 0.01,
|
66 |
+
"step": 1608
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 3.73,
|
70 |
+
"learning_rate": 0.01,
|
71 |
+
"loss": 0.2187,
|
72 |
+
"step": 2000
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 4.0,
|
76 |
+
"eval_accuracy": 0.4137191854233655,
|
77 |
+
"eval_f1_macro": 0.606113442290164,
|
78 |
+
"eval_f1_micro": 0.7318466985527052,
|
79 |
+
"eval_loss": 0.1766517609357834,
|
80 |
+
"eval_roc_auc": 0.8268705246501159,
|
81 |
+
"eval_runtime": 668.3118,
|
82 |
+
"eval_samples_per_second": 4.188,
|
83 |
+
"eval_steps_per_second": 0.262,
|
84 |
+
"learning_rate": 0.01,
|
85 |
+
"step": 2144
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 4.66,
|
89 |
+
"learning_rate": 0.01,
|
90 |
+
"loss": 0.2128,
|
91 |
+
"step": 2500
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 5.0,
|
95 |
+
"eval_accuracy": 0.43265451947123973,
|
96 |
+
"eval_f1_macro": 0.5526692766303807,
|
97 |
+
"eval_f1_micro": 0.7173007124613524,
|
98 |
+
"eval_loss": 0.17600227892398834,
|
99 |
+
"eval_roc_auc": 0.8011887560107374,
|
100 |
+
"eval_runtime": 647.0188,
|
101 |
+
"eval_samples_per_second": 4.326,
|
102 |
+
"eval_steps_per_second": 0.27,
|
103 |
+
"learning_rate": 0.01,
|
104 |
+
"step": 2680
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 5.6,
|
108 |
+
"learning_rate": 0.01,
|
109 |
+
"loss": 0.2171,
|
110 |
+
"step": 3000
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 6.0,
|
114 |
+
"eval_accuracy": 0.4226509467667024,
|
115 |
+
"eval_f1_macro": 0.5183527283365527,
|
116 |
+
"eval_f1_micro": 0.7108839135498899,
|
117 |
+
"eval_loss": 0.1992170512676239,
|
118 |
+
"eval_roc_auc": 0.8061565099276957,
|
119 |
+
"eval_runtime": 650.6673,
|
120 |
+
"eval_samples_per_second": 4.302,
|
121 |
+
"eval_steps_per_second": 0.269,
|
122 |
+
"learning_rate": 0.01,
|
123 |
+
"step": 3216
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 6.53,
|
127 |
+
"learning_rate": 0.01,
|
128 |
+
"loss": 0.2108,
|
129 |
+
"step": 3500
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 7.0,
|
133 |
+
"eval_accuracy": 0.41729188996070027,
|
134 |
+
"eval_f1_macro": 0.6054999995580052,
|
135 |
+
"eval_f1_micro": 0.7338740337092892,
|
136 |
+
"eval_loss": 0.16948619484901428,
|
137 |
+
"eval_roc_auc": 0.8232094683978447,
|
138 |
+
"eval_runtime": 649.3312,
|
139 |
+
"eval_samples_per_second": 4.311,
|
140 |
+
"eval_steps_per_second": 0.27,
|
141 |
+
"learning_rate": 0.01,
|
142 |
+
"step": 3752
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 7.46,
|
146 |
+
"learning_rate": 0.01,
|
147 |
+
"loss": 0.2147,
|
148 |
+
"step": 4000
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 8.0,
|
152 |
+
"eval_accuracy": 0.4226509467667024,
|
153 |
+
"eval_f1_macro": 0.6176021212641017,
|
154 |
+
"eval_f1_micro": 0.7440822902466228,
|
155 |
+
"eval_loss": 0.16185873746871948,
|
156 |
+
"eval_roc_auc": 0.8341874074405569,
|
157 |
+
"eval_runtime": 663.7112,
|
158 |
+
"eval_samples_per_second": 4.217,
|
159 |
+
"eval_steps_per_second": 0.264,
|
160 |
+
"learning_rate": 0.01,
|
161 |
+
"step": 4288
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 8.4,
|
165 |
+
"learning_rate": 0.01,
|
166 |
+
"loss": 0.2112,
|
167 |
+
"step": 4500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 9.0,
|
171 |
+
"eval_accuracy": 0.43086816720257237,
|
172 |
+
"eval_f1_macro": 0.6402287128969387,
|
173 |
+
"eval_f1_micro": 0.7336987336987336,
|
174 |
+
"eval_loss": 0.17079614102840424,
|
175 |
+
"eval_roc_auc": 0.8244850070361233,
|
176 |
+
"eval_runtime": 656.7931,
|
177 |
+
"eval_samples_per_second": 4.262,
|
178 |
+
"eval_steps_per_second": 0.266,
|
179 |
+
"learning_rate": 0.01,
|
180 |
+
"step": 4824
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 9.33,
|
184 |
+
"learning_rate": 0.01,
|
185 |
+
"loss": 0.216,
|
186 |
+
"step": 5000
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 10.0,
|
190 |
+
"eval_accuracy": 0.42836727402643804,
|
191 |
+
"eval_f1_macro": 0.643393584513474,
|
192 |
+
"eval_f1_micro": 0.7493341591824095,
|
193 |
+
"eval_loss": 0.1750514954328537,
|
194 |
+
"eval_roc_auc": 0.83718619371997,
|
195 |
+
"eval_runtime": 657.1366,
|
196 |
+
"eval_samples_per_second": 4.259,
|
197 |
+
"eval_steps_per_second": 0.266,
|
198 |
+
"learning_rate": 0.01,
|
199 |
+
"step": 5360
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 10.26,
|
203 |
+
"learning_rate": 0.01,
|
204 |
+
"loss": 0.2151,
|
205 |
+
"step": 5500
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 11.0,
|
209 |
+
"eval_accuracy": 0.43372633083244017,
|
210 |
+
"eval_f1_macro": 0.6292914445814428,
|
211 |
+
"eval_f1_micro": 0.7241767706883986,
|
212 |
+
"eval_loss": 0.17014054954051971,
|
213 |
+
"eval_roc_auc": 0.8078434001509203,
|
214 |
+
"eval_runtime": 662.7207,
|
215 |
+
"eval_samples_per_second": 4.223,
|
216 |
+
"eval_steps_per_second": 0.264,
|
217 |
+
"learning_rate": 0.01,
|
218 |
+
"step": 5896
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 11.19,
|
222 |
+
"learning_rate": 0.01,
|
223 |
+
"loss": 0.213,
|
224 |
+
"step": 6000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 12.0,
|
228 |
+
"eval_accuracy": 0.3565559128260093,
|
229 |
+
"eval_f1_macro": 0.5371100777543761,
|
230 |
+
"eval_f1_micro": 0.6629161350191394,
|
231 |
+
"eval_loss": 0.2409001588821411,
|
232 |
+
"eval_roc_auc": 0.8034829311226885,
|
233 |
+
"eval_runtime": 657.3534,
|
234 |
+
"eval_samples_per_second": 4.258,
|
235 |
+
"eval_steps_per_second": 0.266,
|
236 |
+
"learning_rate": 0.01,
|
237 |
+
"step": 6432
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 12.13,
|
241 |
+
"learning_rate": 0.01,
|
242 |
+
"loss": 0.2161,
|
243 |
+
"step": 6500
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 13.0,
|
247 |
+
"eval_accuracy": 0.4194355126831011,
|
248 |
+
"eval_f1_macro": 0.6243034191518838,
|
249 |
+
"eval_f1_micro": 0.7290690310322989,
|
250 |
+
"eval_loss": 0.1704823523759842,
|
251 |
+
"eval_roc_auc": 0.820743148916754,
|
252 |
+
"eval_runtime": 665.336,
|
253 |
+
"eval_samples_per_second": 4.207,
|
254 |
+
"eval_steps_per_second": 0.263,
|
255 |
+
"learning_rate": 0.01,
|
256 |
+
"step": 6968
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 13.06,
|
260 |
+
"learning_rate": 0.01,
|
261 |
+
"loss": 0.2145,
|
262 |
+
"step": 7000
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 13.99,
|
266 |
+
"learning_rate": 0.01,
|
267 |
+
"loss": 0.2136,
|
268 |
+
"step": 7500
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 14.0,
|
272 |
+
"eval_accuracy": 0.44265809217577706,
|
273 |
+
"eval_f1_macro": 0.6215367292632343,
|
274 |
+
"eval_f1_micro": 0.734259379462547,
|
275 |
+
"eval_loss": 0.16933664679527283,
|
276 |
+
"eval_roc_auc": 0.8177495297062425,
|
277 |
+
"eval_runtime": 654.4339,
|
278 |
+
"eval_samples_per_second": 4.277,
|
279 |
+
"eval_steps_per_second": 0.267,
|
280 |
+
"learning_rate": 0.01,
|
281 |
+
"step": 7504
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 14.93,
|
285 |
+
"learning_rate": 0.001,
|
286 |
+
"loss": 0.1826,
|
287 |
+
"step": 8000
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 15.0,
|
291 |
+
"eval_accuracy": 0.4969632011432655,
|
292 |
+
"eval_f1_macro": 0.7459554779815883,
|
293 |
+
"eval_f1_micro": 0.7967720076387605,
|
294 |
+
"eval_loss": 0.13880470395088196,
|
295 |
+
"eval_roc_auc": 0.8648126874646008,
|
296 |
+
"eval_runtime": 668.3546,
|
297 |
+
"eval_samples_per_second": 4.188,
|
298 |
+
"eval_steps_per_second": 0.262,
|
299 |
+
"learning_rate": 0.001,
|
300 |
+
"step": 8040
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 15.86,
|
304 |
+
"learning_rate": 0.001,
|
305 |
+
"loss": 0.1731,
|
306 |
+
"step": 8500
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 16.0,
|
310 |
+
"eval_accuracy": 0.5155412647374062,
|
311 |
+
"eval_f1_macro": 0.7385190479208986,
|
312 |
+
"eval_f1_micro": 0.8016270337922403,
|
313 |
+
"eval_loss": 0.14762958884239197,
|
314 |
+
"eval_roc_auc": 0.8631052728177114,
|
315 |
+
"eval_runtime": 657.5159,
|
316 |
+
"eval_samples_per_second": 4.257,
|
317 |
+
"eval_steps_per_second": 0.266,
|
318 |
+
"learning_rate": 0.001,
|
319 |
+
"step": 8576
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 16.79,
|
323 |
+
"learning_rate": 0.001,
|
324 |
+
"loss": 0.1649,
|
325 |
+
"step": 9000
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 17.0,
|
329 |
+
"eval_accuracy": 0.5023222579492675,
|
330 |
+
"eval_f1_macro": 0.7692855132157861,
|
331 |
+
"eval_f1_micro": 0.813318908522659,
|
332 |
+
"eval_loss": 0.1351146697998047,
|
333 |
+
"eval_roc_auc": 0.8811667807242409,
|
334 |
+
"eval_runtime": 655.4487,
|
335 |
+
"eval_samples_per_second": 4.27,
|
336 |
+
"eval_steps_per_second": 0.267,
|
337 |
+
"learning_rate": 0.001,
|
338 |
+
"step": 9112
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 17.72,
|
342 |
+
"learning_rate": 0.001,
|
343 |
+
"loss": 0.1624,
|
344 |
+
"step": 9500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 18.0,
|
348 |
+
"eval_accuracy": 0.5212575919971418,
|
349 |
+
"eval_f1_macro": 0.7714136364578609,
|
350 |
+
"eval_f1_micro": 0.8185593067340675,
|
351 |
+
"eval_loss": 0.13848340511322021,
|
352 |
+
"eval_roc_auc": 0.8837626029149266,
|
353 |
+
"eval_runtime": 662.3967,
|
354 |
+
"eval_samples_per_second": 4.226,
|
355 |
+
"eval_steps_per_second": 0.264,
|
356 |
+
"learning_rate": 0.001,
|
357 |
+
"step": 9648
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 18.66,
|
361 |
+
"learning_rate": 0.001,
|
362 |
+
"loss": 0.1576,
|
363 |
+
"step": 10000
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 19.0,
|
367 |
+
"eval_accuracy": 0.5205430510896749,
|
368 |
+
"eval_f1_macro": 0.7631806100991755,
|
369 |
+
"eval_f1_micro": 0.8175969609705288,
|
370 |
+
"eval_loss": 0.13018544018268585,
|
371 |
+
"eval_roc_auc": 0.8779148602729905,
|
372 |
+
"eval_runtime": 679.9157,
|
373 |
+
"eval_samples_per_second": 4.117,
|
374 |
+
"eval_steps_per_second": 0.257,
|
375 |
+
"learning_rate": 0.001,
|
376 |
+
"step": 10184
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 19.59,
|
380 |
+
"learning_rate": 0.001,
|
381 |
+
"loss": 0.1544,
|
382 |
+
"step": 10500
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 20.0,
|
386 |
+
"eval_accuracy": 0.5319757056091461,
|
387 |
+
"eval_f1_macro": 0.7721701876863485,
|
388 |
+
"eval_f1_micro": 0.8229907814143413,
|
389 |
+
"eval_loss": 0.12343110144138336,
|
390 |
+
"eval_roc_auc": 0.8780568409783432,
|
391 |
+
"eval_runtime": 658.9399,
|
392 |
+
"eval_samples_per_second": 4.248,
|
393 |
+
"eval_steps_per_second": 0.266,
|
394 |
+
"learning_rate": 0.001,
|
395 |
+
"step": 10720
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 20.52,
|
399 |
+
"learning_rate": 0.001,
|
400 |
+
"loss": 0.1542,
|
401 |
+
"step": 11000
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 21.0,
|
405 |
+
"eval_accuracy": 0.525902107895677,
|
406 |
+
"eval_f1_macro": 0.7754587726434372,
|
407 |
+
"eval_f1_micro": 0.8267645466032855,
|
408 |
+
"eval_loss": 0.13044790923595428,
|
409 |
+
"eval_roc_auc": 0.8884477411328827,
|
410 |
+
"eval_runtime": 659.4337,
|
411 |
+
"eval_samples_per_second": 4.245,
|
412 |
+
"eval_steps_per_second": 0.265,
|
413 |
+
"learning_rate": 0.001,
|
414 |
+
"step": 11256
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 21.46,
|
418 |
+
"learning_rate": 0.001,
|
419 |
+
"loss": 0.1525,
|
420 |
+
"step": 11500
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 22.0,
|
424 |
+
"eval_accuracy": 0.5376920328688818,
|
425 |
+
"eval_f1_macro": 0.76461964189607,
|
426 |
+
"eval_f1_micro": 0.8176823176823177,
|
427 |
+
"eval_loss": 0.12196117639541626,
|
428 |
+
"eval_roc_auc": 0.8724977198539325,
|
429 |
+
"eval_runtime": 674.7285,
|
430 |
+
"eval_samples_per_second": 4.148,
|
431 |
+
"eval_steps_per_second": 0.259,
|
432 |
+
"learning_rate": 0.001,
|
433 |
+
"step": 11792
|
434 |
+
},
|
435 |
+
{
|
436 |
+
"epoch": 22.39,
|
437 |
+
"learning_rate": 0.001,
|
438 |
+
"loss": 0.1505,
|
439 |
+
"step": 12000
|
440 |
+
},
|
441 |
+
{
|
442 |
+
"epoch": 23.0,
|
443 |
+
"eval_accuracy": 0.5255448374419436,
|
444 |
+
"eval_f1_macro": 0.7872363331907107,
|
445 |
+
"eval_f1_micro": 0.8264049955396968,
|
446 |
+
"eval_loss": 0.13109837472438812,
|
447 |
+
"eval_roc_auc": 0.891814004155335,
|
448 |
+
"eval_runtime": 658.8901,
|
449 |
+
"eval_samples_per_second": 4.248,
|
450 |
+
"eval_steps_per_second": 0.266,
|
451 |
+
"learning_rate": 0.001,
|
452 |
+
"step": 12328
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 23.32,
|
456 |
+
"learning_rate": 0.001,
|
457 |
+
"loss": 0.1515,
|
458 |
+
"step": 12500
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 24.0,
|
462 |
+
"eval_accuracy": 0.5316184351554126,
|
463 |
+
"eval_f1_macro": 0.7629977302664256,
|
464 |
+
"eval_f1_micro": 0.8188446438586652,
|
465 |
+
"eval_loss": 0.12468679994344711,
|
466 |
+
"eval_roc_auc": 0.8736890703040258,
|
467 |
+
"eval_runtime": 649.926,
|
468 |
+
"eval_samples_per_second": 4.307,
|
469 |
+
"eval_steps_per_second": 0.269,
|
470 |
+
"learning_rate": 0.001,
|
471 |
+
"step": 12864
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 24.25,
|
475 |
+
"learning_rate": 0.001,
|
476 |
+
"loss": 0.1471,
|
477 |
+
"step": 13000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 25.0,
|
481 |
+
"eval_accuracy": 0.5226866738120758,
|
482 |
+
"eval_f1_macro": 0.7726394339166129,
|
483 |
+
"eval_f1_micro": 0.8258118937157316,
|
484 |
+
"eval_loss": 0.1264602392911911,
|
485 |
+
"eval_roc_auc": 0.8875076814616129,
|
486 |
+
"eval_runtime": 661.9223,
|
487 |
+
"eval_samples_per_second": 4.229,
|
488 |
+
"eval_steps_per_second": 0.264,
|
489 |
+
"learning_rate": 0.001,
|
490 |
+
"step": 13400
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 25.19,
|
494 |
+
"learning_rate": 0.001,
|
495 |
+
"loss": 0.1475,
|
496 |
+
"step": 13500
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 26.0,
|
500 |
+
"eval_accuracy": 0.5330475169703466,
|
501 |
+
"eval_f1_macro": 0.7834057788273457,
|
502 |
+
"eval_f1_micro": 0.8301059554256485,
|
503 |
+
"eval_loss": 0.12767212092876434,
|
504 |
+
"eval_roc_auc": 0.8867279830614062,
|
505 |
+
"eval_runtime": 654.8587,
|
506 |
+
"eval_samples_per_second": 4.274,
|
507 |
+
"eval_steps_per_second": 0.267,
|
508 |
+
"learning_rate": 0.001,
|
509 |
+
"step": 13936
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 26.12,
|
513 |
+
"learning_rate": 0.001,
|
514 |
+
"loss": 0.1484,
|
515 |
+
"step": 14000
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 27.0,
|
519 |
+
"eval_accuracy": 0.5398356555912825,
|
520 |
+
"eval_f1_macro": 0.7661233818608114,
|
521 |
+
"eval_f1_micro": 0.8218290555693994,
|
522 |
+
"eval_loss": 0.12360195070505142,
|
523 |
+
"eval_roc_auc": 0.8754121833739898,
|
524 |
+
"eval_runtime": 658.0747,
|
525 |
+
"eval_samples_per_second": 4.253,
|
526 |
+
"eval_steps_per_second": 0.266,
|
527 |
+
"learning_rate": 0.001,
|
528 |
+
"step": 14472
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 27.05,
|
532 |
+
"learning_rate": 0.001,
|
533 |
+
"loss": 0.1475,
|
534 |
+
"step": 14500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 27.99,
|
538 |
+
"learning_rate": 0.001,
|
539 |
+
"loss": 0.1472,
|
540 |
+
"step": 15000
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 28.0,
|
544 |
+
"eval_accuracy": 0.5401929260450161,
|
545 |
+
"eval_f1_macro": 0.7728524137767288,
|
546 |
+
"eval_f1_micro": 0.8228913409388442,
|
547 |
+
"eval_loss": 0.1256585270166397,
|
548 |
+
"eval_roc_auc": 0.8757945197904818,
|
549 |
+
"eval_runtime": 655.1438,
|
550 |
+
"eval_samples_per_second": 4.272,
|
551 |
+
"eval_steps_per_second": 0.267,
|
552 |
+
"learning_rate": 0.001,
|
553 |
+
"step": 15008
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"epoch": 28.92,
|
557 |
+
"learning_rate": 0.0001,
|
558 |
+
"loss": 0.1379,
|
559 |
+
"step": 15500
|
560 |
+
},
|
561 |
+
{
|
562 |
+
"epoch": 29.0,
|
563 |
+
"eval_accuracy": 0.5451947123972848,
|
564 |
+
"eval_f1_macro": 0.7891009081675242,
|
565 |
+
"eval_f1_micro": 0.8352094482376824,
|
566 |
+
"eval_loss": 0.11994459480047226,
|
567 |
+
"eval_roc_auc": 0.8865305601939816,
|
568 |
+
"eval_runtime": 657.6068,
|
569 |
+
"eval_samples_per_second": 4.256,
|
570 |
+
"eval_steps_per_second": 0.266,
|
571 |
+
"learning_rate": 0.0001,
|
572 |
+
"step": 15544
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 29.85,
|
576 |
+
"learning_rate": 0.0001,
|
577 |
+
"loss": 0.1349,
|
578 |
+
"step": 16000
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 30.0,
|
582 |
+
"eval_accuracy": 0.5487674169346195,
|
583 |
+
"eval_f1_macro": 0.7943922901606704,
|
584 |
+
"eval_f1_micro": 0.841315916787615,
|
585 |
+
"eval_loss": 0.11564121395349503,
|
586 |
+
"eval_roc_auc": 0.895138845939546,
|
587 |
+
"eval_runtime": 659.4245,
|
588 |
+
"eval_samples_per_second": 4.245,
|
589 |
+
"eval_steps_per_second": 0.265,
|
590 |
+
"learning_rate": 0.0001,
|
591 |
+
"step": 16080
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 30.78,
|
595 |
+
"learning_rate": 0.0001,
|
596 |
+
"loss": 0.1326,
|
597 |
+
"step": 16500
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 31.0,
|
601 |
+
"eval_accuracy": 0.5498392282958199,
|
602 |
+
"eval_f1_macro": 0.7983364672353336,
|
603 |
+
"eval_f1_micro": 0.84037558685446,
|
604 |
+
"eval_loss": 0.11515345424413681,
|
605 |
+
"eval_roc_auc": 0.8960665621862288,
|
606 |
+
"eval_runtime": 652.6109,
|
607 |
+
"eval_samples_per_second": 4.289,
|
608 |
+
"eval_steps_per_second": 0.268,
|
609 |
+
"learning_rate": 0.0001,
|
610 |
+
"step": 16616
|
611 |
+
},
|
612 |
+
{
|
613 |
+
"epoch": 31.72,
|
614 |
+
"learning_rate": 0.0001,
|
615 |
+
"loss": 0.1321,
|
616 |
+
"step": 17000
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 32.0,
|
620 |
+
"eval_accuracy": 0.5509110396570204,
|
621 |
+
"eval_f1_macro": 0.7911364775051637,
|
622 |
+
"eval_f1_micro": 0.8385913426265589,
|
623 |
+
"eval_loss": 0.11371538788080215,
|
624 |
+
"eval_roc_auc": 0.89024940896342,
|
625 |
+
"eval_runtime": 663.8067,
|
626 |
+
"eval_samples_per_second": 4.217,
|
627 |
+
"eval_steps_per_second": 0.264,
|
628 |
+
"learning_rate": 0.0001,
|
629 |
+
"step": 17152
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 32.65,
|
633 |
+
"learning_rate": 0.0001,
|
634 |
+
"loss": 0.1294,
|
635 |
+
"step": 17500
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 33.0,
|
639 |
+
"eval_accuracy": 0.5512683101107538,
|
640 |
+
"eval_f1_macro": 0.7924466456431516,
|
641 |
+
"eval_f1_micro": 0.8406133545115768,
|
642 |
+
"eval_loss": 0.11363548040390015,
|
643 |
+
"eval_roc_auc": 0.891620070857568,
|
644 |
+
"eval_runtime": 641.5792,
|
645 |
+
"eval_samples_per_second": 4.363,
|
646 |
+
"eval_steps_per_second": 0.273,
|
647 |
+
"learning_rate": 0.0001,
|
648 |
+
"step": 17688
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 33.58,
|
652 |
+
"learning_rate": 0.0001,
|
653 |
+
"loss": 0.1297,
|
654 |
+
"step": 18000
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 34.0,
|
658 |
+
"eval_accuracy": 0.5551982851018221,
|
659 |
+
"eval_f1_macro": 0.7994952163993189,
|
660 |
+
"eval_f1_micro": 0.8438879816125325,
|
661 |
+
"eval_loss": 0.1100151389837265,
|
662 |
+
"eval_roc_auc": 0.8965325171836247,
|
663 |
+
"eval_runtime": 666.5011,
|
664 |
+
"eval_samples_per_second": 4.2,
|
665 |
+
"eval_steps_per_second": 0.263,
|
666 |
+
"learning_rate": 0.0001,
|
667 |
+
"step": 18224
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 34.51,
|
671 |
+
"learning_rate": 0.0001,
|
672 |
+
"loss": 0.1296,
|
673 |
+
"step": 18500
|
674 |
+
},
|
675 |
+
{
|
676 |
+
"epoch": 35.0,
|
677 |
+
"eval_accuracy": 0.5569846373704894,
|
678 |
+
"eval_f1_macro": 0.7958547104209327,
|
679 |
+
"eval_f1_micro": 0.8430563978168587,
|
680 |
+
"eval_loss": 0.11022897809743881,
|
681 |
+
"eval_roc_auc": 0.8952561397957243,
|
682 |
+
"eval_runtime": 651.2057,
|
683 |
+
"eval_samples_per_second": 4.298,
|
684 |
+
"eval_steps_per_second": 0.269,
|
685 |
+
"learning_rate": 0.0001,
|
686 |
+
"step": 18760
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 35.45,
|
690 |
+
"learning_rate": 0.0001,
|
691 |
+
"loss": 0.1276,
|
692 |
+
"step": 19000
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 36.0,
|
696 |
+
"eval_accuracy": 0.5584137191854234,
|
697 |
+
"eval_f1_macro": 0.7954026209465306,
|
698 |
+
"eval_f1_micro": 0.8428501708150318,
|
699 |
+
"eval_loss": 0.11038191616535187,
|
700 |
+
"eval_roc_auc": 0.8933002510838177,
|
701 |
+
"eval_runtime": 659.4359,
|
702 |
+
"eval_samples_per_second": 4.245,
|
703 |
+
"eval_steps_per_second": 0.265,
|
704 |
+
"learning_rate": 0.0001,
|
705 |
+
"step": 19296
|
706 |
+
},
|
707 |
+
{
|
708 |
+
"epoch": 36.38,
|
709 |
+
"learning_rate": 0.0001,
|
710 |
+
"loss": 0.1264,
|
711 |
+
"step": 19500
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 37.0,
|
715 |
+
"eval_accuracy": 0.5644873168988924,
|
716 |
+
"eval_f1_macro": 0.8073166114826349,
|
717 |
+
"eval_f1_micro": 0.8467775842392937,
|
718 |
+
"eval_loss": 0.11108729988336563,
|
719 |
+
"eval_roc_auc": 0.9003547734565809,
|
720 |
+
"eval_runtime": 649.1518,
|
721 |
+
"eval_samples_per_second": 4.312,
|
722 |
+
"eval_steps_per_second": 0.27,
|
723 |
+
"learning_rate": 0.0001,
|
724 |
+
"step": 19832
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 37.31,
|
728 |
+
"learning_rate": 0.0001,
|
729 |
+
"loss": 0.1279,
|
730 |
+
"step": 20000
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 38.0,
|
734 |
+
"eval_accuracy": 0.5662736691675598,
|
735 |
+
"eval_f1_macro": 0.8059718169069215,
|
736 |
+
"eval_f1_micro": 0.8457475869604808,
|
737 |
+
"eval_loss": 0.11050034314393997,
|
738 |
+
"eval_roc_auc": 0.8964483289954395,
|
739 |
+
"eval_runtime": 661.6908,
|
740 |
+
"eval_samples_per_second": 4.23,
|
741 |
+
"eval_steps_per_second": 0.264,
|
742 |
+
"learning_rate": 0.0001,
|
743 |
+
"step": 20368
|
744 |
+
},
|
745 |
+
{
|
746 |
+
"epoch": 38.25,
|
747 |
+
"learning_rate": 0.0001,
|
748 |
+
"loss": 0.1231,
|
749 |
+
"step": 20500
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 39.0,
|
753 |
+
"eval_accuracy": 0.5623436941764915,
|
754 |
+
"eval_f1_macro": 0.810452280144873,
|
755 |
+
"eval_f1_micro": 0.8481141692150868,
|
756 |
+
"eval_loss": 0.1114969253540039,
|
757 |
+
"eval_roc_auc": 0.9016441984475202,
|
758 |
+
"eval_runtime": 645.0365,
|
759 |
+
"eval_samples_per_second": 4.339,
|
760 |
+
"eval_steps_per_second": 0.271,
|
761 |
+
"learning_rate": 0.0001,
|
762 |
+
"step": 20904
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 39.18,
|
766 |
+
"learning_rate": 0.0001,
|
767 |
+
"loss": 0.1276,
|
768 |
+
"step": 21000
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 40.0,
|
772 |
+
"eval_accuracy": 0.564844587352626,
|
773 |
+
"eval_f1_macro": 0.7998757908744976,
|
774 |
+
"eval_f1_micro": 0.8442948914040991,
|
775 |
+
"eval_loss": 0.10886894911527634,
|
776 |
+
"eval_roc_auc": 0.8932229387863774,
|
777 |
+
"eval_runtime": 664.3024,
|
778 |
+
"eval_samples_per_second": 4.213,
|
779 |
+
"eval_steps_per_second": 0.263,
|
780 |
+
"learning_rate": 0.0001,
|
781 |
+
"step": 21440
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 40.11,
|
785 |
+
"learning_rate": 0.0001,
|
786 |
+
"loss": 0.121,
|
787 |
+
"step": 21500
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 41.0,
|
791 |
+
"eval_accuracy": 0.5627009646302251,
|
792 |
+
"eval_f1_macro": 0.8014681171854418,
|
793 |
+
"eval_f1_micro": 0.8454512239678481,
|
794 |
+
"eval_loss": 0.10981705039739609,
|
795 |
+
"eval_roc_auc": 0.8953198348828445,
|
796 |
+
"eval_runtime": 652.9462,
|
797 |
+
"eval_samples_per_second": 4.287,
|
798 |
+
"eval_steps_per_second": 0.268,
|
799 |
+
"learning_rate": 0.0001,
|
800 |
+
"step": 21976
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 41.04,
|
804 |
+
"learning_rate": 0.0001,
|
805 |
+
"loss": 0.1241,
|
806 |
+
"step": 22000
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 41.98,
|
810 |
+
"learning_rate": 0.0001,
|
811 |
+
"loss": 0.1229,
|
812 |
+
"step": 22500
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 42.0,
|
816 |
+
"eval_accuracy": 0.5619864237227581,
|
817 |
+
"eval_f1_macro": 0.8009464266100834,
|
818 |
+
"eval_f1_micro": 0.8459390554813646,
|
819 |
+
"eval_loss": 0.10872387140989304,
|
820 |
+
"eval_roc_auc": 0.8965747458488879,
|
821 |
+
"eval_runtime": 655.0026,
|
822 |
+
"eval_samples_per_second": 4.273,
|
823 |
+
"eval_steps_per_second": 0.267,
|
824 |
+
"learning_rate": 0.0001,
|
825 |
+
"step": 22512
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 42.91,
|
829 |
+
"learning_rate": 0.0001,
|
830 |
+
"loss": 0.1227,
|
831 |
+
"step": 23000
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 43.0,
|
835 |
+
"eval_accuracy": 0.5687745623436942,
|
836 |
+
"eval_f1_macro": 0.8067169938735436,
|
837 |
+
"eval_f1_micro": 0.8468292682926829,
|
838 |
+
"eval_loss": 0.10882638394832611,
|
839 |
+
"eval_roc_auc": 0.8956753719148074,
|
840 |
+
"eval_runtime": 660.8489,
|
841 |
+
"eval_samples_per_second": 4.235,
|
842 |
+
"eval_steps_per_second": 0.265,
|
843 |
+
"learning_rate": 0.0001,
|
844 |
+
"step": 23048
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 43.84,
|
848 |
+
"learning_rate": 0.0001,
|
849 |
+
"loss": 0.1221,
|
850 |
+
"step": 23500
|
851 |
+
},
|
852 |
+
{
|
853 |
+
"epoch": 44.0,
|
854 |
+
"eval_accuracy": 0.5673454805287603,
|
855 |
+
"eval_f1_macro": 0.8065744332804647,
|
856 |
+
"eval_f1_micro": 0.8476346632659257,
|
857 |
+
"eval_loss": 0.10762665420770645,
|
858 |
+
"eval_roc_auc": 0.897394560379003,
|
859 |
+
"eval_runtime": 657.9808,
|
860 |
+
"eval_samples_per_second": 4.254,
|
861 |
+
"eval_steps_per_second": 0.266,
|
862 |
+
"learning_rate": 0.0001,
|
863 |
+
"step": 23584
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 44.78,
|
867 |
+
"learning_rate": 0.0001,
|
868 |
+
"loss": 0.1191,
|
869 |
+
"step": 24000
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 45.0,
|
873 |
+
"eval_accuracy": 0.5698463737048947,
|
874 |
+
"eval_f1_macro": 0.817339787108748,
|
875 |
+
"eval_f1_micro": 0.8507955568898229,
|
876 |
+
"eval_loss": 0.10689569264650345,
|
877 |
+
"eval_roc_auc": 0.9027498177813545,
|
878 |
+
"eval_runtime": 671.4032,
|
879 |
+
"eval_samples_per_second": 4.169,
|
880 |
+
"eval_steps_per_second": 0.261,
|
881 |
+
"learning_rate": 0.0001,
|
882 |
+
"step": 24120
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 45.71,
|
886 |
+
"learning_rate": 0.0001,
|
887 |
+
"loss": 0.1212,
|
888 |
+
"step": 24500
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 46.0,
|
892 |
+
"eval_accuracy": 0.563415505537692,
|
893 |
+
"eval_f1_macro": 0.8173548628519673,
|
894 |
+
"eval_f1_micro": 0.8508932256352808,
|
895 |
+
"eval_loss": 0.10721632838249207,
|
896 |
+
"eval_roc_auc": 0.9086343215547117,
|
897 |
+
"eval_runtime": 661.9534,
|
898 |
+
"eval_samples_per_second": 4.228,
|
899 |
+
"eval_steps_per_second": 0.264,
|
900 |
+
"learning_rate": 0.0001,
|
901 |
+
"step": 24656
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 46.64,
|
905 |
+
"learning_rate": 0.0001,
|
906 |
+
"loss": 0.1198,
|
907 |
+
"step": 25000
|
908 |
+
},
|
909 |
+
{
|
910 |
+
"epoch": 47.0,
|
911 |
+
"eval_accuracy": 0.5687745623436942,
|
912 |
+
"eval_f1_macro": 0.8090465345914333,
|
913 |
+
"eval_f1_micro": 0.849293563579278,
|
914 |
+
"eval_loss": 0.10663535445928574,
|
915 |
+
"eval_roc_auc": 0.900130416633457,
|
916 |
+
"eval_runtime": 673.8512,
|
917 |
+
"eval_samples_per_second": 4.154,
|
918 |
+
"eval_steps_per_second": 0.26,
|
919 |
+
"learning_rate": 0.0001,
|
920 |
+
"step": 25192
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 47.57,
|
924 |
+
"learning_rate": 0.0001,
|
925 |
+
"loss": 0.1201,
|
926 |
+
"step": 25500
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 48.0,
|
930 |
+
"eval_accuracy": 0.5652018578063595,
|
931 |
+
"eval_f1_macro": 0.8082582901487069,
|
932 |
+
"eval_f1_micro": 0.8484811957569913,
|
933 |
+
"eval_loss": 0.10762892663478851,
|
934 |
+
"eval_roc_auc": 0.9002407000277199,
|
935 |
+
"eval_runtime": 662.4725,
|
936 |
+
"eval_samples_per_second": 4.225,
|
937 |
+
"eval_steps_per_second": 0.264,
|
938 |
+
"learning_rate": 0.0001,
|
939 |
+
"step": 25728
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 48.51,
|
943 |
+
"learning_rate": 0.0001,
|
944 |
+
"loss": 0.1189,
|
945 |
+
"step": 26000
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 49.0,
|
949 |
+
"eval_accuracy": 0.5687745623436942,
|
950 |
+
"eval_f1_macro": 0.8152048365492516,
|
951 |
+
"eval_f1_micro": 0.8507597141312836,
|
952 |
+
"eval_loss": 0.10654111951589584,
|
953 |
+
"eval_roc_auc": 0.9026532861685821,
|
954 |
+
"eval_runtime": 679.062,
|
955 |
+
"eval_samples_per_second": 4.122,
|
956 |
+
"eval_steps_per_second": 0.258,
|
957 |
+
"learning_rate": 0.0001,
|
958 |
+
"step": 26264
|
959 |
+
},
|
960 |
+
{
|
961 |
+
"epoch": 49.44,
|
962 |
+
"learning_rate": 0.0001,
|
963 |
+
"loss": 0.1176,
|
964 |
+
"step": 26500
|
965 |
+
},
|
966 |
+
{
|
967 |
+
"epoch": 50.0,
|
968 |
+
"eval_accuracy": 0.563415505537692,
|
969 |
+
"eval_f1_macro": 0.8033757987633436,
|
970 |
+
"eval_f1_micro": 0.8462940461725396,
|
971 |
+
"eval_loss": 0.10730718821287155,
|
972 |
+
"eval_roc_auc": 0.8965096424048645,
|
973 |
+
"eval_runtime": 643.0995,
|
974 |
+
"eval_samples_per_second": 4.352,
|
975 |
+
"eval_steps_per_second": 0.272,
|
976 |
+
"learning_rate": 0.0001,
|
977 |
+
"step": 26800
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 50.37,
|
981 |
+
"learning_rate": 0.0001,
|
982 |
+
"loss": 0.1202,
|
983 |
+
"step": 27000
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 51.0,
|
987 |
+
"eval_accuracy": 0.5727045373347625,
|
988 |
+
"eval_f1_macro": 0.8102272806912906,
|
989 |
+
"eval_f1_micro": 0.8480676328502416,
|
990 |
+
"eval_loss": 0.10732194036245346,
|
991 |
+
"eval_roc_auc": 0.8994002578253693,
|
992 |
+
"eval_runtime": 645.0782,
|
993 |
+
"eval_samples_per_second": 4.339,
|
994 |
+
"eval_steps_per_second": 0.271,
|
995 |
+
"learning_rate": 0.0001,
|
996 |
+
"step": 27336
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 51.31,
|
1000 |
+
"learning_rate": 0.0001,
|
1001 |
+
"loss": 0.1167,
|
1002 |
+
"step": 27500
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 52.0,
|
1006 |
+
"eval_accuracy": 0.571632725973562,
|
1007 |
+
"eval_f1_macro": 0.8179079350328715,
|
1008 |
+
"eval_f1_micro": 0.8521687462863933,
|
1009 |
+
"eval_loss": 0.10600127279758453,
|
1010 |
+
"eval_roc_auc": 0.9068707918898985,
|
1011 |
+
"eval_runtime": 661.3351,
|
1012 |
+
"eval_samples_per_second": 4.232,
|
1013 |
+
"eval_steps_per_second": 0.265,
|
1014 |
+
"learning_rate": 0.0001,
|
1015 |
+
"step": 27872
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 52.24,
|
1019 |
+
"learning_rate": 0.0001,
|
1020 |
+
"loss": 0.1192,
|
1021 |
+
"step": 28000
|
1022 |
+
},
|
1023 |
+
{
|
1024 |
+
"epoch": 53.0,
|
1025 |
+
"eval_accuracy": 0.5712754555198285,
|
1026 |
+
"eval_f1_macro": 0.8127939311574505,
|
1027 |
+
"eval_f1_micro": 0.8507093268940539,
|
1028 |
+
"eval_loss": 0.10628383606672287,
|
1029 |
+
"eval_roc_auc": 0.9009869922949931,
|
1030 |
+
"eval_runtime": 649.2151,
|
1031 |
+
"eval_samples_per_second": 4.311,
|
1032 |
+
"eval_steps_per_second": 0.27,
|
1033 |
+
"learning_rate": 0.0001,
|
1034 |
+
"step": 28408
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 53.17,
|
1038 |
+
"learning_rate": 0.0001,
|
1039 |
+
"loss": 0.1156,
|
1040 |
+
"step": 28500
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 54.0,
|
1044 |
+
"eval_accuracy": 0.5719899964272954,
|
1045 |
+
"eval_f1_macro": 0.8113280265904282,
|
1046 |
+
"eval_f1_micro": 0.8493415488703637,
|
1047 |
+
"eval_loss": 0.10670817643404007,
|
1048 |
+
"eval_roc_auc": 0.9000055427080821,
|
1049 |
+
"eval_runtime": 659.1604,
|
1050 |
+
"eval_samples_per_second": 4.246,
|
1051 |
+
"eval_steps_per_second": 0.265,
|
1052 |
+
"learning_rate": 0.0001,
|
1053 |
+
"step": 28944
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 54.1,
|
1057 |
+
"learning_rate": 0.0001,
|
1058 |
+
"loss": 0.1193,
|
1059 |
+
"step": 29000
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 55.0,
|
1063 |
+
"eval_accuracy": 0.5727045373347625,
|
1064 |
+
"eval_f1_macro": 0.8116343899877551,
|
1065 |
+
"eval_f1_micro": 0.8490052609300356,
|
1066 |
+
"eval_loss": 0.10690104961395264,
|
1067 |
+
"eval_roc_auc": 0.8994929994035434,
|
1068 |
+
"eval_runtime": 663.4912,
|
1069 |
+
"eval_samples_per_second": 4.219,
|
1070 |
+
"eval_steps_per_second": 0.264,
|
1071 |
+
"learning_rate": 0.0001,
|
1072 |
+
"step": 29480
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 55.04,
|
1076 |
+
"learning_rate": 0.0001,
|
1077 |
+
"loss": 0.1161,
|
1078 |
+
"step": 29500
|
1079 |
+
},
|
1080 |
+
{
|
1081 |
+
"epoch": 55.97,
|
1082 |
+
"learning_rate": 0.0001,
|
1083 |
+
"loss": 0.116,
|
1084 |
+
"step": 30000
|
1085 |
+
},
|
1086 |
+
{
|
1087 |
+
"epoch": 56.0,
|
1088 |
+
"eval_accuracy": 0.5744908896034298,
|
1089 |
+
"eval_f1_macro": 0.818624257330544,
|
1090 |
+
"eval_f1_micro": 0.8542707589816796,
|
1091 |
+
"eval_loss": 0.10558204352855682,
|
1092 |
+
"eval_roc_auc": 0.9077319963356549,
|
1093 |
+
"eval_runtime": 659.0048,
|
1094 |
+
"eval_samples_per_second": 4.247,
|
1095 |
+
"eval_steps_per_second": 0.266,
|
1096 |
+
"learning_rate": 0.0001,
|
1097 |
+
"step": 30016
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 56.9,
|
1101 |
+
"learning_rate": 0.0001,
|
1102 |
+
"loss": 0.1147,
|
1103 |
+
"step": 30500
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 57.0,
|
1107 |
+
"eval_accuracy": 0.5730618077884959,
|
1108 |
+
"eval_f1_macro": 0.8113712664273885,
|
1109 |
+
"eval_f1_micro": 0.8505116959064327,
|
1110 |
+
"eval_loss": 0.10627623647451401,
|
1111 |
+
"eval_roc_auc": 0.8980389880708319,
|
1112 |
+
"eval_runtime": 669.8276,
|
1113 |
+
"eval_samples_per_second": 4.179,
|
1114 |
+
"eval_steps_per_second": 0.261,
|
1115 |
+
"learning_rate": 0.0001,
|
1116 |
+
"step": 30552
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 57.84,
|
1120 |
+
"learning_rate": 0.0001,
|
1121 |
+
"loss": 0.1139,
|
1122 |
+
"step": 31000
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 58.0,
|
1126 |
+
"eval_accuracy": 0.5705609146123616,
|
1127 |
+
"eval_f1_macro": 0.8074486566040442,
|
1128 |
+
"eval_f1_micro": 0.8488632919066383,
|
1129 |
+
"eval_loss": 0.10657747834920883,
|
1130 |
+
"eval_roc_auc": 0.898615796069772,
|
1131 |
+
"eval_runtime": 654.902,
|
1132 |
+
"eval_samples_per_second": 4.274,
|
1133 |
+
"eval_steps_per_second": 0.267,
|
1134 |
+
"learning_rate": 0.0001,
|
1135 |
+
"step": 31088
|
1136 |
+
},
|
1137 |
+
{
|
1138 |
+
"epoch": 58.77,
|
1139 |
+
"learning_rate": 0.0001,
|
1140 |
+
"loss": 0.1143,
|
1141 |
+
"step": 31500
|
1142 |
+
},
|
1143 |
+
{
|
1144 |
+
"epoch": 59.0,
|
1145 |
+
"eval_accuracy": 0.5727045373347625,
|
1146 |
+
"eval_f1_macro": 0.8065194300593265,
|
1147 |
+
"eval_f1_micro": 0.8490646517579673,
|
1148 |
+
"eval_loss": 0.10738535225391388,
|
1149 |
+
"eval_roc_auc": 0.8971341466029096,
|
1150 |
+
"eval_runtime": 661.8186,
|
1151 |
+
"eval_samples_per_second": 4.229,
|
1152 |
+
"eval_steps_per_second": 0.264,
|
1153 |
+
"learning_rate": 0.0001,
|
1154 |
+
"step": 31624
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 59.7,
|
1158 |
+
"learning_rate": 0.0001,
|
1159 |
+
"loss": 0.1148,
|
1160 |
+
"step": 32000
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 60.0,
|
1164 |
+
"eval_accuracy": 0.5694891032511611,
|
1165 |
+
"eval_f1_macro": 0.8079607267977419,
|
1166 |
+
"eval_f1_micro": 0.8498935199269851,
|
1167 |
+
"eval_loss": 0.10777446627616882,
|
1168 |
+
"eval_roc_auc": 0.8980542827803957,
|
1169 |
+
"eval_runtime": 663.1043,
|
1170 |
+
"eval_samples_per_second": 4.221,
|
1171 |
+
"eval_steps_per_second": 0.264,
|
1172 |
+
"learning_rate": 0.0001,
|
1173 |
+
"step": 32160
|
1174 |
+
},
|
1175 |
+
{
|
1176 |
+
"epoch": 60.63,
|
1177 |
+
"learning_rate": 0.0001,
|
1178 |
+
"loss": 0.1143,
|
1179 |
+
"step": 32500
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 61.0,
|
1183 |
+
"eval_accuracy": 0.572347266881029,
|
1184 |
+
"eval_f1_macro": 0.8159017507862669,
|
1185 |
+
"eval_f1_micro": 0.8512052195976559,
|
1186 |
+
"eval_loss": 0.10536229610443115,
|
1187 |
+
"eval_roc_auc": 0.9010383439575261,
|
1188 |
+
"eval_runtime": 660.7796,
|
1189 |
+
"eval_samples_per_second": 4.236,
|
1190 |
+
"eval_steps_per_second": 0.265,
|
1191 |
+
"learning_rate": 0.0001,
|
1192 |
+
"step": 32696
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 61.57,
|
1196 |
+
"learning_rate": 0.0001,
|
1197 |
+
"loss": 0.1133,
|
1198 |
+
"step": 33000
|
1199 |
+
},
|
1200 |
+
{
|
1201 |
+
"epoch": 62.0,
|
1202 |
+
"eval_accuracy": 0.5737763486959628,
|
1203 |
+
"eval_f1_macro": 0.808341771577789,
|
1204 |
+
"eval_f1_micro": 0.8495672315006705,
|
1205 |
+
"eval_loss": 0.10581369698047638,
|
1206 |
+
"eval_roc_auc": 0.8973203339377094,
|
1207 |
+
"eval_runtime": 655.4703,
|
1208 |
+
"eval_samples_per_second": 4.27,
|
1209 |
+
"eval_steps_per_second": 0.267,
|
1210 |
+
"learning_rate": 0.0001,
|
1211 |
+
"step": 33232
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 62.5,
|
1215 |
+
"learning_rate": 0.0001,
|
1216 |
+
"loss": 0.1134,
|
1217 |
+
"step": 33500
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 63.0,
|
1221 |
+
"eval_accuracy": 0.5680600214362272,
|
1222 |
+
"eval_f1_macro": 0.8087952156757714,
|
1223 |
+
"eval_f1_micro": 0.847873368777187,
|
1224 |
+
"eval_loss": 0.1063385158777237,
|
1225 |
+
"eval_roc_auc": 0.899139005299654,
|
1226 |
+
"eval_runtime": 659.92,
|
1227 |
+
"eval_samples_per_second": 4.241,
|
1228 |
+
"eval_steps_per_second": 0.265,
|
1229 |
+
"learning_rate": 0.0001,
|
1230 |
+
"step": 33768
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 63.43,
|
1234 |
+
"learning_rate": 0.0001,
|
1235 |
+
"loss": 0.1123,
|
1236 |
+
"step": 34000
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 64.0,
|
1240 |
+
"eval_accuracy": 0.5702036441586281,
|
1241 |
+
"eval_f1_macro": 0.8121242156828758,
|
1242 |
+
"eval_f1_micro": 0.8503755754785559,
|
1243 |
+
"eval_loss": 0.10543316602706909,
|
1244 |
+
"eval_roc_auc": 0.8997137007632394,
|
1245 |
+
"eval_runtime": 647.0637,
|
1246 |
+
"eval_samples_per_second": 4.326,
|
1247 |
+
"eval_steps_per_second": 0.27,
|
1248 |
+
"learning_rate": 0.0001,
|
1249 |
+
"step": 34304
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 64.37,
|
1253 |
+
"learning_rate": 0.0001,
|
1254 |
+
"loss": 0.1141,
|
1255 |
+
"step": 34500
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 65.0,
|
1259 |
+
"eval_accuracy": 0.5730618077884959,
|
1260 |
+
"eval_f1_macro": 0.8099395952334975,
|
1261 |
+
"eval_f1_micro": 0.8494330240737371,
|
1262 |
+
"eval_loss": 0.10500979423522949,
|
1263 |
+
"eval_roc_auc": 0.8988602109578501,
|
1264 |
+
"eval_runtime": 649.7608,
|
1265 |
+
"eval_samples_per_second": 4.308,
|
1266 |
+
"eval_steps_per_second": 0.269,
|
1267 |
+
"learning_rate": 0.0001,
|
1268 |
+
"step": 34840
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 65.3,
|
1272 |
+
"learning_rate": 0.0001,
|
1273 |
+
"loss": 0.1104,
|
1274 |
+
"step": 35000
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 66.0,
|
1278 |
+
"eval_accuracy": 0.5762772418720972,
|
1279 |
+
"eval_f1_macro": 0.8132610218333434,
|
1280 |
+
"eval_f1_micro": 0.8506980430409072,
|
1281 |
+
"eval_loss": 0.10500740259885788,
|
1282 |
+
"eval_roc_auc": 0.8978957335797464,
|
1283 |
+
"eval_runtime": 652.9454,
|
1284 |
+
"eval_samples_per_second": 4.287,
|
1285 |
+
"eval_steps_per_second": 0.268,
|
1286 |
+
"learning_rate": 0.0001,
|
1287 |
+
"step": 35376
|
1288 |
+
},
|
1289 |
+
{
|
1290 |
+
"epoch": 66.23,
|
1291 |
+
"learning_rate": 0.0001,
|
1292 |
+
"loss": 0.1124,
|
1293 |
+
"step": 35500
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 67.0,
|
1297 |
+
"eval_accuracy": 0.5669882100750268,
|
1298 |
+
"eval_f1_macro": 0.8163481281680508,
|
1299 |
+
"eval_f1_micro": 0.8512857399748246,
|
1300 |
+
"eval_loss": 0.10600199550390244,
|
1301 |
+
"eval_roc_auc": 0.9035619176711028,
|
1302 |
+
"eval_runtime": 657.7451,
|
1303 |
+
"eval_samples_per_second": 4.255,
|
1304 |
+
"eval_steps_per_second": 0.266,
|
1305 |
+
"learning_rate": 0.0001,
|
1306 |
+
"step": 35912
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 67.16,
|
1310 |
+
"learning_rate": 0.0001,
|
1311 |
+
"loss": 0.1111,
|
1312 |
+
"step": 36000
|
1313 |
+
},
|
1314 |
+
{
|
1315 |
+
"epoch": 68.0,
|
1316 |
+
"eval_accuracy": 0.5680600214362272,
|
1317 |
+
"eval_f1_macro": 0.8156635542444352,
|
1318 |
+
"eval_f1_micro": 0.8511868899867453,
|
1319 |
+
"eval_loss": 0.10536548495292664,
|
1320 |
+
"eval_roc_auc": 0.9018856620506218,
|
1321 |
+
"eval_runtime": 656.6156,
|
1322 |
+
"eval_samples_per_second": 4.263,
|
1323 |
+
"eval_steps_per_second": 0.267,
|
1324 |
+
"learning_rate": 0.0001,
|
1325 |
+
"step": 36448
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 68.1,
|
1329 |
+
"learning_rate": 0.0001,
|
1330 |
+
"loss": 0.1097,
|
1331 |
+
"step": 36500
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 69.0,
|
1335 |
+
"eval_accuracy": 0.5673454805287603,
|
1336 |
+
"eval_f1_macro": 0.8109923908741982,
|
1337 |
+
"eval_f1_micro": 0.8500871237156763,
|
1338 |
+
"eval_loss": 0.1056470051407814,
|
1339 |
+
"eval_roc_auc": 0.9021192764421861,
|
1340 |
+
"eval_runtime": 657.5814,
|
1341 |
+
"eval_samples_per_second": 4.257,
|
1342 |
+
"eval_steps_per_second": 0.266,
|
1343 |
+
"learning_rate": 0.0001,
|
1344 |
+
"step": 36984
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 69.03,
|
1348 |
+
"learning_rate": 0.0001,
|
1349 |
+
"loss": 0.1106,
|
1350 |
+
"step": 37000
|
1351 |
+
},
|
1352 |
+
{
|
1353 |
+
"epoch": 69.96,
|
1354 |
+
"learning_rate": 0.0001,
|
1355 |
+
"loss": 0.1096,
|
1356 |
+
"step": 37500
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 70.0,
|
1360 |
+
"eval_accuracy": 0.5673454805287603,
|
1361 |
+
"eval_f1_macro": 0.8119494369118779,
|
1362 |
+
"eval_f1_micro": 0.8500514558992676,
|
1363 |
+
"eval_loss": 0.1059202253818512,
|
1364 |
+
"eval_roc_auc": 0.8997405001477684,
|
1365 |
+
"eval_runtime": 661.9686,
|
1366 |
+
"eval_samples_per_second": 4.228,
|
1367 |
+
"eval_steps_per_second": 0.264,
|
1368 |
+
"learning_rate": 0.0001,
|
1369 |
+
"step": 37520
|
1370 |
+
},
|
1371 |
+
{
|
1372 |
+
"epoch": 70.9,
|
1373 |
+
"learning_rate": 0.0001,
|
1374 |
+
"loss": 0.1097,
|
1375 |
+
"step": 38000
|
1376 |
+
},
|
1377 |
+
{
|
1378 |
+
"epoch": 71.0,
|
1379 |
+
"eval_accuracy": 0.5737763486959628,
|
1380 |
+
"eval_f1_macro": 0.8171777200998823,
|
1381 |
+
"eval_f1_micro": 0.8517297200071947,
|
1382 |
+
"eval_loss": 0.10546696186065674,
|
1383 |
+
"eval_roc_auc": 0.9037381432590107,
|
1384 |
+
"eval_runtime": 654.6273,
|
1385 |
+
"eval_samples_per_second": 4.276,
|
1386 |
+
"eval_steps_per_second": 0.267,
|
1387 |
+
"learning_rate": 0.0001,
|
1388 |
+
"step": 38056
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 71.83,
|
1392 |
+
"learning_rate": 1e-05,
|
1393 |
+
"loss": 0.1084,
|
1394 |
+
"step": 38500
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 72.0,
|
1398 |
+
"eval_accuracy": 0.5630582350839586,
|
1399 |
+
"eval_f1_macro": 0.8063147670569071,
|
1400 |
+
"eval_f1_micro": 0.84692696594993,
|
1401 |
+
"eval_loss": 0.10735420882701874,
|
1402 |
+
"eval_roc_auc": 0.8960515366190057,
|
1403 |
+
"eval_runtime": 658.0374,
|
1404 |
+
"eval_samples_per_second": 4.254,
|
1405 |
+
"eval_steps_per_second": 0.266,
|
1406 |
+
"learning_rate": 1e-05,
|
1407 |
+
"step": 38592
|
1408 |
+
},
|
1409 |
+
{
|
1410 |
+
"epoch": 72.76,
|
1411 |
+
"learning_rate": 1e-05,
|
1412 |
+
"loss": 0.1091,
|
1413 |
+
"step": 39000
|
1414 |
+
},
|
1415 |
+
{
|
1416 |
+
"epoch": 73.0,
|
1417 |
+
"eval_accuracy": 0.5734190782422294,
|
1418 |
+
"eval_f1_macro": 0.8171103344103756,
|
1419 |
+
"eval_f1_micro": 0.8525143029208069,
|
1420 |
+
"eval_loss": 0.10438621789216995,
|
1421 |
+
"eval_roc_auc": 0.9027705800247602,
|
1422 |
+
"eval_runtime": 664.8235,
|
1423 |
+
"eval_samples_per_second": 4.21,
|
1424 |
+
"eval_steps_per_second": 0.263,
|
1425 |
+
"learning_rate": 1e-05,
|
1426 |
+
"step": 39128
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 73.69,
|
1430 |
+
"learning_rate": 1e-05,
|
1431 |
+
"loss": 0.1051,
|
1432 |
+
"step": 39500
|
1433 |
+
},
|
1434 |
+
{
|
1435 |
+
"epoch": 74.0,
|
1436 |
+
"eval_accuracy": 0.571632725973562,
|
1437 |
+
"eval_f1_macro": 0.8187356068324974,
|
1438 |
+
"eval_f1_micro": 0.8532934131736527,
|
1439 |
+
"eval_loss": 0.10408420860767365,
|
1440 |
+
"eval_roc_auc": 0.9050291111780234,
|
1441 |
+
"eval_runtime": 673.5967,
|
1442 |
+
"eval_samples_per_second": 4.155,
|
1443 |
+
"eval_steps_per_second": 0.26,
|
1444 |
+
"learning_rate": 1e-05,
|
1445 |
+
"step": 39664
|
1446 |
+
},
|
1447 |
+
{
|
1448 |
+
"epoch": 74.63,
|
1449 |
+
"learning_rate": 1e-05,
|
1450 |
+
"loss": 0.1069,
|
1451 |
+
"step": 40000
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 75.0,
|
1455 |
+
"eval_accuracy": 0.5698463737048947,
|
1456 |
+
"eval_f1_macro": 0.8154818599661411,
|
1457 |
+
"eval_f1_micro": 0.8505788712011578,
|
1458 |
+
"eval_loss": 0.10555566847324371,
|
1459 |
+
"eval_roc_auc": 0.9012750442052373,
|
1460 |
+
"eval_runtime": 657.1771,
|
1461 |
+
"eval_samples_per_second": 4.259,
|
1462 |
+
"eval_steps_per_second": 0.266,
|
1463 |
+
"learning_rate": 1e-05,
|
1464 |
+
"step": 40200
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 75.56,
|
1468 |
+
"learning_rate": 1e-05,
|
1469 |
+
"loss": 0.1079,
|
1470 |
+
"step": 40500
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 76.0,
|
1474 |
+
"eval_accuracy": 0.572347266881029,
|
1475 |
+
"eval_f1_macro": 0.8153806167901401,
|
1476 |
+
"eval_f1_micro": 0.8517114840883114,
|
1477 |
+
"eval_loss": 0.10429207235574722,
|
1478 |
+
"eval_roc_auc": 0.9026594578808756,
|
1479 |
+
"eval_runtime": 664.6613,
|
1480 |
+
"eval_samples_per_second": 4.211,
|
1481 |
+
"eval_steps_per_second": 0.263,
|
1482 |
+
"learning_rate": 1e-05,
|
1483 |
+
"step": 40736
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 76.49,
|
1487 |
+
"learning_rate": 1e-05,
|
1488 |
+
"loss": 0.1072,
|
1489 |
+
"step": 41000
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 77.0,
|
1493 |
+
"eval_accuracy": 0.5780635941407646,
|
1494 |
+
"eval_f1_macro": 0.8187874152985335,
|
1495 |
+
"eval_f1_micro": 0.853607002716571,
|
1496 |
+
"eval_loss": 0.10398340970277786,
|
1497 |
+
"eval_roc_auc": 0.9026235354990767,
|
1498 |
+
"eval_runtime": 664.3977,
|
1499 |
+
"eval_samples_per_second": 4.213,
|
1500 |
+
"eval_steps_per_second": 0.263,
|
1501 |
+
"learning_rate": 1e-05,
|
1502 |
+
"step": 41272
|
1503 |
+
},
|
1504 |
+
{
|
1505 |
+
"epoch": 77.43,
|
1506 |
+
"learning_rate": 1e-05,
|
1507 |
+
"loss": 0.105,
|
1508 |
+
"step": 41500
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 78.0,
|
1512 |
+
"eval_accuracy": 0.5712754555198285,
|
1513 |
+
"eval_f1_macro": 0.8150330415154026,
|
1514 |
+
"eval_f1_micro": 0.8514409743156879,
|
1515 |
+
"eval_loss": 0.10431113094091415,
|
1516 |
+
"eval_roc_auc": 0.9018006351128146,
|
1517 |
+
"eval_runtime": 660.4258,
|
1518 |
+
"eval_samples_per_second": 4.238,
|
1519 |
+
"eval_steps_per_second": 0.265,
|
1520 |
+
"learning_rate": 1e-05,
|
1521 |
+
"step": 41808
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 78.36,
|
1525 |
+
"learning_rate": 1e-05,
|
1526 |
+
"loss": 0.1061,
|
1527 |
+
"step": 42000
|
1528 |
+
},
|
1529 |
+
{
|
1530 |
+
"epoch": 79.0,
|
1531 |
+
"eval_accuracy": 0.5734190782422294,
|
1532 |
+
"eval_f1_macro": 0.8180685346674,
|
1533 |
+
"eval_f1_micro": 0.8525934861278649,
|
1534 |
+
"eval_loss": 0.10427288711071014,
|
1535 |
+
"eval_roc_auc": 0.9023377306953573,
|
1536 |
+
"eval_runtime": 662.4252,
|
1537 |
+
"eval_samples_per_second": 4.225,
|
1538 |
+
"eval_steps_per_second": 0.264,
|
1539 |
+
"learning_rate": 1e-05,
|
1540 |
+
"step": 42344
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 79.29,
|
1544 |
+
"learning_rate": 1e-05,
|
1545 |
+
"loss": 0.1045,
|
1546 |
+
"step": 42500
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 80.0,
|
1550 |
+
"eval_accuracy": 0.5719899964272954,
|
1551 |
+
"eval_f1_macro": 0.8176554657248621,
|
1552 |
+
"eval_f1_micro": 0.8512023142288917,
|
1553 |
+
"eval_loss": 0.10507169365882874,
|
1554 |
+
"eval_roc_auc": 0.9017990921847413,
|
1555 |
+
"eval_runtime": 662.5216,
|
1556 |
+
"eval_samples_per_second": 4.225,
|
1557 |
+
"eval_steps_per_second": 0.264,
|
1558 |
+
"learning_rate": 1e-05,
|
1559 |
+
"step": 42880
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 80.22,
|
1563 |
+
"learning_rate": 1e-05,
|
1564 |
+
"loss": 0.1062,
|
1565 |
+
"step": 43000
|
1566 |
+
},
|
1567 |
+
{
|
1568 |
+
"epoch": 81.0,
|
1569 |
+
"eval_accuracy": 0.5694891032511611,
|
1570 |
+
"eval_f1_macro": 0.8157488543073257,
|
1571 |
+
"eval_f1_micro": 0.8500577402297453,
|
1572 |
+
"eval_loss": 0.10686225444078445,
|
1573 |
+
"eval_roc_auc": 0.8984886750378719,
|
1574 |
+
"eval_runtime": 655.8239,
|
1575 |
+
"eval_samples_per_second": 4.268,
|
1576 |
+
"eval_steps_per_second": 0.267,
|
1577 |
+
"learning_rate": 1e-05,
|
1578 |
+
"step": 43416
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 81.16,
|
1582 |
+
"learning_rate": 1e-05,
|
1583 |
+
"loss": 0.1057,
|
1584 |
+
"step": 43500
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 82.0,
|
1588 |
+
"eval_accuracy": 0.5734190782422294,
|
1589 |
+
"eval_f1_macro": 0.8184447896335341,
|
1590 |
+
"eval_f1_micro": 0.8530242056579975,
|
1591 |
+
"eval_loss": 0.1039666160941124,
|
1592 |
+
"eval_roc_auc": 0.9039006170654286,
|
1593 |
+
"eval_runtime": 659.9314,
|
1594 |
+
"eval_samples_per_second": 4.241,
|
1595 |
+
"eval_steps_per_second": 0.265,
|
1596 |
+
"learning_rate": 1e-05,
|
1597 |
+
"step": 43952
|
1598 |
+
},
|
1599 |
+
{
|
1600 |
+
"epoch": 82.09,
|
1601 |
+
"learning_rate": 1e-05,
|
1602 |
+
"loss": 0.1073,
|
1603 |
+
"step": 44000
|
1604 |
+
},
|
1605 |
+
{
|
1606 |
+
"epoch": 83.0,
|
1607 |
+
"eval_accuracy": 0.5712754555198285,
|
1608 |
+
"eval_f1_macro": 0.8151349204991644,
|
1609 |
+
"eval_f1_micro": 0.8503600634688149,
|
1610 |
+
"eval_loss": 0.10477207601070404,
|
1611 |
+
"eval_roc_auc": 0.8973831902752079,
|
1612 |
+
"eval_runtime": 660.9853,
|
1613 |
+
"eval_samples_per_second": 4.235,
|
1614 |
+
"eval_steps_per_second": 0.265,
|
1615 |
+
"learning_rate": 1e-05,
|
1616 |
+
"step": 44488
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 83.02,
|
1620 |
+
"learning_rate": 1e-05,
|
1621 |
+
"loss": 0.1053,
|
1622 |
+
"step": 44500
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 83.96,
|
1626 |
+
"learning_rate": 1e-05,
|
1627 |
+
"loss": 0.1059,
|
1628 |
+
"step": 45000
|
1629 |
+
},
|
1630 |
+
{
|
1631 |
+
"epoch": 84.0,
|
1632 |
+
"eval_accuracy": 0.5741336191496963,
|
1633 |
+
"eval_f1_macro": 0.8177615611039802,
|
1634 |
+
"eval_f1_micro": 0.8525637163342773,
|
1635 |
+
"eval_loss": 0.1043851226568222,
|
1636 |
+
"eval_roc_auc": 0.9026457060993854,
|
1637 |
+
"eval_runtime": 661.1385,
|
1638 |
+
"eval_samples_per_second": 4.234,
|
1639 |
+
"eval_steps_per_second": 0.265,
|
1640 |
+
"learning_rate": 1e-05,
|
1641 |
+
"step": 45024
|
1642 |
+
},
|
1643 |
+
{
|
1644 |
+
"epoch": 84.89,
|
1645 |
+
"learning_rate": 1e-05,
|
1646 |
+
"loss": 0.1054,
|
1647 |
+
"step": 45500
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 85.0,
|
1651 |
+
"eval_accuracy": 0.5702036441586281,
|
1652 |
+
"eval_f1_macro": 0.8148620661290662,
|
1653 |
+
"eval_f1_micro": 0.8505413415593056,
|
1654 |
+
"eval_loss": 0.10626183450222015,
|
1655 |
+
"eval_roc_auc": 0.9002829286929828,
|
1656 |
+
"eval_runtime": 667.6418,
|
1657 |
+
"eval_samples_per_second": 4.192,
|
1658 |
+
"eval_steps_per_second": 0.262,
|
1659 |
+
"learning_rate": 1e-05,
|
1660 |
+
"step": 45560
|
1661 |
+
},
|
1662 |
+
{
|
1663 |
+
"epoch": 85.82,
|
1664 |
+
"learning_rate": 1e-05,
|
1665 |
+
"loss": 0.1046,
|
1666 |
+
"step": 46000
|
1667 |
+
},
|
1668 |
+
{
|
1669 |
+
"epoch": 86.0,
|
1670 |
+
"eval_accuracy": 0.5727045373347625,
|
1671 |
+
"eval_f1_macro": 0.8160800717420196,
|
1672 |
+
"eval_f1_micro": 0.8525306469231703,
|
1673 |
+
"eval_loss": 0.10432148724794388,
|
1674 |
+
"eval_roc_auc": 0.9003534996708482,
|
1675 |
+
"eval_runtime": 656.1196,
|
1676 |
+
"eval_samples_per_second": 4.266,
|
1677 |
+
"eval_steps_per_second": 0.267,
|
1678 |
+
"learning_rate": 1e-05,
|
1679 |
+
"step": 46096
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 86.75,
|
1683 |
+
"learning_rate": 1e-05,
|
1684 |
+
"loss": 0.105,
|
1685 |
+
"step": 46500
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 87.0,
|
1689 |
+
"eval_accuracy": 0.5719899964272954,
|
1690 |
+
"eval_f1_macro": 0.8185429098530308,
|
1691 |
+
"eval_f1_micro": 0.8532148646214804,
|
1692 |
+
"eval_loss": 0.10474765300750732,
|
1693 |
+
"eval_roc_auc": 0.904161869591144,
|
1694 |
+
"eval_runtime": 662.4652,
|
1695 |
+
"eval_samples_per_second": 4.225,
|
1696 |
+
"eval_steps_per_second": 0.264,
|
1697 |
+
"learning_rate": 1e-05,
|
1698 |
+
"step": 46632
|
1699 |
+
},
|
1700 |
+
{
|
1701 |
+
"epoch": 87.69,
|
1702 |
+
"learning_rate": 1e-05,
|
1703 |
+
"loss": 0.1029,
|
1704 |
+
"step": 47000
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 88.0,
|
1708 |
+
"eval_accuracy": 0.5759199714183637,
|
1709 |
+
"eval_f1_macro": 0.8155503181233792,
|
1710 |
+
"eval_f1_micro": 0.8518207954000958,
|
1711 |
+
"eval_loss": 0.10432733595371246,
|
1712 |
+
"eval_roc_auc": 0.9041143079632091,
|
1713 |
+
"eval_runtime": 659.661,
|
1714 |
+
"eval_samples_per_second": 4.243,
|
1715 |
+
"eval_steps_per_second": 0.265,
|
1716 |
+
"learning_rate": 1e-05,
|
1717 |
+
"step": 47168
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 88.62,
|
1721 |
+
"learning_rate": 1.0000000000000002e-06,
|
1722 |
+
"loss": 0.1059,
|
1723 |
+
"step": 47500
|
1724 |
+
},
|
1725 |
+
{
|
1726 |
+
"epoch": 89.0,
|
1727 |
+
"eval_accuracy": 0.5787781350482315,
|
1728 |
+
"eval_f1_macro": 0.8178092875894164,
|
1729 |
+
"eval_f1_micro": 0.8538892570946557,
|
1730 |
+
"eval_loss": 0.10400809347629547,
|
1731 |
+
"eval_roc_auc": 0.9033957884012571,
|
1732 |
+
"eval_runtime": 659.0133,
|
1733 |
+
"eval_samples_per_second": 4.247,
|
1734 |
+
"eval_steps_per_second": 0.266,
|
1735 |
+
"learning_rate": 1.0000000000000002e-06,
|
1736 |
+
"step": 47704
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 89.55,
|
1740 |
+
"learning_rate": 1.0000000000000002e-06,
|
1741 |
+
"loss": 0.1047,
|
1742 |
+
"step": 48000
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 90.0,
|
1746 |
+
"eval_accuracy": 0.5694891032511611,
|
1747 |
+
"eval_f1_macro": 0.8136056007290207,
|
1748 |
+
"eval_f1_micro": 0.8504655833485485,
|
1749 |
+
"eval_loss": 0.10469033569097519,
|
1750 |
+
"eval_roc_auc": 0.8982986976684738,
|
1751 |
+
"eval_runtime": 653.5683,
|
1752 |
+
"eval_samples_per_second": 4.283,
|
1753 |
+
"eval_steps_per_second": 0.268,
|
1754 |
+
"learning_rate": 1.0000000000000002e-06,
|
1755 |
+
"step": 48240
|
1756 |
+
},
|
1757 |
+
{
|
1758 |
+
"epoch": 90.0,
|
1759 |
+
"learning_rate": 1.0000000000000002e-06,
|
1760 |
+
"step": 48240,
|
1761 |
+
"total_flos": 6.13340503159962e+19,
|
1762 |
+
"train_loss": 0.13765035268679188,
|
1763 |
+
"train_runtime": 238196.7751,
|
1764 |
+
"train_samples_per_second": 3.237,
|
1765 |
+
"train_steps_per_second": 0.203
|
1766 |
+
}
|
1767 |
+
],
|
1768 |
+
"logging_steps": 500,
|
1769 |
+
"max_steps": 48240,
|
1770 |
+
"num_train_epochs": 90,
|
1771 |
+
"save_steps": 500,
|
1772 |
+
"total_flos": 6.13340503159962e+19,
|
1773 |
+
"trial_name": null,
|
1774 |
+
"trial_params": null
|
1775 |
+
}
|