Commit
•
c2dc5e3
1
Parent(s):
1fa23ae
🍻 cheers
Browse files- README.md +10 -6
- all_results.json +16 -0
- config.json +1 -1
- eval_results.json +12 -0
- runs/Mar12_11-48-01_datavisu4/events.out.tfevents.1710311640.datavisu4.62295.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +779 -0
README.md
CHANGED
|
@@ -1,7 +1,11 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: microsoft/resnet-50
|
| 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 |
# resnet-50-linearhead-2024_03_12-with_data_aug_batch-size32_epochs93_freeze
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [microsoft/resnet-50](https://huggingface.co/microsoft/resnet-50) 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: 1e-05
|
| 26 |
|
| 27 |
## Model description
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- eng
|
| 4 |
license: apache-2.0
|
| 5 |
base_model: microsoft/resnet-50
|
| 6 |
tags:
|
| 7 |
+
- multilabel-image-classification
|
| 8 |
+
- multilabel
|
| 9 |
- generated_from_trainer
|
| 10 |
metrics:
|
| 11 |
- accuracy
|
|
|
|
| 19 |
|
| 20 |
# resnet-50-linearhead-2024_03_12-with_data_aug_batch-size32_epochs93_freeze
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [microsoft/resnet-50](https://huggingface.co/microsoft/resnet-50) on the multilabel_complete_dataset dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.1518
|
| 25 |
+
- F1 Micro: 0.7545
|
| 26 |
+
- F1 Macro: 0.6309
|
| 27 |
+
- Roc Auc: 0.8276
|
| 28 |
+
- Accuracy: 0.4069
|
| 29 |
- Learning Rate: 1e-05
|
| 30 |
|
| 31 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"eval_accuracy": 0.40687478077867417,
|
| 4 |
+
"eval_f1_macro": 0.6308665357852737,
|
| 5 |
+
"eval_f1_micro": 0.7544523735650016,
|
| 6 |
+
"eval_loss": 0.15177054703235626,
|
| 7 |
+
"eval_roc_auc": 0.8275818899442217,
|
| 8 |
+
"eval_runtime": 368.9147,
|
| 9 |
+
"eval_samples_per_second": 7.728,
|
| 10 |
+
"eval_steps_per_second": 0.244,
|
| 11 |
+
"learning_rate": 1e-05,
|
| 12 |
+
"train_loss": 0.1827572128274107,
|
| 13 |
+
"train_runtime": 70769.3212,
|
| 14 |
+
"train_samples_per_second": 11.517,
|
| 15 |
+
"train_steps_per_second": 0.36
|
| 16 |
+
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "microsoft/resnet-
|
| 3 |
"architectures": [
|
| 4 |
"ResNetForImageClassification"
|
| 5 |
],
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "microsoft/resnet-502024_03_12",
|
| 3 |
"architectures": [
|
| 4 |
"ResNetForImageClassification"
|
| 5 |
],
|
eval_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"eval_accuracy": 0.40687478077867417,
|
| 4 |
+
"eval_f1_macro": 0.6308665357852737,
|
| 5 |
+
"eval_f1_micro": 0.7544523735650016,
|
| 6 |
+
"eval_loss": 0.15177054703235626,
|
| 7 |
+
"eval_roc_auc": 0.8275818899442217,
|
| 8 |
+
"eval_runtime": 368.9147,
|
| 9 |
+
"eval_samples_per_second": 7.728,
|
| 10 |
+
"eval_steps_per_second": 0.244,
|
| 11 |
+
"learning_rate": 1e-05
|
| 12 |
+
}
|
runs/Mar12_11-48-01_datavisu4/events.out.tfevents.1710311640.datavisu4.62295.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1a62462d2e0eea6e5dfabdb1d9f3e70881b3db0f6c29348f071465f035653cb
|
| 3 |
+
size 624
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"learning_rate": 1e-05,
|
| 4 |
+
"train_loss": 0.1827572128274107,
|
| 5 |
+
"train_runtime": 70769.3212,
|
| 6 |
+
"train_samples_per_second": 11.517,
|
| 7 |
+
"train_steps_per_second": 0.36
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,779 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.15303632616996765,
|
| 3 |
+
"best_model_checkpoint": "/home1/datawork/mcontini/models/multilabel/huggingface/resnet-50-linearhead-2024_03_12-with_data_aug_batch-size32_epochs93_freeze/checkpoint-9864",
|
| 4 |
+
"epoch": 46.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 12604,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.1951983298538622,
|
| 14 |
+
"eval_f1_macro": 0.2834407731254271,
|
| 15 |
+
"eval_f1_micro": 0.5838964773544213,
|
| 16 |
+
"eval_loss": 0.22367511689662933,
|
| 17 |
+
"eval_roc_auc": 0.71756579506086,
|
| 18 |
+
"eval_runtime": 377.3279,
|
| 19 |
+
"eval_samples_per_second": 7.617,
|
| 20 |
+
"eval_steps_per_second": 0.239,
|
| 21 |
+
"learning_rate": 0.001,
|
| 22 |
+
"step": 274
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 1.82,
|
| 26 |
+
"learning_rate": 0.001,
|
| 27 |
+
"loss": 0.2683,
|
| 28 |
+
"step": 500
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_accuracy": 0.30549756437021575,
|
| 33 |
+
"eval_f1_macro": 0.4548607736351322,
|
| 34 |
+
"eval_f1_micro": 0.67727182342414,
|
| 35 |
+
"eval_loss": 0.18945501744747162,
|
| 36 |
+
"eval_roc_auc": 0.7743387387654568,
|
| 37 |
+
"eval_runtime": 387.1639,
|
| 38 |
+
"eval_samples_per_second": 7.423,
|
| 39 |
+
"eval_steps_per_second": 0.232,
|
| 40 |
+
"learning_rate": 0.001,
|
| 41 |
+
"step": 548
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"eval_accuracy": 0.35386221294363257,
|
| 46 |
+
"eval_f1_macro": 0.5202190109990653,
|
| 47 |
+
"eval_f1_micro": 0.7020794195763426,
|
| 48 |
+
"eval_loss": 0.17864234745502472,
|
| 49 |
+
"eval_roc_auc": 0.7911168944948984,
|
| 50 |
+
"eval_runtime": 391.2444,
|
| 51 |
+
"eval_samples_per_second": 7.346,
|
| 52 |
+
"eval_steps_per_second": 0.23,
|
| 53 |
+
"learning_rate": 0.001,
|
| 54 |
+
"step": 822
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.65,
|
| 58 |
+
"learning_rate": 0.001,
|
| 59 |
+
"loss": 0.2058,
|
| 60 |
+
"step": 1000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"eval_accuracy": 0.36673625608907445,
|
| 65 |
+
"eval_f1_macro": 0.5666424404021909,
|
| 66 |
+
"eval_f1_micro": 0.7198384654215043,
|
| 67 |
+
"eval_loss": 0.17152228951454163,
|
| 68 |
+
"eval_roc_auc": 0.8057974071888091,
|
| 69 |
+
"eval_runtime": 384.9412,
|
| 70 |
+
"eval_samples_per_second": 7.466,
|
| 71 |
+
"eval_steps_per_second": 0.234,
|
| 72 |
+
"learning_rate": 0.001,
|
| 73 |
+
"step": 1096
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 5.0,
|
| 77 |
+
"eval_accuracy": 0.3768267223382046,
|
| 78 |
+
"eval_f1_macro": 0.5718368144674506,
|
| 79 |
+
"eval_f1_micro": 0.7220207418718584,
|
| 80 |
+
"eval_loss": 0.16623608767986298,
|
| 81 |
+
"eval_roc_auc": 0.8049842310066343,
|
| 82 |
+
"eval_runtime": 384.6479,
|
| 83 |
+
"eval_samples_per_second": 7.472,
|
| 84 |
+
"eval_steps_per_second": 0.234,
|
| 85 |
+
"learning_rate": 0.001,
|
| 86 |
+
"step": 1370
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 5.47,
|
| 90 |
+
"learning_rate": 0.001,
|
| 91 |
+
"loss": 0.1916,
|
| 92 |
+
"step": 1500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 6.0,
|
| 96 |
+
"eval_accuracy": 0.37961029923451634,
|
| 97 |
+
"eval_f1_macro": 0.5721081100676888,
|
| 98 |
+
"eval_f1_micro": 0.715496603041087,
|
| 99 |
+
"eval_loss": 0.16475693881511688,
|
| 100 |
+
"eval_roc_auc": 0.7979565909429278,
|
| 101 |
+
"eval_runtime": 379.9116,
|
| 102 |
+
"eval_samples_per_second": 7.565,
|
| 103 |
+
"eval_steps_per_second": 0.237,
|
| 104 |
+
"learning_rate": 0.001,
|
| 105 |
+
"step": 1644
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"eval_accuracy": 0.38100208768267224,
|
| 110 |
+
"eval_f1_macro": 0.5973146797781492,
|
| 111 |
+
"eval_f1_micro": 0.7280701754385965,
|
| 112 |
+
"eval_loss": 0.16181063652038574,
|
| 113 |
+
"eval_roc_auc": 0.808226921681461,
|
| 114 |
+
"eval_runtime": 378.5774,
|
| 115 |
+
"eval_samples_per_second": 7.592,
|
| 116 |
+
"eval_steps_per_second": 0.238,
|
| 117 |
+
"learning_rate": 0.001,
|
| 118 |
+
"step": 1918
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 7.3,
|
| 122 |
+
"learning_rate": 0.001,
|
| 123 |
+
"loss": 0.1858,
|
| 124 |
+
"step": 2000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 8.0,
|
| 128 |
+
"eval_accuracy": 0.38552540013917885,
|
| 129 |
+
"eval_f1_macro": 0.6061015606132648,
|
| 130 |
+
"eval_f1_micro": 0.737494522005885,
|
| 131 |
+
"eval_loss": 0.15984570980072021,
|
| 132 |
+
"eval_roc_auc": 0.8165736033401827,
|
| 133 |
+
"eval_runtime": 373.9089,
|
| 134 |
+
"eval_samples_per_second": 7.686,
|
| 135 |
+
"eval_steps_per_second": 0.241,
|
| 136 |
+
"learning_rate": 0.001,
|
| 137 |
+
"step": 2192
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"eval_accuracy": 0.39109255393180237,
|
| 142 |
+
"eval_f1_macro": 0.6209440717973264,
|
| 143 |
+
"eval_f1_micro": 0.7440332279461905,
|
| 144 |
+
"eval_loss": 0.15988709032535553,
|
| 145 |
+
"eval_roc_auc": 0.8223085810562136,
|
| 146 |
+
"eval_runtime": 382.5296,
|
| 147 |
+
"eval_samples_per_second": 7.513,
|
| 148 |
+
"eval_steps_per_second": 0.235,
|
| 149 |
+
"learning_rate": 0.001,
|
| 150 |
+
"step": 2466
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.12,
|
| 154 |
+
"learning_rate": 0.001,
|
| 155 |
+
"loss": 0.1839,
|
| 156 |
+
"step": 2500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 10.0,
|
| 160 |
+
"eval_accuracy": 0.394919972164231,
|
| 161 |
+
"eval_f1_macro": 0.6047260524666048,
|
| 162 |
+
"eval_f1_micro": 0.7382172771596449,
|
| 163 |
+
"eval_loss": 0.15841498970985413,
|
| 164 |
+
"eval_roc_auc": 0.8173181533125079,
|
| 165 |
+
"eval_runtime": 384.4811,
|
| 166 |
+
"eval_samples_per_second": 7.475,
|
| 167 |
+
"eval_steps_per_second": 0.234,
|
| 168 |
+
"learning_rate": 0.001,
|
| 169 |
+
"step": 2740
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 10.95,
|
| 173 |
+
"learning_rate": 0.001,
|
| 174 |
+
"loss": 0.1815,
|
| 175 |
+
"step": 3000
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 11.0,
|
| 179 |
+
"eval_accuracy": 0.3959638135003479,
|
| 180 |
+
"eval_f1_macro": 0.6067885073789792,
|
| 181 |
+
"eval_f1_micro": 0.7413944173238203,
|
| 182 |
+
"eval_loss": 0.15687702596187592,
|
| 183 |
+
"eval_roc_auc": 0.8186288501697149,
|
| 184 |
+
"eval_runtime": 378.7669,
|
| 185 |
+
"eval_samples_per_second": 7.588,
|
| 186 |
+
"eval_steps_per_second": 0.238,
|
| 187 |
+
"learning_rate": 0.001,
|
| 188 |
+
"step": 3014
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 12.0,
|
| 192 |
+
"eval_accuracy": 0.39631176061238693,
|
| 193 |
+
"eval_f1_macro": 0.5952620618269702,
|
| 194 |
+
"eval_f1_micro": 0.7257172262961533,
|
| 195 |
+
"eval_loss": 0.15848152339458466,
|
| 196 |
+
"eval_roc_auc": 0.8043142106532862,
|
| 197 |
+
"eval_runtime": 379.0677,
|
| 198 |
+
"eval_samples_per_second": 7.582,
|
| 199 |
+
"eval_steps_per_second": 0.237,
|
| 200 |
+
"learning_rate": 0.001,
|
| 201 |
+
"step": 3288
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 12.77,
|
| 205 |
+
"learning_rate": 0.001,
|
| 206 |
+
"loss": 0.1807,
|
| 207 |
+
"step": 3500
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 13.0,
|
| 211 |
+
"eval_accuracy": 0.3966597077244259,
|
| 212 |
+
"eval_f1_macro": 0.6286378372829055,
|
| 213 |
+
"eval_f1_micro": 0.7513671162960261,
|
| 214 |
+
"eval_loss": 0.1580551415681839,
|
| 215 |
+
"eval_roc_auc": 0.8311449619256275,
|
| 216 |
+
"eval_runtime": 381.6295,
|
| 217 |
+
"eval_samples_per_second": 7.531,
|
| 218 |
+
"eval_steps_per_second": 0.236,
|
| 219 |
+
"learning_rate": 0.001,
|
| 220 |
+
"step": 3562
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 14.0,
|
| 224 |
+
"eval_accuracy": 0.4022268615170494,
|
| 225 |
+
"eval_f1_macro": 0.6230252873319017,
|
| 226 |
+
"eval_f1_micro": 0.7452783300198808,
|
| 227 |
+
"eval_loss": 0.15653984248638153,
|
| 228 |
+
"eval_roc_auc": 0.8224100131959987,
|
| 229 |
+
"eval_runtime": 376.7587,
|
| 230 |
+
"eval_samples_per_second": 7.628,
|
| 231 |
+
"eval_steps_per_second": 0.239,
|
| 232 |
+
"learning_rate": 0.001,
|
| 233 |
+
"step": 3836
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.6,
|
| 237 |
+
"learning_rate": 0.001,
|
| 238 |
+
"loss": 0.1795,
|
| 239 |
+
"step": 4000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 15.0,
|
| 243 |
+
"eval_accuracy": 0.3990953375086987,
|
| 244 |
+
"eval_f1_macro": 0.6252632316687944,
|
| 245 |
+
"eval_f1_micro": 0.750355574794385,
|
| 246 |
+
"eval_loss": 0.1549140363931656,
|
| 247 |
+
"eval_roc_auc": 0.8261687325810277,
|
| 248 |
+
"eval_runtime": 378.8075,
|
| 249 |
+
"eval_samples_per_second": 7.587,
|
| 250 |
+
"eval_steps_per_second": 0.238,
|
| 251 |
+
"learning_rate": 0.001,
|
| 252 |
+
"step": 4110
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 16.0,
|
| 256 |
+
"eval_accuracy": 0.3938761308281141,
|
| 257 |
+
"eval_f1_macro": 0.6290292313985378,
|
| 258 |
+
"eval_f1_micro": 0.7446265030216187,
|
| 259 |
+
"eval_loss": 0.15725594758987427,
|
| 260 |
+
"eval_roc_auc": 0.8213875460516582,
|
| 261 |
+
"eval_runtime": 387.9526,
|
| 262 |
+
"eval_samples_per_second": 7.408,
|
| 263 |
+
"eval_steps_per_second": 0.232,
|
| 264 |
+
"learning_rate": 0.001,
|
| 265 |
+
"step": 4384
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 16.42,
|
| 269 |
+
"learning_rate": 0.001,
|
| 270 |
+
"loss": 0.178,
|
| 271 |
+
"step": 4500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 17.0,
|
| 275 |
+
"eval_accuracy": 0.4025748086290884,
|
| 276 |
+
"eval_f1_macro": 0.6286656277196884,
|
| 277 |
+
"eval_f1_micro": 0.7519389388157086,
|
| 278 |
+
"eval_loss": 0.15514959394931793,
|
| 279 |
+
"eval_roc_auc": 0.8281482042533274,
|
| 280 |
+
"eval_runtime": 380.4468,
|
| 281 |
+
"eval_samples_per_second": 7.554,
|
| 282 |
+
"eval_steps_per_second": 0.237,
|
| 283 |
+
"learning_rate": 0.001,
|
| 284 |
+
"step": 4658
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 18.0,
|
| 288 |
+
"eval_accuracy": 0.3914405010438413,
|
| 289 |
+
"eval_f1_macro": 0.6155101525089491,
|
| 290 |
+
"eval_f1_micro": 0.7430460271922166,
|
| 291 |
+
"eval_loss": 0.15699204802513123,
|
| 292 |
+
"eval_roc_auc": 0.8203220599321907,
|
| 293 |
+
"eval_runtime": 379.1064,
|
| 294 |
+
"eval_samples_per_second": 7.581,
|
| 295 |
+
"eval_steps_per_second": 0.237,
|
| 296 |
+
"learning_rate": 0.001,
|
| 297 |
+
"step": 4932
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 18.25,
|
| 301 |
+
"learning_rate": 0.001,
|
| 302 |
+
"loss": 0.1764,
|
| 303 |
+
"step": 5000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 19.0,
|
| 307 |
+
"eval_accuracy": 0.3990953375086987,
|
| 308 |
+
"eval_f1_macro": 0.6286877926596727,
|
| 309 |
+
"eval_f1_micro": 0.7480261112838048,
|
| 310 |
+
"eval_loss": 0.15578077733516693,
|
| 311 |
+
"eval_roc_auc": 0.8236456074129308,
|
| 312 |
+
"eval_runtime": 380.5374,
|
| 313 |
+
"eval_samples_per_second": 7.552,
|
| 314 |
+
"eval_steps_per_second": 0.237,
|
| 315 |
+
"learning_rate": 0.001,
|
| 316 |
+
"step": 5206
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 20.0,
|
| 320 |
+
"eval_accuracy": 0.4001391788448156,
|
| 321 |
+
"eval_f1_macro": 0.608545078452072,
|
| 322 |
+
"eval_f1_micro": 0.7402785655763534,
|
| 323 |
+
"eval_loss": 0.15735512971878052,
|
| 324 |
+
"eval_roc_auc": 0.8163778880079953,
|
| 325 |
+
"eval_runtime": 371.4966,
|
| 326 |
+
"eval_samples_per_second": 7.736,
|
| 327 |
+
"eval_steps_per_second": 0.242,
|
| 328 |
+
"learning_rate": 0.001,
|
| 329 |
+
"step": 5480
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.07,
|
| 333 |
+
"learning_rate": 0.001,
|
| 334 |
+
"loss": 0.1775,
|
| 335 |
+
"step": 5500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 21.0,
|
| 339 |
+
"eval_accuracy": 0.40292275574112735,
|
| 340 |
+
"eval_f1_macro": 0.6245521933830414,
|
| 341 |
+
"eval_f1_micro": 0.753183153770813,
|
| 342 |
+
"eval_loss": 0.15607939660549164,
|
| 343 |
+
"eval_roc_auc": 0.8301931385420087,
|
| 344 |
+
"eval_runtime": 382.1339,
|
| 345 |
+
"eval_samples_per_second": 7.521,
|
| 346 |
+
"eval_steps_per_second": 0.236,
|
| 347 |
+
"learning_rate": 0.001,
|
| 348 |
+
"step": 5754
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 21.9,
|
| 352 |
+
"learning_rate": 0.0001,
|
| 353 |
+
"loss": 0.177,
|
| 354 |
+
"step": 6000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 22.0,
|
| 358 |
+
"eval_accuracy": 0.39735560194850383,
|
| 359 |
+
"eval_f1_macro": 0.6430549553924351,
|
| 360 |
+
"eval_f1_micro": 0.7596049620619053,
|
| 361 |
+
"eval_loss": 0.1545252650976181,
|
| 362 |
+
"eval_roc_auc": 0.8377894663149374,
|
| 363 |
+
"eval_runtime": 374.7337,
|
| 364 |
+
"eval_samples_per_second": 7.669,
|
| 365 |
+
"eval_steps_per_second": 0.24,
|
| 366 |
+
"learning_rate": 0.0001,
|
| 367 |
+
"step": 6028
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 23.0,
|
| 371 |
+
"eval_accuracy": 0.4025748086290884,
|
| 372 |
+
"eval_f1_macro": 0.6291738798521619,
|
| 373 |
+
"eval_f1_micro": 0.7472035794183445,
|
| 374 |
+
"eval_loss": 0.1555616706609726,
|
| 375 |
+
"eval_roc_auc": 0.8233330160035968,
|
| 376 |
+
"eval_runtime": 372.3825,
|
| 377 |
+
"eval_samples_per_second": 7.718,
|
| 378 |
+
"eval_steps_per_second": 0.242,
|
| 379 |
+
"learning_rate": 0.0001,
|
| 380 |
+
"step": 6302
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 23.72,
|
| 384 |
+
"learning_rate": 0.0001,
|
| 385 |
+
"loss": 0.1762,
|
| 386 |
+
"step": 6500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 24.0,
|
| 390 |
+
"eval_accuracy": 0.39944328462073764,
|
| 391 |
+
"eval_f1_macro": 0.6343316351234961,
|
| 392 |
+
"eval_f1_micro": 0.752788906009245,
|
| 393 |
+
"eval_loss": 0.1547899842262268,
|
| 394 |
+
"eval_roc_auc": 0.8282598991861336,
|
| 395 |
+
"eval_runtime": 369.1278,
|
| 396 |
+
"eval_samples_per_second": 7.786,
|
| 397 |
+
"eval_steps_per_second": 0.244,
|
| 398 |
+
"learning_rate": 0.0001,
|
| 399 |
+
"step": 6576
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 25.0,
|
| 403 |
+
"eval_accuracy": 0.39944328462073764,
|
| 404 |
+
"eval_f1_macro": 0.6224788217146914,
|
| 405 |
+
"eval_f1_micro": 0.7467564870259481,
|
| 406 |
+
"eval_loss": 0.15538595616817474,
|
| 407 |
+
"eval_roc_auc": 0.8221815416817054,
|
| 408 |
+
"eval_runtime": 366.9271,
|
| 409 |
+
"eval_samples_per_second": 7.833,
|
| 410 |
+
"eval_steps_per_second": 0.245,
|
| 411 |
+
"learning_rate": 0.0001,
|
| 412 |
+
"step": 6850
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 25.55,
|
| 416 |
+
"learning_rate": 0.0001,
|
| 417 |
+
"loss": 0.1759,
|
| 418 |
+
"step": 7000
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 26.0,
|
| 422 |
+
"eval_accuracy": 0.3977035490605428,
|
| 423 |
+
"eval_f1_macro": 0.632595682527143,
|
| 424 |
+
"eval_f1_micro": 0.7529426189308485,
|
| 425 |
+
"eval_loss": 0.1547958254814148,
|
| 426 |
+
"eval_roc_auc": 0.8296944220765509,
|
| 427 |
+
"eval_runtime": 367.9424,
|
| 428 |
+
"eval_samples_per_second": 7.811,
|
| 429 |
+
"eval_steps_per_second": 0.245,
|
| 430 |
+
"learning_rate": 0.0001,
|
| 431 |
+
"step": 7124
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 27.0,
|
| 435 |
+
"eval_accuracy": 0.3970076548364649,
|
| 436 |
+
"eval_f1_macro": 0.6352159754424679,
|
| 437 |
+
"eval_f1_micro": 0.7516299667855824,
|
| 438 |
+
"eval_loss": 0.1552300751209259,
|
| 439 |
+
"eval_roc_auc": 0.8281751565219652,
|
| 440 |
+
"eval_runtime": 367.5636,
|
| 441 |
+
"eval_samples_per_second": 7.819,
|
| 442 |
+
"eval_steps_per_second": 0.245,
|
| 443 |
+
"learning_rate": 0.0001,
|
| 444 |
+
"step": 7398
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 27.37,
|
| 448 |
+
"learning_rate": 0.0001,
|
| 449 |
+
"loss": 0.1752,
|
| 450 |
+
"step": 7500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 28.0,
|
| 454 |
+
"eval_accuracy": 0.4091858037578288,
|
| 455 |
+
"eval_f1_macro": 0.6327633387519097,
|
| 456 |
+
"eval_f1_micro": 0.7523139577934099,
|
| 457 |
+
"eval_loss": 0.1543245166540146,
|
| 458 |
+
"eval_roc_auc": 0.8277194590316342,
|
| 459 |
+
"eval_runtime": 363.6815,
|
| 460 |
+
"eval_samples_per_second": 7.903,
|
| 461 |
+
"eval_steps_per_second": 0.247,
|
| 462 |
+
"learning_rate": 0.0001,
|
| 463 |
+
"step": 7672
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 29.0,
|
| 467 |
+
"eval_accuracy": 0.40187891440501045,
|
| 468 |
+
"eval_f1_macro": 0.6311913970379929,
|
| 469 |
+
"eval_f1_micro": 0.7505715168365771,
|
| 470 |
+
"eval_loss": 0.15445660054683685,
|
| 471 |
+
"eval_roc_auc": 0.8264967181798933,
|
| 472 |
+
"eval_runtime": 370.9747,
|
| 473 |
+
"eval_samples_per_second": 7.747,
|
| 474 |
+
"eval_steps_per_second": 0.243,
|
| 475 |
+
"learning_rate": 0.0001,
|
| 476 |
+
"step": 7946
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 29.2,
|
| 480 |
+
"learning_rate": 0.0001,
|
| 481 |
+
"loss": 0.1757,
|
| 482 |
+
"step": 8000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 30.0,
|
| 486 |
+
"eval_accuracy": 0.40396659707724425,
|
| 487 |
+
"eval_f1_macro": 0.6394187347167111,
|
| 488 |
+
"eval_f1_micro": 0.7554370872962985,
|
| 489 |
+
"eval_loss": 0.15497015416622162,
|
| 490 |
+
"eval_roc_auc": 0.834040386953802,
|
| 491 |
+
"eval_runtime": 370.7394,
|
| 492 |
+
"eval_samples_per_second": 7.752,
|
| 493 |
+
"eval_steps_per_second": 0.243,
|
| 494 |
+
"learning_rate": 0.0001,
|
| 495 |
+
"step": 8220
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 31.0,
|
| 499 |
+
"eval_accuracy": 0.4022268615170494,
|
| 500 |
+
"eval_f1_macro": 0.6345022832926616,
|
| 501 |
+
"eval_f1_micro": 0.7511534912334668,
|
| 502 |
+
"eval_loss": 0.15540161728858948,
|
| 503 |
+
"eval_roc_auc": 0.8278824679295457,
|
| 504 |
+
"eval_runtime": 373.8253,
|
| 505 |
+
"eval_samples_per_second": 7.688,
|
| 506 |
+
"eval_steps_per_second": 0.241,
|
| 507 |
+
"learning_rate": 0.0001,
|
| 508 |
+
"step": 8494
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 31.02,
|
| 512 |
+
"learning_rate": 0.0001,
|
| 513 |
+
"loss": 0.1758,
|
| 514 |
+
"step": 8500
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 32.0,
|
| 518 |
+
"eval_accuracy": 0.4032707028531663,
|
| 519 |
+
"eval_f1_macro": 0.6301888508291651,
|
| 520 |
+
"eval_f1_micro": 0.7513253606213783,
|
| 521 |
+
"eval_loss": 0.15450650453567505,
|
| 522 |
+
"eval_roc_auc": 0.8274588541043629,
|
| 523 |
+
"eval_runtime": 366.8538,
|
| 524 |
+
"eval_samples_per_second": 7.834,
|
| 525 |
+
"eval_steps_per_second": 0.245,
|
| 526 |
+
"learning_rate": 0.0001,
|
| 527 |
+
"step": 8768
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 32.85,
|
| 531 |
+
"learning_rate": 0.0001,
|
| 532 |
+
"loss": 0.1755,
|
| 533 |
+
"step": 9000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 33.0,
|
| 537 |
+
"eval_accuracy": 0.3977035490605428,
|
| 538 |
+
"eval_f1_macro": 0.6261364069131695,
|
| 539 |
+
"eval_f1_micro": 0.7456298600311042,
|
| 540 |
+
"eval_loss": 0.15546494722366333,
|
| 541 |
+
"eval_roc_auc": 0.8222675298841294,
|
| 542 |
+
"eval_runtime": 372.6997,
|
| 543 |
+
"eval_samples_per_second": 7.711,
|
| 544 |
+
"eval_steps_per_second": 0.241,
|
| 545 |
+
"learning_rate": 0.0001,
|
| 546 |
+
"step": 9042
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 34.0,
|
| 550 |
+
"eval_accuracy": 0.41092553931802367,
|
| 551 |
+
"eval_f1_macro": 0.6307477868895636,
|
| 552 |
+
"eval_f1_micro": 0.7514572739675058,
|
| 553 |
+
"eval_loss": 0.15329033136367798,
|
| 554 |
+
"eval_roc_auc": 0.8260358892930517,
|
| 555 |
+
"eval_runtime": 370.3661,
|
| 556 |
+
"eval_samples_per_second": 7.76,
|
| 557 |
+
"eval_steps_per_second": 0.243,
|
| 558 |
+
"learning_rate": 0.0001,
|
| 559 |
+
"step": 9316
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 34.67,
|
| 563 |
+
"learning_rate": 0.0001,
|
| 564 |
+
"loss": 0.1752,
|
| 565 |
+
"step": 9500
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 35.0,
|
| 569 |
+
"eval_accuracy": 0.40535838552540016,
|
| 570 |
+
"eval_f1_macro": 0.6325032556817062,
|
| 571 |
+
"eval_f1_micro": 0.7506499938095826,
|
| 572 |
+
"eval_loss": 0.15512260794639587,
|
| 573 |
+
"eval_roc_auc": 0.8260564148790938,
|
| 574 |
+
"eval_runtime": 369.1819,
|
| 575 |
+
"eval_samples_per_second": 7.785,
|
| 576 |
+
"eval_steps_per_second": 0.244,
|
| 577 |
+
"learning_rate": 0.0001,
|
| 578 |
+
"step": 9590
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 36.0,
|
| 582 |
+
"eval_accuracy": 0.4025748086290884,
|
| 583 |
+
"eval_f1_macro": 0.6299014754104307,
|
| 584 |
+
"eval_f1_micro": 0.7539002281556392,
|
| 585 |
+
"eval_loss": 0.15303632616996765,
|
| 586 |
+
"eval_roc_auc": 0.8287079244556137,
|
| 587 |
+
"eval_runtime": 369.3754,
|
| 588 |
+
"eval_samples_per_second": 7.781,
|
| 589 |
+
"eval_steps_per_second": 0.244,
|
| 590 |
+
"learning_rate": 0.0001,
|
| 591 |
+
"step": 9864
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 36.5,
|
| 595 |
+
"learning_rate": 0.0001,
|
| 596 |
+
"loss": 0.1752,
|
| 597 |
+
"step": 10000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 37.0,
|
| 601 |
+
"eval_accuracy": 0.4036186499652053,
|
| 602 |
+
"eval_f1_macro": 0.6270201439545436,
|
| 603 |
+
"eval_f1_micro": 0.7464025415810128,
|
| 604 |
+
"eval_loss": 0.1545841097831726,
|
| 605 |
+
"eval_roc_auc": 0.8223240249935749,
|
| 606 |
+
"eval_runtime": 381.9233,
|
| 607 |
+
"eval_samples_per_second": 7.525,
|
| 608 |
+
"eval_steps_per_second": 0.236,
|
| 609 |
+
"learning_rate": 0.0001,
|
| 610 |
+
"step": 10138
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 38.0,
|
| 614 |
+
"eval_accuracy": 0.3987473903966597,
|
| 615 |
+
"eval_f1_macro": 0.6363851637994423,
|
| 616 |
+
"eval_f1_micro": 0.7538583541755627,
|
| 617 |
+
"eval_loss": 0.1548989862203598,
|
| 618 |
+
"eval_roc_auc": 0.8314331916384938,
|
| 619 |
+
"eval_runtime": 372.9722,
|
| 620 |
+
"eval_samples_per_second": 7.706,
|
| 621 |
+
"eval_steps_per_second": 0.241,
|
| 622 |
+
"learning_rate": 0.0001,
|
| 623 |
+
"step": 10412
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 38.32,
|
| 627 |
+
"learning_rate": 0.0001,
|
| 628 |
+
"loss": 0.1763,
|
| 629 |
+
"step": 10500
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 39.0,
|
| 633 |
+
"eval_accuracy": 0.3977035490605428,
|
| 634 |
+
"eval_f1_macro": 0.6420887435658987,
|
| 635 |
+
"eval_f1_micro": 0.7579290763003684,
|
| 636 |
+
"eval_loss": 0.15474249422550201,
|
| 637 |
+
"eval_roc_auc": 0.8360712223390366,
|
| 638 |
+
"eval_runtime": 367.6796,
|
| 639 |
+
"eval_samples_per_second": 7.817,
|
| 640 |
+
"eval_steps_per_second": 0.245,
|
| 641 |
+
"learning_rate": 0.0001,
|
| 642 |
+
"step": 10686
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 40.0,
|
| 646 |
+
"eval_accuracy": 0.40048712595685454,
|
| 647 |
+
"eval_f1_macro": 0.6344591092641424,
|
| 648 |
+
"eval_f1_micro": 0.7539234919077978,
|
| 649 |
+
"eval_loss": 0.15435411036014557,
|
| 650 |
+
"eval_roc_auc": 0.8302027785659021,
|
| 651 |
+
"eval_runtime": 366.3255,
|
| 652 |
+
"eval_samples_per_second": 7.845,
|
| 653 |
+
"eval_steps_per_second": 0.246,
|
| 654 |
+
"learning_rate": 0.0001,
|
| 655 |
+
"step": 10960
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 40.15,
|
| 659 |
+
"learning_rate": 0.0001,
|
| 660 |
+
"loss": 0.176,
|
| 661 |
+
"step": 11000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 41.0,
|
| 665 |
+
"eval_accuracy": 0.40153096729297144,
|
| 666 |
+
"eval_f1_macro": 0.6347263815695223,
|
| 667 |
+
"eval_f1_micro": 0.7535591556210112,
|
| 668 |
+
"eval_loss": 0.15571445226669312,
|
| 669 |
+
"eval_roc_auc": 0.8297643933203154,
|
| 670 |
+
"eval_runtime": 368.1721,
|
| 671 |
+
"eval_samples_per_second": 7.806,
|
| 672 |
+
"eval_steps_per_second": 0.244,
|
| 673 |
+
"learning_rate": 0.0001,
|
| 674 |
+
"step": 11234
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 41.97,
|
| 678 |
+
"learning_rate": 0.0001,
|
| 679 |
+
"loss": 0.1758,
|
| 680 |
+
"step": 11500
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 42.0,
|
| 684 |
+
"eval_accuracy": 0.3959638135003479,
|
| 685 |
+
"eval_f1_macro": 0.627716108835686,
|
| 686 |
+
"eval_f1_micro": 0.7474281438992456,
|
| 687 |
+
"eval_loss": 0.15399669110774994,
|
| 688 |
+
"eval_roc_auc": 0.8226314850063985,
|
| 689 |
+
"eval_runtime": 371.1239,
|
| 690 |
+
"eval_samples_per_second": 7.744,
|
| 691 |
+
"eval_steps_per_second": 0.243,
|
| 692 |
+
"learning_rate": 0.0001,
|
| 693 |
+
"step": 11508
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 43.0,
|
| 697 |
+
"eval_accuracy": 0.3970076548364649,
|
| 698 |
+
"eval_f1_macro": 0.6383655319176039,
|
| 699 |
+
"eval_f1_micro": 0.7577960704199963,
|
| 700 |
+
"eval_loss": 0.1547509878873825,
|
| 701 |
+
"eval_roc_auc": 0.83741716645486,
|
| 702 |
+
"eval_runtime": 365.888,
|
| 703 |
+
"eval_samples_per_second": 7.855,
|
| 704 |
+
"eval_steps_per_second": 0.246,
|
| 705 |
+
"learning_rate": 1e-05,
|
| 706 |
+
"step": 11782
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 43.8,
|
| 710 |
+
"learning_rate": 1e-05,
|
| 711 |
+
"loss": 0.1764,
|
| 712 |
+
"step": 12000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 44.0,
|
| 716 |
+
"eval_accuracy": 0.4011830201809325,
|
| 717 |
+
"eval_f1_macro": 0.6398260147869291,
|
| 718 |
+
"eval_f1_micro": 0.758185787048266,
|
| 719 |
+
"eval_loss": 0.15433941781520844,
|
| 720 |
+
"eval_roc_auc": 0.8352458155607984,
|
| 721 |
+
"eval_runtime": 377.6162,
|
| 722 |
+
"eval_samples_per_second": 7.611,
|
| 723 |
+
"eval_steps_per_second": 0.238,
|
| 724 |
+
"learning_rate": 1e-05,
|
| 725 |
+
"step": 12056
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 45.0,
|
| 729 |
+
"eval_accuracy": 0.3990953375086987,
|
| 730 |
+
"eval_f1_macro": 0.6206082021561579,
|
| 731 |
+
"eval_f1_micro": 0.7447583176396736,
|
| 732 |
+
"eval_loss": 0.1543821096420288,
|
| 733 |
+
"eval_roc_auc": 0.8196012986350353,
|
| 734 |
+
"eval_runtime": 369.6762,
|
| 735 |
+
"eval_samples_per_second": 7.774,
|
| 736 |
+
"eval_steps_per_second": 0.243,
|
| 737 |
+
"learning_rate": 1e-05,
|
| 738 |
+
"step": 12330
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 45.62,
|
| 742 |
+
"learning_rate": 1e-05,
|
| 743 |
+
"loss": 0.1746,
|
| 744 |
+
"step": 12500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 46.0,
|
| 748 |
+
"eval_accuracy": 0.40501043841336115,
|
| 749 |
+
"eval_f1_macro": 0.6223405225541471,
|
| 750 |
+
"eval_f1_micro": 0.7452157598499062,
|
| 751 |
+
"eval_loss": 0.15461167693138123,
|
| 752 |
+
"eval_roc_auc": 0.8207527729569265,
|
| 753 |
+
"eval_runtime": 369.5577,
|
| 754 |
+
"eval_samples_per_second": 7.777,
|
| 755 |
+
"eval_steps_per_second": 0.244,
|
| 756 |
+
"learning_rate": 1e-05,
|
| 757 |
+
"step": 12604
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 46.0,
|
| 761 |
+
"learning_rate": 1e-05,
|
| 762 |
+
"step": 12604,
|
| 763 |
+
"total_flos": 4.586865882212175e+19,
|
| 764 |
+
"train_loss": 0.1827572128274107,
|
| 765 |
+
"train_runtime": 70769.3212,
|
| 766 |
+
"train_samples_per_second": 11.517,
|
| 767 |
+
"train_steps_per_second": 0.36
|
| 768 |
+
}
|
| 769 |
+
],
|
| 770 |
+
"logging_steps": 500,
|
| 771 |
+
"max_steps": 25482,
|
| 772 |
+
"num_input_tokens_seen": 0,
|
| 773 |
+
"num_train_epochs": 93,
|
| 774 |
+
"save_steps": 500,
|
| 775 |
+
"total_flos": 4.586865882212175e+19,
|
| 776 |
+
"train_batch_size": 32,
|
| 777 |
+
"trial_name": null,
|
| 778 |
+
"trial_params": null
|
| 779 |
+
}
|