Commit
•
f6be8c4
1
Parent(s):
d06839c
Evaluation on the test set completed on 2024_05_23.
Browse files- README.md +5 -5
- all_results.json +17 -0
- logs/events.out.tfevents.1716454187.u22redoip03 +3 -0
- test_results.json +12 -0
- train_results.json +9 -0
- trainer_state.json +716 -0
README.md
CHANGED
|
@@ -17,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/dinov2-large](https://huggingface.co/facebook/dinov2-large) on the None dataset.
|
| 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: 0.0001
|
| 26 |
|
| 27 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/dinov2-large](https://huggingface.co/facebook/dinov2-large) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.2361
|
| 21 |
+
- F1 Micro: 0.7694
|
| 22 |
+
- F1 Macro: 0.4048
|
| 23 |
+
- Roc Auc: 0.8448
|
| 24 |
+
- Accuracy: 0.1449
|
| 25 |
- Learning Rate: 0.0001
|
| 26 |
|
| 27 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.14489458453906573,
|
| 4 |
+
"eval_f1_macro": 0.40480966865155954,
|
| 5 |
+
"eval_f1_micro": 0.7693991831922866,
|
| 6 |
+
"eval_loss": 0.23611676692962646,
|
| 7 |
+
"eval_roc_auc": 0.8447725516744388,
|
| 8 |
+
"eval_runtime": 101.4033,
|
| 9 |
+
"eval_samples_per_second": 47.71,
|
| 10 |
+
"eval_steps_per_second": 0.099,
|
| 11 |
+
"learning_rate": 0.0001,
|
| 12 |
+
"total_flos": 9.998216165993483e+19,
|
| 13 |
+
"train_loss": 0.2689315250941685,
|
| 14 |
+
"train_runtime": 8372.8151,
|
| 15 |
+
"train_samples_per_second": 84.762,
|
| 16 |
+
"train_steps_per_second": 0.167
|
| 17 |
+
}
|
logs/events.out.tfevents.1716454187.u22redoip03
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:992db895ca927b61dcf6adb36128bd8acf26c1e8d6a8b3b9a676f7aeb2a9adfc
|
| 3 |
+
size 526
|
test_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.14489458453906573,
|
| 4 |
+
"eval_f1_macro": 0.40480966865155954,
|
| 5 |
+
"eval_f1_micro": 0.7693991831922866,
|
| 6 |
+
"eval_loss": 0.23611676692962646,
|
| 7 |
+
"eval_roc_auc": 0.8447725516744388,
|
| 8 |
+
"eval_runtime": 101.4033,
|
| 9 |
+
"eval_samples_per_second": 47.71,
|
| 10 |
+
"eval_steps_per_second": 0.099,
|
| 11 |
+
"learning_rate": 0.0001
|
| 12 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"learning_rate": 0.0001,
|
| 4 |
+
"total_flos": 9.998216165993483e+19,
|
| 5 |
+
"train_loss": 0.2689315250941685,
|
| 6 |
+
"train_runtime": 8372.8151,
|
| 7 |
+
"train_samples_per_second": 84.762,
|
| 8 |
+
"train_steps_per_second": 0.167
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,716 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.23673731088638306,
|
| 3 |
+
"best_model_checkpoint": "/mnt/disk_victorlebos/data/datarmor/models/dinov2-large-2024_05_23-drone_batch-size512_epochs50_freeze/checkpoint-1400",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1400,
|
| 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": 0.012405938580435224,
|
| 14 |
+
"eval_f1_macro": 0.40667627783698285,
|
| 15 |
+
"eval_f1_micro": 0.5738973203699311,
|
| 16 |
+
"eval_loss": 0.5951732397079468,
|
| 17 |
+
"eval_roc_auc": 0.752833094536879,
|
| 18 |
+
"eval_runtime": 32.6366,
|
| 19 |
+
"eval_samples_per_second": 150.659,
|
| 20 |
+
"eval_steps_per_second": 0.306,
|
| 21 |
+
"learning_rate": 0.001,
|
| 22 |
+
"step": 28
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"eval_accuracy": 0.06975798250966037,
|
| 27 |
+
"eval_f1_macro": 0.4367882492755507,
|
| 28 |
+
"eval_f1_micro": 0.7307120964254151,
|
| 29 |
+
"eval_loss": 0.4730209410190582,
|
| 30 |
+
"eval_roc_auc": 0.8401493620198748,
|
| 31 |
+
"eval_runtime": 32.801,
|
| 32 |
+
"eval_samples_per_second": 149.904,
|
| 33 |
+
"eval_steps_per_second": 0.305,
|
| 34 |
+
"learning_rate": 0.001,
|
| 35 |
+
"step": 56
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 3.0,
|
| 39 |
+
"eval_accuracy": 0.10738255033557047,
|
| 40 |
+
"eval_f1_macro": 0.37702459211637257,
|
| 41 |
+
"eval_f1_micro": 0.7498981835953409,
|
| 42 |
+
"eval_loss": 0.3240152895450592,
|
| 43 |
+
"eval_roc_auc": 0.8377639463906806,
|
| 44 |
+
"eval_runtime": 33.9297,
|
| 45 |
+
"eval_samples_per_second": 144.917,
|
| 46 |
+
"eval_steps_per_second": 0.295,
|
| 47 |
+
"learning_rate": 0.001,
|
| 48 |
+
"step": 84
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 4.0,
|
| 52 |
+
"eval_accuracy": 0.11795810453528574,
|
| 53 |
+
"eval_f1_macro": 0.3710481900670742,
|
| 54 |
+
"eval_f1_micro": 0.7521195160095482,
|
| 55 |
+
"eval_loss": 0.2770342230796814,
|
| 56 |
+
"eval_roc_auc": 0.8371968879000634,
|
| 57 |
+
"eval_runtime": 33.1578,
|
| 58 |
+
"eval_samples_per_second": 148.291,
|
| 59 |
+
"eval_steps_per_second": 0.302,
|
| 60 |
+
"learning_rate": 0.001,
|
| 61 |
+
"step": 112
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 5.0,
|
| 65 |
+
"eval_accuracy": 0.11958511287370348,
|
| 66 |
+
"eval_f1_macro": 0.3714736793659693,
|
| 67 |
+
"eval_f1_micro": 0.7507292550220328,
|
| 68 |
+
"eval_loss": 0.25879302620887756,
|
| 69 |
+
"eval_roc_auc": 0.8353003094576528,
|
| 70 |
+
"eval_runtime": 33.1064,
|
| 71 |
+
"eval_samples_per_second": 148.521,
|
| 72 |
+
"eval_steps_per_second": 0.302,
|
| 73 |
+
"learning_rate": 0.001,
|
| 74 |
+
"step": 140
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 6.0,
|
| 78 |
+
"eval_accuracy": 0.12182224933902787,
|
| 79 |
+
"eval_f1_macro": 0.36304822534346387,
|
| 80 |
+
"eval_f1_micro": 0.7520252586099456,
|
| 81 |
+
"eval_loss": 0.25328728556632996,
|
| 82 |
+
"eval_roc_auc": 0.8353620986749458,
|
| 83 |
+
"eval_runtime": 33.4017,
|
| 84 |
+
"eval_samples_per_second": 147.208,
|
| 85 |
+
"eval_steps_per_second": 0.299,
|
| 86 |
+
"learning_rate": 0.001,
|
| 87 |
+
"step": 168
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 7.0,
|
| 91 |
+
"eval_accuracy": 0.11531421598535692,
|
| 92 |
+
"eval_f1_macro": 0.3646331511607325,
|
| 93 |
+
"eval_f1_micro": 0.7517183920016662,
|
| 94 |
+
"eval_loss": 0.25132349133491516,
|
| 95 |
+
"eval_roc_auc": 0.834652371152471,
|
| 96 |
+
"eval_runtime": 33.338,
|
| 97 |
+
"eval_samples_per_second": 147.489,
|
| 98 |
+
"eval_steps_per_second": 0.3,
|
| 99 |
+
"learning_rate": 0.001,
|
| 100 |
+
"step": 196
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 8.0,
|
| 104 |
+
"eval_accuracy": 0.12283912955053895,
|
| 105 |
+
"eval_f1_macro": 0.38940077262215617,
|
| 106 |
+
"eval_f1_micro": 0.7576399892988702,
|
| 107 |
+
"eval_loss": 0.2507544159889221,
|
| 108 |
+
"eval_roc_auc": 0.8407135898239845,
|
| 109 |
+
"eval_runtime": 33.4587,
|
| 110 |
+
"eval_samples_per_second": 146.957,
|
| 111 |
+
"eval_steps_per_second": 0.299,
|
| 112 |
+
"learning_rate": 0.001,
|
| 113 |
+
"step": 224
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 9.0,
|
| 117 |
+
"eval_accuracy": 0.12751677852348994,
|
| 118 |
+
"eval_f1_macro": 0.38290945223752887,
|
| 119 |
+
"eval_f1_micro": 0.7549859932265752,
|
| 120 |
+
"eval_loss": 0.24785615503787994,
|
| 121 |
+
"eval_roc_auc": 0.8360210323976333,
|
| 122 |
+
"eval_runtime": 34.4389,
|
| 123 |
+
"eval_samples_per_second": 142.775,
|
| 124 |
+
"eval_steps_per_second": 0.29,
|
| 125 |
+
"learning_rate": 0.001,
|
| 126 |
+
"step": 252
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 10.0,
|
| 130 |
+
"eval_accuracy": 0.12649989831197886,
|
| 131 |
+
"eval_f1_macro": 0.37973264961121395,
|
| 132 |
+
"eval_f1_micro": 0.7583163191651716,
|
| 133 |
+
"eval_loss": 0.2480766475200653,
|
| 134 |
+
"eval_roc_auc": 0.8407307602236042,
|
| 135 |
+
"eval_runtime": 34.0205,
|
| 136 |
+
"eval_samples_per_second": 144.531,
|
| 137 |
+
"eval_steps_per_second": 0.294,
|
| 138 |
+
"learning_rate": 0.001,
|
| 139 |
+
"step": 280
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 11.0,
|
| 143 |
+
"eval_accuracy": 0.12426276184665447,
|
| 144 |
+
"eval_f1_macro": 0.3964288145693209,
|
| 145 |
+
"eval_f1_micro": 0.7600958878849345,
|
| 146 |
+
"eval_loss": 0.24667006731033325,
|
| 147 |
+
"eval_roc_auc": 0.8431424613547966,
|
| 148 |
+
"eval_runtime": 34.2677,
|
| 149 |
+
"eval_samples_per_second": 143.488,
|
| 150 |
+
"eval_steps_per_second": 0.292,
|
| 151 |
+
"learning_rate": 0.001,
|
| 152 |
+
"step": 308
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 12.0,
|
| 156 |
+
"eval_accuracy": 0.12507626601586333,
|
| 157 |
+
"eval_f1_macro": 0.3733203958034908,
|
| 158 |
+
"eval_f1_micro": 0.7564640698455339,
|
| 159 |
+
"eval_loss": 0.2459569126367569,
|
| 160 |
+
"eval_roc_auc": 0.836213120741293,
|
| 161 |
+
"eval_runtime": 34.8153,
|
| 162 |
+
"eval_samples_per_second": 141.231,
|
| 163 |
+
"eval_steps_per_second": 0.287,
|
| 164 |
+
"learning_rate": 0.001,
|
| 165 |
+
"step": 336
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 13.0,
|
| 169 |
+
"eval_accuracy": 0.1297539149888143,
|
| 170 |
+
"eval_f1_macro": 0.38618999344086086,
|
| 171 |
+
"eval_f1_micro": 0.7581923944769908,
|
| 172 |
+
"eval_loss": 0.245611771941185,
|
| 173 |
+
"eval_roc_auc": 0.839925579714772,
|
| 174 |
+
"eval_runtime": 34.1156,
|
| 175 |
+
"eval_samples_per_second": 144.128,
|
| 176 |
+
"eval_steps_per_second": 0.293,
|
| 177 |
+
"learning_rate": 0.001,
|
| 178 |
+
"step": 364
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 14.0,
|
| 182 |
+
"eval_accuracy": 0.13707545251169412,
|
| 183 |
+
"eval_f1_macro": 0.37084554766098704,
|
| 184 |
+
"eval_f1_micro": 0.7526021832952525,
|
| 185 |
+
"eval_loss": 0.24649737775325775,
|
| 186 |
+
"eval_roc_auc": 0.8323046017083424,
|
| 187 |
+
"eval_runtime": 34.0576,
|
| 188 |
+
"eval_samples_per_second": 144.373,
|
| 189 |
+
"eval_steps_per_second": 0.294,
|
| 190 |
+
"learning_rate": 0.001,
|
| 191 |
+
"step": 392
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 15.0,
|
| 195 |
+
"eval_accuracy": 0.1271100264388855,
|
| 196 |
+
"eval_f1_macro": 0.37953234953900117,
|
| 197 |
+
"eval_f1_micro": 0.7540528606572888,
|
| 198 |
+
"eval_loss": 0.24523988366127014,
|
| 199 |
+
"eval_roc_auc": 0.8343517118764396,
|
| 200 |
+
"eval_runtime": 34.2408,
|
| 201 |
+
"eval_samples_per_second": 143.601,
|
| 202 |
+
"eval_steps_per_second": 0.292,
|
| 203 |
+
"learning_rate": 0.001,
|
| 204 |
+
"step": 420
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 16.0,
|
| 208 |
+
"eval_accuracy": 0.1293471629042099,
|
| 209 |
+
"eval_f1_macro": 0.39042586476441543,
|
| 210 |
+
"eval_f1_micro": 0.7597242635642867,
|
| 211 |
+
"eval_loss": 0.24370642006397247,
|
| 212 |
+
"eval_roc_auc": 0.8408989171069663,
|
| 213 |
+
"eval_runtime": 34.1492,
|
| 214 |
+
"eval_samples_per_second": 143.986,
|
| 215 |
+
"eval_steps_per_second": 0.293,
|
| 216 |
+
"learning_rate": 0.001,
|
| 217 |
+
"step": 448
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 17.0,
|
| 221 |
+
"eval_accuracy": 0.13158429936953428,
|
| 222 |
+
"eval_f1_macro": 0.38542350135117487,
|
| 223 |
+
"eval_f1_micro": 0.7525727259224682,
|
| 224 |
+
"eval_loss": 0.24466517567634583,
|
| 225 |
+
"eval_roc_auc": 0.8317419799054098,
|
| 226 |
+
"eval_runtime": 33.8536,
|
| 227 |
+
"eval_samples_per_second": 145.243,
|
| 228 |
+
"eval_steps_per_second": 0.295,
|
| 229 |
+
"learning_rate": 0.001,
|
| 230 |
+
"step": 476
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 17.857142857142858,
|
| 234 |
+
"grad_norm": 0.09434398263692856,
|
| 235 |
+
"learning_rate": 0.001,
|
| 236 |
+
"loss": 0.3126,
|
| 237 |
+
"step": 500
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 18.0,
|
| 241 |
+
"eval_accuracy": 0.133211307707952,
|
| 242 |
+
"eval_f1_macro": 0.35783734462173733,
|
| 243 |
+
"eval_f1_micro": 0.7534316217590239,
|
| 244 |
+
"eval_loss": 0.24544650316238403,
|
| 245 |
+
"eval_roc_auc": 0.8325846543647923,
|
| 246 |
+
"eval_runtime": 33.8618,
|
| 247 |
+
"eval_samples_per_second": 145.208,
|
| 248 |
+
"eval_steps_per_second": 0.295,
|
| 249 |
+
"learning_rate": 0.001,
|
| 250 |
+
"step": 504
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 19.0,
|
| 254 |
+
"eval_accuracy": 0.13239780353874314,
|
| 255 |
+
"eval_f1_macro": 0.3694145346248099,
|
| 256 |
+
"eval_f1_micro": 0.7568417082268136,
|
| 257 |
+
"eval_loss": 0.2440878450870514,
|
| 258 |
+
"eval_roc_auc": 0.8367437261179794,
|
| 259 |
+
"eval_runtime": 33.9612,
|
| 260 |
+
"eval_samples_per_second": 144.783,
|
| 261 |
+
"eval_steps_per_second": 0.294,
|
| 262 |
+
"learning_rate": 0.001,
|
| 263 |
+
"step": 532
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 20.0,
|
| 267 |
+
"eval_accuracy": 0.13605857230018303,
|
| 268 |
+
"eval_f1_macro": 0.3768127127776539,
|
| 269 |
+
"eval_f1_micro": 0.750895096799091,
|
| 270 |
+
"eval_loss": 0.2453632354736328,
|
| 271 |
+
"eval_roc_auc": 0.8288096279784166,
|
| 272 |
+
"eval_runtime": 33.6713,
|
| 273 |
+
"eval_samples_per_second": 146.03,
|
| 274 |
+
"eval_steps_per_second": 0.297,
|
| 275 |
+
"learning_rate": 0.001,
|
| 276 |
+
"step": 560
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 21.0,
|
| 280 |
+
"eval_accuracy": 0.12487288997356112,
|
| 281 |
+
"eval_f1_macro": 0.38961590782494593,
|
| 282 |
+
"eval_f1_micro": 0.760243826841616,
|
| 283 |
+
"eval_loss": 0.24377579987049103,
|
| 284 |
+
"eval_roc_auc": 0.8415825201166579,
|
| 285 |
+
"eval_runtime": 33.7544,
|
| 286 |
+
"eval_samples_per_second": 145.67,
|
| 287 |
+
"eval_steps_per_second": 0.296,
|
| 288 |
+
"learning_rate": 0.001,
|
| 289 |
+
"step": 588
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 22.0,
|
| 293 |
+
"eval_accuracy": 0.13016066707341875,
|
| 294 |
+
"eval_f1_macro": 0.3715634230883189,
|
| 295 |
+
"eval_f1_micro": 0.7576183975637929,
|
| 296 |
+
"eval_loss": 0.24192409217357635,
|
| 297 |
+
"eval_roc_auc": 0.8368051515288276,
|
| 298 |
+
"eval_runtime": 33.7095,
|
| 299 |
+
"eval_samples_per_second": 145.864,
|
| 300 |
+
"eval_steps_per_second": 0.297,
|
| 301 |
+
"learning_rate": 0.001,
|
| 302 |
+
"step": 616
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 23.0,
|
| 306 |
+
"eval_accuracy": 0.12649989831197886,
|
| 307 |
+
"eval_f1_macro": 0.3880375815747224,
|
| 308 |
+
"eval_f1_micro": 0.7628996647313762,
|
| 309 |
+
"eval_loss": 0.24348826706409454,
|
| 310 |
+
"eval_roc_auc": 0.845384034822261,
|
| 311 |
+
"eval_runtime": 33.9782,
|
| 312 |
+
"eval_samples_per_second": 144.711,
|
| 313 |
+
"eval_steps_per_second": 0.294,
|
| 314 |
+
"learning_rate": 0.001,
|
| 315 |
+
"step": 644
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 24.0,
|
| 319 |
+
"eval_accuracy": 0.1342281879194631,
|
| 320 |
+
"eval_f1_macro": 0.3896884130115941,
|
| 321 |
+
"eval_f1_micro": 0.7561114991428027,
|
| 322 |
+
"eval_loss": 0.2413305789232254,
|
| 323 |
+
"eval_roc_auc": 0.8344032230752988,
|
| 324 |
+
"eval_runtime": 34.4343,
|
| 325 |
+
"eval_samples_per_second": 142.794,
|
| 326 |
+
"eval_steps_per_second": 0.29,
|
| 327 |
+
"learning_rate": 0.001,
|
| 328 |
+
"step": 672
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 25.0,
|
| 332 |
+
"eval_accuracy": 0.1297539149888143,
|
| 333 |
+
"eval_f1_macro": 0.38267978517684004,
|
| 334 |
+
"eval_f1_micro": 0.7599182173024102,
|
| 335 |
+
"eval_loss": 0.24189460277557373,
|
| 336 |
+
"eval_roc_auc": 0.8414795784567333,
|
| 337 |
+
"eval_runtime": 34.3901,
|
| 338 |
+
"eval_samples_per_second": 142.977,
|
| 339 |
+
"eval_steps_per_second": 0.291,
|
| 340 |
+
"learning_rate": 0.001,
|
| 341 |
+
"step": 700
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 26.0,
|
| 345 |
+
"eval_accuracy": 0.12670327435428105,
|
| 346 |
+
"eval_f1_macro": 0.3971421437602147,
|
| 347 |
+
"eval_f1_micro": 0.7593076827294236,
|
| 348 |
+
"eval_loss": 0.2437727451324463,
|
| 349 |
+
"eval_roc_auc": 0.8401496844846781,
|
| 350 |
+
"eval_runtime": 34.2926,
|
| 351 |
+
"eval_samples_per_second": 143.384,
|
| 352 |
+
"eval_steps_per_second": 0.292,
|
| 353 |
+
"learning_rate": 0.001,
|
| 354 |
+
"step": 728
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 27.0,
|
| 358 |
+
"eval_accuracy": 0.1309741712426276,
|
| 359 |
+
"eval_f1_macro": 0.38383597863653807,
|
| 360 |
+
"eval_f1_micro": 0.761437908496732,
|
| 361 |
+
"eval_loss": 0.24182096123695374,
|
| 362 |
+
"eval_roc_auc": 0.8421530853512137,
|
| 363 |
+
"eval_runtime": 34.6404,
|
| 364 |
+
"eval_samples_per_second": 141.944,
|
| 365 |
+
"eval_steps_per_second": 0.289,
|
| 366 |
+
"learning_rate": 0.001,
|
| 367 |
+
"step": 756
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 28.0,
|
| 371 |
+
"eval_accuracy": 0.13341468375025423,
|
| 372 |
+
"eval_f1_macro": 0.3792682503180625,
|
| 373 |
+
"eval_f1_micro": 0.7498440155769273,
|
| 374 |
+
"eval_loss": 0.24316559731960297,
|
| 375 |
+
"eval_roc_auc": 0.8275284660741716,
|
| 376 |
+
"eval_runtime": 33.9704,
|
| 377 |
+
"eval_samples_per_second": 144.743,
|
| 378 |
+
"eval_steps_per_second": 0.294,
|
| 379 |
+
"learning_rate": 0.001,
|
| 380 |
+
"step": 784
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 29.0,
|
| 384 |
+
"eval_accuracy": 0.1366687004270897,
|
| 385 |
+
"eval_f1_macro": 0.39596794972011545,
|
| 386 |
+
"eval_f1_micro": 0.7621594930458399,
|
| 387 |
+
"eval_loss": 0.24201267957687378,
|
| 388 |
+
"eval_roc_auc": 0.8436467097327637,
|
| 389 |
+
"eval_runtime": 34.1508,
|
| 390 |
+
"eval_samples_per_second": 143.979,
|
| 391 |
+
"eval_steps_per_second": 0.293,
|
| 392 |
+
"learning_rate": 0.001,
|
| 393 |
+
"step": 812
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 30.0,
|
| 397 |
+
"eval_accuracy": 0.14236322961155176,
|
| 398 |
+
"eval_f1_macro": 0.38596411111358153,
|
| 399 |
+
"eval_f1_micro": 0.7619565217391304,
|
| 400 |
+
"eval_loss": 0.2406790852546692,
|
| 401 |
+
"eval_roc_auc": 0.840430919512073,
|
| 402 |
+
"eval_runtime": 34.2312,
|
| 403 |
+
"eval_samples_per_second": 143.641,
|
| 404 |
+
"eval_steps_per_second": 0.292,
|
| 405 |
+
"learning_rate": 0.001,
|
| 406 |
+
"step": 840
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 31.0,
|
| 410 |
+
"eval_accuracy": 0.13280455562334756,
|
| 411 |
+
"eval_f1_macro": 0.3928781445591724,
|
| 412 |
+
"eval_f1_micro": 0.7611869607298037,
|
| 413 |
+
"eval_loss": 0.24222084879875183,
|
| 414 |
+
"eval_roc_auc": 0.8428517162907542,
|
| 415 |
+
"eval_runtime": 34.0721,
|
| 416 |
+
"eval_samples_per_second": 144.311,
|
| 417 |
+
"eval_steps_per_second": 0.293,
|
| 418 |
+
"learning_rate": 0.001,
|
| 419 |
+
"step": 868
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 32.0,
|
| 423 |
+
"eval_accuracy": 0.13117754728492984,
|
| 424 |
+
"eval_f1_macro": 0.3912203123758987,
|
| 425 |
+
"eval_f1_micro": 0.7516135926480015,
|
| 426 |
+
"eval_loss": 0.24304261803627014,
|
| 427 |
+
"eval_roc_auc": 0.8298242087648536,
|
| 428 |
+
"eval_runtime": 33.969,
|
| 429 |
+
"eval_samples_per_second": 144.75,
|
| 430 |
+
"eval_steps_per_second": 0.294,
|
| 431 |
+
"learning_rate": 0.001,
|
| 432 |
+
"step": 896
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 33.0,
|
| 436 |
+
"eval_accuracy": 0.13016066707341875,
|
| 437 |
+
"eval_f1_macro": 0.38844227395152936,
|
| 438 |
+
"eval_f1_micro": 0.758885526453094,
|
| 439 |
+
"eval_loss": 0.24139608442783356,
|
| 440 |
+
"eval_roc_auc": 0.838798415911255,
|
| 441 |
+
"eval_runtime": 34.3634,
|
| 442 |
+
"eval_samples_per_second": 143.088,
|
| 443 |
+
"eval_steps_per_second": 0.291,
|
| 444 |
+
"learning_rate": 0.001,
|
| 445 |
+
"step": 924
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 34.0,
|
| 449 |
+
"eval_accuracy": 0.1354484441732764,
|
| 450 |
+
"eval_f1_macro": 0.4037409737349212,
|
| 451 |
+
"eval_f1_micro": 0.7624706542289075,
|
| 452 |
+
"eval_loss": 0.24039919674396515,
|
| 453 |
+
"eval_roc_auc": 0.8419399356299166,
|
| 454 |
+
"eval_runtime": 34.2464,
|
| 455 |
+
"eval_samples_per_second": 143.577,
|
| 456 |
+
"eval_steps_per_second": 0.292,
|
| 457 |
+
"learning_rate": 0.001,
|
| 458 |
+
"step": 952
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 35.0,
|
| 462 |
+
"eval_accuracy": 0.12995729103111653,
|
| 463 |
+
"eval_f1_macro": 0.3973020120442106,
|
| 464 |
+
"eval_f1_micro": 0.7601615858737297,
|
| 465 |
+
"eval_loss": 0.24134761095046997,
|
| 466 |
+
"eval_roc_auc": 0.8400403455705201,
|
| 467 |
+
"eval_runtime": 34.3251,
|
| 468 |
+
"eval_samples_per_second": 143.248,
|
| 469 |
+
"eval_steps_per_second": 0.291,
|
| 470 |
+
"learning_rate": 0.001,
|
| 471 |
+
"step": 980
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 35.714285714285715,
|
| 475 |
+
"grad_norm": 0.08046075701713562,
|
| 476 |
+
"learning_rate": 0.001,
|
| 477 |
+
"loss": 0.2465,
|
| 478 |
+
"step": 1000
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 36.0,
|
| 482 |
+
"eval_accuracy": 0.13565182021557862,
|
| 483 |
+
"eval_f1_macro": 0.38761085480429286,
|
| 484 |
+
"eval_f1_micro": 0.7622066694112803,
|
| 485 |
+
"eval_loss": 0.24192169308662415,
|
| 486 |
+
"eval_roc_auc": 0.8435798471665079,
|
| 487 |
+
"eval_runtime": 34.1038,
|
| 488 |
+
"eval_samples_per_second": 144.177,
|
| 489 |
+
"eval_steps_per_second": 0.293,
|
| 490 |
+
"learning_rate": 0.001,
|
| 491 |
+
"step": 1008
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 37.0,
|
| 495 |
+
"eval_accuracy": 0.1342281879194631,
|
| 496 |
+
"eval_f1_macro": 0.3992187594370792,
|
| 497 |
+
"eval_f1_micro": 0.7598352387357096,
|
| 498 |
+
"eval_loss": 0.2399486005306244,
|
| 499 |
+
"eval_roc_auc": 0.8380691834023499,
|
| 500 |
+
"eval_runtime": 34.6394,
|
| 501 |
+
"eval_samples_per_second": 141.948,
|
| 502 |
+
"eval_steps_per_second": 0.289,
|
| 503 |
+
"learning_rate": 0.001,
|
| 504 |
+
"step": 1036
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 38.0,
|
| 508 |
+
"eval_accuracy": 0.13300793166564978,
|
| 509 |
+
"eval_f1_macro": 0.3932700433432016,
|
| 510 |
+
"eval_f1_micro": 0.7607364527387098,
|
| 511 |
+
"eval_loss": 0.24004822969436646,
|
| 512 |
+
"eval_roc_auc": 0.8397074475961233,
|
| 513 |
+
"eval_runtime": 34.2918,
|
| 514 |
+
"eval_samples_per_second": 143.387,
|
| 515 |
+
"eval_steps_per_second": 0.292,
|
| 516 |
+
"learning_rate": 0.001,
|
| 517 |
+
"step": 1064
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 39.0,
|
| 521 |
+
"eval_accuracy": 0.13890583689241406,
|
| 522 |
+
"eval_f1_macro": 0.4007929579258356,
|
| 523 |
+
"eval_f1_micro": 0.7619087275149901,
|
| 524 |
+
"eval_loss": 0.24091550707817078,
|
| 525 |
+
"eval_roc_auc": 0.8411762713701608,
|
| 526 |
+
"eval_runtime": 34.4221,
|
| 527 |
+
"eval_samples_per_second": 142.844,
|
| 528 |
+
"eval_steps_per_second": 0.291,
|
| 529 |
+
"learning_rate": 0.001,
|
| 530 |
+
"step": 1092
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 40.0,
|
| 534 |
+
"eval_accuracy": 0.1354484441732764,
|
| 535 |
+
"eval_f1_macro": 0.39250375468507387,
|
| 536 |
+
"eval_f1_micro": 0.76,
|
| 537 |
+
"eval_loss": 0.23991511762142181,
|
| 538 |
+
"eval_roc_auc": 0.8378351644204551,
|
| 539 |
+
"eval_runtime": 35.4899,
|
| 540 |
+
"eval_samples_per_second": 138.546,
|
| 541 |
+
"eval_steps_per_second": 0.282,
|
| 542 |
+
"learning_rate": 0.001,
|
| 543 |
+
"step": 1120
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 41.0,
|
| 547 |
+
"eval_accuracy": 0.12487288997356112,
|
| 548 |
+
"eval_f1_macro": 0.40608061408264917,
|
| 549 |
+
"eval_f1_micro": 0.7639710985018574,
|
| 550 |
+
"eval_loss": 0.2422637641429901,
|
| 551 |
+
"eval_roc_auc": 0.8463905306427496,
|
| 552 |
+
"eval_runtime": 35.0605,
|
| 553 |
+
"eval_samples_per_second": 140.243,
|
| 554 |
+
"eval_steps_per_second": 0.285,
|
| 555 |
+
"learning_rate": 0.001,
|
| 556 |
+
"step": 1148
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 42.0,
|
| 560 |
+
"eval_accuracy": 0.1309741712426276,
|
| 561 |
+
"eval_f1_macro": 0.4005098857996497,
|
| 562 |
+
"eval_f1_micro": 0.7568840806286871,
|
| 563 |
+
"eval_loss": 0.24256455898284912,
|
| 564 |
+
"eval_roc_auc": 0.8377674429208019,
|
| 565 |
+
"eval_runtime": 35.1895,
|
| 566 |
+
"eval_samples_per_second": 139.729,
|
| 567 |
+
"eval_steps_per_second": 0.284,
|
| 568 |
+
"learning_rate": 0.001,
|
| 569 |
+
"step": 1176
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 43.0,
|
| 573 |
+
"eval_accuracy": 0.13361805979255645,
|
| 574 |
+
"eval_f1_macro": 0.4007981173529554,
|
| 575 |
+
"eval_f1_micro": 0.7594289817122102,
|
| 576 |
+
"eval_loss": 0.23922023177146912,
|
| 577 |
+
"eval_roc_auc": 0.8368935200169594,
|
| 578 |
+
"eval_runtime": 34.9621,
|
| 579 |
+
"eval_samples_per_second": 140.638,
|
| 580 |
+
"eval_steps_per_second": 0.286,
|
| 581 |
+
"learning_rate": 0.001,
|
| 582 |
+
"step": 1204
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 44.0,
|
| 586 |
+
"eval_accuracy": 0.13036404311572097,
|
| 587 |
+
"eval_f1_macro": 0.40641694858015515,
|
| 588 |
+
"eval_f1_micro": 0.7576905272903253,
|
| 589 |
+
"eval_loss": 0.24184103310108185,
|
| 590 |
+
"eval_roc_auc": 0.8364958212082838,
|
| 591 |
+
"eval_runtime": 35.0243,
|
| 592 |
+
"eval_samples_per_second": 140.388,
|
| 593 |
+
"eval_steps_per_second": 0.286,
|
| 594 |
+
"learning_rate": 0.001,
|
| 595 |
+
"step": 1232
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 45.0,
|
| 599 |
+
"eval_accuracy": 0.13788895668090298,
|
| 600 |
+
"eval_f1_macro": 0.39055068831340933,
|
| 601 |
+
"eval_f1_micro": 0.7591085068536151,
|
| 602 |
+
"eval_loss": 0.24105145037174225,
|
| 603 |
+
"eval_roc_auc": 0.8384187060666997,
|
| 604 |
+
"eval_runtime": 34.7759,
|
| 605 |
+
"eval_samples_per_second": 141.391,
|
| 606 |
+
"eval_steps_per_second": 0.288,
|
| 607 |
+
"learning_rate": 0.001,
|
| 608 |
+
"step": 1260
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 46.0,
|
| 612 |
+
"eval_accuracy": 0.13626194834248526,
|
| 613 |
+
"eval_f1_macro": 0.4106196361694743,
|
| 614 |
+
"eval_f1_micro": 0.7653508320819534,
|
| 615 |
+
"eval_loss": 0.23963303864002228,
|
| 616 |
+
"eval_roc_auc": 0.8457236836435186,
|
| 617 |
+
"eval_runtime": 34.9314,
|
| 618 |
+
"eval_samples_per_second": 140.762,
|
| 619 |
+
"eval_steps_per_second": 0.286,
|
| 620 |
+
"learning_rate": 0.001,
|
| 621 |
+
"step": 1288
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 47.0,
|
| 625 |
+
"eval_accuracy": 0.13260117958104536,
|
| 626 |
+
"eval_f1_macro": 0.3967990889217657,
|
| 627 |
+
"eval_f1_micro": 0.7575076348829317,
|
| 628 |
+
"eval_loss": 0.23957742750644684,
|
| 629 |
+
"eval_roc_auc": 0.8348926760093537,
|
| 630 |
+
"eval_runtime": 35.1971,
|
| 631 |
+
"eval_samples_per_second": 139.699,
|
| 632 |
+
"eval_steps_per_second": 0.284,
|
| 633 |
+
"learning_rate": 0.001,
|
| 634 |
+
"step": 1316
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 48.0,
|
| 638 |
+
"eval_accuracy": 0.12873703477730322,
|
| 639 |
+
"eval_f1_macro": 0.38777990454974365,
|
| 640 |
+
"eval_f1_micro": 0.7563947423325684,
|
| 641 |
+
"eval_loss": 0.24231907725334167,
|
| 642 |
+
"eval_roc_auc": 0.8372881264708554,
|
| 643 |
+
"eval_runtime": 106.544,
|
| 644 |
+
"eval_samples_per_second": 46.15,
|
| 645 |
+
"eval_steps_per_second": 0.094,
|
| 646 |
+
"learning_rate": 0.001,
|
| 647 |
+
"step": 1344
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 49.0,
|
| 651 |
+
"eval_accuracy": 0.13300793166564978,
|
| 652 |
+
"eval_f1_macro": 0.40266317126303974,
|
| 653 |
+
"eval_f1_micro": 0.7608376348147216,
|
| 654 |
+
"eval_loss": 0.23978127539157867,
|
| 655 |
+
"eval_roc_auc": 0.8389943850191152,
|
| 656 |
+
"eval_runtime": 103.5302,
|
| 657 |
+
"eval_samples_per_second": 47.493,
|
| 658 |
+
"eval_steps_per_second": 0.097,
|
| 659 |
+
"learning_rate": 0.001,
|
| 660 |
+
"step": 1372
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 50.0,
|
| 664 |
+
"eval_accuracy": 0.14236322961155176,
|
| 665 |
+
"eval_f1_macro": 0.4087415721658059,
|
| 666 |
+
"eval_f1_micro": 0.7652267908369019,
|
| 667 |
+
"eval_loss": 0.23673731088638306,
|
| 668 |
+
"eval_roc_auc": 0.8435746322832197,
|
| 669 |
+
"eval_runtime": 102.3756,
|
| 670 |
+
"eval_samples_per_second": 48.029,
|
| 671 |
+
"eval_steps_per_second": 0.098,
|
| 672 |
+
"learning_rate": 0.0001,
|
| 673 |
+
"step": 1400
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 50.0,
|
| 677 |
+
"learning_rate": 0.0001,
|
| 678 |
+
"step": 1400,
|
| 679 |
+
"total_flos": 9.998216165993483e+19,
|
| 680 |
+
"train_loss": 0.2689315250941685,
|
| 681 |
+
"train_runtime": 8372.8151,
|
| 682 |
+
"train_samples_per_second": 84.762,
|
| 683 |
+
"train_steps_per_second": 0.167
|
| 684 |
+
}
|
| 685 |
+
],
|
| 686 |
+
"logging_steps": 500,
|
| 687 |
+
"max_steps": 1400,
|
| 688 |
+
"num_input_tokens_seen": 0,
|
| 689 |
+
"num_train_epochs": 50,
|
| 690 |
+
"save_steps": 500,
|
| 691 |
+
"stateful_callbacks": {
|
| 692 |
+
"EarlyStoppingCallback": {
|
| 693 |
+
"args": {
|
| 694 |
+
"early_stopping_patience": 10,
|
| 695 |
+
"early_stopping_threshold": 0.0
|
| 696 |
+
},
|
| 697 |
+
"attributes": {
|
| 698 |
+
"early_stopping_patience_counter": 0
|
| 699 |
+
}
|
| 700 |
+
},
|
| 701 |
+
"TrainerControl": {
|
| 702 |
+
"args": {
|
| 703 |
+
"should_epoch_stop": false,
|
| 704 |
+
"should_evaluate": false,
|
| 705 |
+
"should_log": false,
|
| 706 |
+
"should_save": true,
|
| 707 |
+
"should_training_stop": true
|
| 708 |
+
},
|
| 709 |
+
"attributes": {}
|
| 710 |
+
}
|
| 711 |
+
},
|
| 712 |
+
"total_flos": 9.998216165993483e+19,
|
| 713 |
+
"train_batch_size": 512,
|
| 714 |
+
"trial_name": null,
|
| 715 |
+
"trial_params": null
|
| 716 |
+
}
|