Nguyen Tien
commited on
Commit
•
c2f2911
1
Parent(s):
13e9bbb
vuihocrnd/teacher-status-van-tiny-256
Browse files- README.md +8 -8
- all_results.json +14 -5
- eval_results.json +9 -5
- train_results.json +6 -6
- trainer_state.json +557 -47
README.md
CHANGED
|
@@ -24,13 +24,13 @@ model-index:
|
|
| 24 |
metrics:
|
| 25 |
- name: Accuracy
|
| 26 |
type: accuracy
|
| 27 |
-
value: 0.
|
| 28 |
- name: Recall
|
| 29 |
type: recall
|
| 30 |
-
value: 0.
|
| 31 |
- name: Precision
|
| 32 |
type: precision
|
| 33 |
-
value: 0.
|
| 34 |
---
|
| 35 |
|
| 36 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -40,11 +40,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 40 |
|
| 41 |
This model is a fine-tuned version of [Visual-Attention-Network/van-tiny](https://huggingface.co/Visual-Attention-Network/van-tiny) on the imagefolder dataset.
|
| 42 |
It achieves the following results on the evaluation set:
|
| 43 |
-
- Loss: 0.
|
| 44 |
-
- Accuracy: 0.
|
| 45 |
-
- F1 Score: 0.
|
| 46 |
-
- Recall: 0.
|
| 47 |
-
- Precision: 0.
|
| 48 |
|
| 49 |
## Model description
|
| 50 |
|
|
|
|
| 24 |
metrics:
|
| 25 |
- name: Accuracy
|
| 26 |
type: accuracy
|
| 27 |
+
value: 0.9831460674157303
|
| 28 |
- name: Recall
|
| 29 |
type: recall
|
| 30 |
+
value: 0.9789473684210527
|
| 31 |
- name: Precision
|
| 32 |
type: precision
|
| 33 |
+
value: 0.9893617021276596
|
| 34 |
---
|
| 35 |
|
| 36 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 40 |
|
| 41 |
This model is a fine-tuned version of [Visual-Attention-Network/van-tiny](https://huggingface.co/Visual-Attention-Network/van-tiny) on the imagefolder dataset.
|
| 42 |
It achieves the following results on the evaluation set:
|
| 43 |
+
- Loss: 0.0988
|
| 44 |
+
- Accuracy: 0.9831
|
| 45 |
+
- F1 Score: 0.9841
|
| 46 |
+
- Recall: 0.9789
|
| 47 |
+
- Precision: 0.9894
|
| 48 |
|
| 49 |
## Model description
|
| 50 |
|
all_results.json
CHANGED
|
@@ -1,7 +1,16 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 28.8,
|
| 3 |
+
"eval_accuracy": 0.9831460674157303,
|
| 4 |
+
"eval_f1_score": 0.9841269841269842,
|
| 5 |
+
"eval_loss": 0.09884575009346008,
|
| 6 |
+
"eval_precision": 0.9893617021276596,
|
| 7 |
+
"eval_recall": 0.9789473684210527,
|
| 8 |
+
"eval_runtime": 1.7039,
|
| 9 |
+
"eval_samples_per_second": 104.466,
|
| 10 |
+
"eval_steps_per_second": 3.521,
|
| 11 |
+
"total_flos": 2.0837941191062323e+17,
|
| 12 |
+
"train_loss": 0.3229101174407535,
|
| 13 |
+
"train_runtime": 715.5601,
|
| 14 |
+
"train_samples_per_second": 66.787,
|
| 15 |
+
"train_steps_per_second": 0.503
|
| 16 |
}
|
eval_results.json
CHANGED
|
@@ -1,7 +1,11 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 28.8,
|
| 3 |
+
"eval_accuracy": 0.9831460674157303,
|
| 4 |
+
"eval_f1_score": 0.9841269841269842,
|
| 5 |
+
"eval_loss": 0.09884575009346008,
|
| 6 |
+
"eval_precision": 0.9893617021276596,
|
| 7 |
+
"eval_recall": 0.9789473684210527,
|
| 8 |
+
"eval_runtime": 1.7039,
|
| 9 |
+
"eval_samples_per_second": 104.466,
|
| 10 |
+
"eval_steps_per_second": 3.521
|
| 11 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 28.8,
|
| 3 |
+
"total_flos": 2.0837941191062323e+17,
|
| 4 |
+
"train_loss": 0.3229101174407535,
|
| 5 |
+
"train_runtime": 715.5601,
|
| 6 |
+
"train_samples_per_second": 66.787,
|
| 7 |
+
"train_steps_per_second": 0.503
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,83 +1,593 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "teacher-status-van-tiny-256/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.8,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.96,
|
| 19 |
-
"eval_accuracy": 0.
|
| 20 |
-
"eval_f1_score": 0.
|
| 21 |
-
"eval_loss": 0.
|
| 22 |
-
"eval_precision": 0.
|
| 23 |
-
"eval_recall": 0
|
| 24 |
-
"eval_runtime": 1.
|
| 25 |
-
"eval_samples_per_second":
|
| 26 |
-
"eval_steps_per_second": 4.
|
| 27 |
"step": 12
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 1.6,
|
| 31 |
-
"learning_rate": 2.
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 20
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 2.0,
|
| 37 |
-
"eval_accuracy": 0.
|
| 38 |
-
"eval_f1_score": 0.
|
| 39 |
-
"eval_loss": 0.
|
| 40 |
-
"eval_precision": 0.
|
| 41 |
-
"eval_recall": 0
|
| 42 |
-
"eval_runtime": 1.
|
| 43 |
-
"eval_samples_per_second":
|
| 44 |
-
"eval_steps_per_second": 5.
|
| 45 |
"step": 25
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.4,
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 2.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
"eval_accuracy": 0.9438202247191011,
|
| 56 |
-
"eval_f1_score": 0.
|
| 57 |
-
"eval_loss": 0.
|
| 58 |
-
"eval_precision": 0.
|
| 59 |
-
"eval_recall": 0.
|
| 60 |
-
"eval_runtime": 1.
|
| 61 |
-
"eval_samples_per_second":
|
| 62 |
-
"eval_steps_per_second": 5.
|
| 63 |
-
"step":
|
| 64 |
-
},
|
| 65 |
-
{
|
| 66 |
-
"epoch": 2
|
| 67 |
-
"
|
| 68 |
-
"
|
| 69 |
-
"
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
}
|
| 74 |
],
|
| 75 |
"logging_steps": 10,
|
| 76 |
-
"max_steps":
|
| 77 |
"num_input_tokens_seen": 0,
|
| 78 |
-
"num_train_epochs":
|
| 79 |
"save_steps": 500,
|
| 80 |
-
"total_flos": 2.
|
| 81 |
"train_batch_size": 32,
|
| 82 |
"trial_name": null,
|
| 83 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9841269841269842,
|
| 3 |
+
"best_model_checkpoint": "teacher-status-van-tiny-256/checkpoint-187",
|
| 4 |
+
"epoch": 28.8,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 360,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.8,
|
| 13 |
+
"learning_rate": 1.388888888888889e-05,
|
| 14 |
+
"loss": 0.6928,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.96,
|
| 19 |
+
"eval_accuracy": 0.6685393258426966,
|
| 20 |
+
"eval_f1_score": 0.7630522088353413,
|
| 21 |
+
"eval_loss": 0.6904417276382446,
|
| 22 |
+
"eval_precision": 0.6168831168831169,
|
| 23 |
+
"eval_recall": 1.0,
|
| 24 |
+
"eval_runtime": 1.2798,
|
| 25 |
+
"eval_samples_per_second": 139.083,
|
| 26 |
+
"eval_steps_per_second": 4.688,
|
| 27 |
"step": 12
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 1.6,
|
| 31 |
+
"learning_rate": 2.777777777777778e-05,
|
| 32 |
+
"loss": 0.6893,
|
| 33 |
"step": 20
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 2.0,
|
| 37 |
+
"eval_accuracy": 0.5393258426966292,
|
| 38 |
+
"eval_f1_score": 0.698529411764706,
|
| 39 |
+
"eval_loss": 0.6683324575424194,
|
| 40 |
+
"eval_precision": 0.536723163841808,
|
| 41 |
+
"eval_recall": 1.0,
|
| 42 |
+
"eval_runtime": 1.0413,
|
| 43 |
+
"eval_samples_per_second": 170.94,
|
| 44 |
+
"eval_steps_per_second": 5.762,
|
| 45 |
"step": 25
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.4,
|
| 49 |
+
"learning_rate": 4.166666666666667e-05,
|
| 50 |
+
"loss": 0.6726,
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 2.96,
|
| 55 |
+
"eval_accuracy": 0.5842696629213483,
|
| 56 |
+
"eval_f1_score": 0.7196969696969697,
|
| 57 |
+
"eval_loss": 0.5703710913658142,
|
| 58 |
+
"eval_precision": 0.5621301775147929,
|
| 59 |
+
"eval_recall": 1.0,
|
| 60 |
+
"eval_runtime": 1.3234,
|
| 61 |
+
"eval_samples_per_second": 134.506,
|
| 62 |
+
"eval_steps_per_second": 4.534,
|
| 63 |
+
"step": 37
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 3.2,
|
| 67 |
+
"learning_rate": 4.938271604938271e-05,
|
| 68 |
+
"loss": 0.6047,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 4.0,
|
| 73 |
+
"learning_rate": 4.783950617283951e-05,
|
| 74 |
+
"loss": 0.5295,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 4.0,
|
| 79 |
+
"eval_accuracy": 0.9213483146067416,
|
| 80 |
+
"eval_f1_score": 0.9263157894736842,
|
| 81 |
+
"eval_loss": 0.4148499667644501,
|
| 82 |
+
"eval_precision": 0.9263157894736842,
|
| 83 |
+
"eval_recall": 0.9263157894736842,
|
| 84 |
+
"eval_runtime": 1.0116,
|
| 85 |
+
"eval_samples_per_second": 175.963,
|
| 86 |
+
"eval_steps_per_second": 5.931,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 4.8,
|
| 91 |
+
"learning_rate": 4.62962962962963e-05,
|
| 92 |
+
"loss": 0.4745,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.96,
|
| 97 |
+
"eval_accuracy": 0.9382022471910112,
|
| 98 |
+
"eval_f1_score": 0.9430051813471502,
|
| 99 |
+
"eval_loss": 0.31081339716911316,
|
| 100 |
+
"eval_precision": 0.9285714285714286,
|
| 101 |
+
"eval_recall": 0.9578947368421052,
|
| 102 |
+
"eval_runtime": 1.3432,
|
| 103 |
+
"eval_samples_per_second": 132.515,
|
| 104 |
+
"eval_steps_per_second": 4.467,
|
| 105 |
+
"step": 62
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 5.6,
|
| 109 |
+
"learning_rate": 4.4753086419753084e-05,
|
| 110 |
+
"loss": 0.4206,
|
| 111 |
+
"step": 70
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.0,
|
| 115 |
+
"eval_accuracy": 0.9438202247191011,
|
| 116 |
+
"eval_f1_score": 0.9473684210526315,
|
| 117 |
+
"eval_loss": 0.2301165908575058,
|
| 118 |
+
"eval_precision": 0.9473684210526315,
|
| 119 |
+
"eval_recall": 0.9473684210526315,
|
| 120 |
+
"eval_runtime": 1.0463,
|
| 121 |
+
"eval_samples_per_second": 170.129,
|
| 122 |
+
"eval_steps_per_second": 5.735,
|
| 123 |
+
"step": 75
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 6.4,
|
| 127 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 128 |
+
"loss": 0.3898,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.96,
|
| 133 |
+
"eval_accuracy": 0.949438202247191,
|
| 134 |
+
"eval_f1_score": 0.9518716577540107,
|
| 135 |
+
"eval_loss": 0.1820228397846222,
|
| 136 |
+
"eval_precision": 0.967391304347826,
|
| 137 |
+
"eval_recall": 0.9368421052631579,
|
| 138 |
+
"eval_runtime": 1.0432,
|
| 139 |
+
"eval_samples_per_second": 170.633,
|
| 140 |
+
"eval_steps_per_second": 5.752,
|
| 141 |
+
"step": 87
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 7.2,
|
| 145 |
+
"learning_rate": 4.166666666666667e-05,
|
| 146 |
+
"loss": 0.3557,
|
| 147 |
+
"step": 90
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 8.0,
|
| 151 |
+
"learning_rate": 4.012345679012346e-05,
|
| 152 |
+
"loss": 0.3153,
|
| 153 |
+
"step": 100
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 8.0,
|
| 157 |
+
"eval_accuracy": 0.949438202247191,
|
| 158 |
+
"eval_f1_score": 0.9538461538461539,
|
| 159 |
+
"eval_loss": 0.1545208990573883,
|
| 160 |
+
"eval_precision": 0.93,
|
| 161 |
+
"eval_recall": 0.9789473684210527,
|
| 162 |
+
"eval_runtime": 1.2226,
|
| 163 |
+
"eval_samples_per_second": 145.595,
|
| 164 |
+
"eval_steps_per_second": 4.908,
|
| 165 |
+
"step": 100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 8.8,
|
| 169 |
+
"learning_rate": 3.8580246913580246e-05,
|
| 170 |
+
"loss": 0.3077,
|
| 171 |
+
"step": 110
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 8.96,
|
| 175 |
+
"eval_accuracy": 0.9606741573033708,
|
| 176 |
+
"eval_f1_score": 0.9621621621621621,
|
| 177 |
+
"eval_loss": 0.1521005630493164,
|
| 178 |
+
"eval_precision": 0.9888888888888889,
|
| 179 |
+
"eval_recall": 0.9368421052631579,
|
| 180 |
+
"eval_runtime": 1.2664,
|
| 181 |
+
"eval_samples_per_second": 140.559,
|
| 182 |
+
"eval_steps_per_second": 4.738,
|
| 183 |
+
"step": 112
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 9.6,
|
| 187 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 188 |
+
"loss": 0.3048,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 10.0,
|
| 193 |
+
"eval_accuracy": 0.9606741573033708,
|
| 194 |
+
"eval_f1_score": 0.9625668449197862,
|
| 195 |
+
"eval_loss": 0.133090078830719,
|
| 196 |
+
"eval_precision": 0.9782608695652174,
|
| 197 |
+
"eval_recall": 0.9473684210526315,
|
| 198 |
+
"eval_runtime": 1.032,
|
| 199 |
+
"eval_samples_per_second": 172.478,
|
| 200 |
+
"eval_steps_per_second": 5.814,
|
| 201 |
+
"step": 125
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 10.4,
|
| 205 |
+
"learning_rate": 3.5493827160493834e-05,
|
| 206 |
+
"loss": 0.3004,
|
| 207 |
+
"step": 130
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 10.96,
|
| 211 |
+
"eval_accuracy": 0.9606741573033708,
|
| 212 |
+
"eval_f1_score": 0.9633507853403142,
|
| 213 |
+
"eval_loss": 0.13144755363464355,
|
| 214 |
+
"eval_precision": 0.9583333333333334,
|
| 215 |
+
"eval_recall": 0.968421052631579,
|
| 216 |
+
"eval_runtime": 1.0224,
|
| 217 |
+
"eval_samples_per_second": 174.103,
|
| 218 |
+
"eval_steps_per_second": 5.869,
|
| 219 |
+
"step": 137
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 11.2,
|
| 223 |
+
"learning_rate": 3.395061728395062e-05,
|
| 224 |
+
"loss": 0.2685,
|
| 225 |
+
"step": 140
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 12.0,
|
| 229 |
+
"learning_rate": 3.240740740740741e-05,
|
| 230 |
+
"loss": 0.2839,
|
| 231 |
+
"step": 150
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 12.0,
|
| 235 |
+
"eval_accuracy": 0.9606741573033708,
|
| 236 |
+
"eval_f1_score": 0.9621621621621621,
|
| 237 |
+
"eval_loss": 0.1272086501121521,
|
| 238 |
+
"eval_precision": 0.9888888888888889,
|
| 239 |
+
"eval_recall": 0.9368421052631579,
|
| 240 |
+
"eval_runtime": 1.0248,
|
| 241 |
+
"eval_samples_per_second": 173.691,
|
| 242 |
+
"eval_steps_per_second": 5.855,
|
| 243 |
+
"step": 150
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 12.8,
|
| 247 |
+
"learning_rate": 3.08641975308642e-05,
|
| 248 |
+
"loss": 0.286,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 12.96,
|
| 253 |
+
"eval_accuracy": 0.9606741573033708,
|
| 254 |
+
"eval_f1_score": 0.9621621621621621,
|
| 255 |
+
"eval_loss": 0.11889710277318954,
|
| 256 |
+
"eval_precision": 0.9888888888888889,
|
| 257 |
+
"eval_recall": 0.9368421052631579,
|
| 258 |
+
"eval_runtime": 1.0343,
|
| 259 |
+
"eval_samples_per_second": 172.102,
|
| 260 |
+
"eval_steps_per_second": 5.801,
|
| 261 |
+
"step": 162
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 13.6,
|
| 265 |
+
"learning_rate": 2.9320987654320992e-05,
|
| 266 |
+
"loss": 0.2473,
|
| 267 |
+
"step": 170
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 14.0,
|
| 271 |
+
"eval_accuracy": 0.9719101123595506,
|
| 272 |
+
"eval_f1_score": 0.9732620320855614,
|
| 273 |
+
"eval_loss": 0.09774552285671234,
|
| 274 |
+
"eval_precision": 0.9891304347826086,
|
| 275 |
+
"eval_recall": 0.9578947368421052,
|
| 276 |
+
"eval_runtime": 1.2533,
|
| 277 |
+
"eval_samples_per_second": 142.026,
|
| 278 |
+
"eval_steps_per_second": 4.787,
|
| 279 |
+
"step": 175
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 14.4,
|
| 283 |
+
"learning_rate": 2.777777777777778e-05,
|
| 284 |
+
"loss": 0.2774,
|
| 285 |
+
"step": 180
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 14.96,
|
| 289 |
+
"eval_accuracy": 0.9831460674157303,
|
| 290 |
+
"eval_f1_score": 0.9841269841269842,
|
| 291 |
+
"eval_loss": 0.09884575009346008,
|
| 292 |
+
"eval_precision": 0.9893617021276596,
|
| 293 |
+
"eval_recall": 0.9789473684210527,
|
| 294 |
+
"eval_runtime": 1.2659,
|
| 295 |
+
"eval_samples_per_second": 140.612,
|
| 296 |
+
"eval_steps_per_second": 4.74,
|
| 297 |
+
"step": 187
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 15.2,
|
| 301 |
+
"learning_rate": 2.623456790123457e-05,
|
| 302 |
+
"loss": 0.2733,
|
| 303 |
+
"step": 190
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 16.0,
|
| 307 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 308 |
+
"loss": 0.2541,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 16.0,
|
| 313 |
+
"eval_accuracy": 0.9719101123595506,
|
| 314 |
+
"eval_f1_score": 0.9732620320855614,
|
| 315 |
+
"eval_loss": 0.09689934551715851,
|
| 316 |
+
"eval_precision": 0.9891304347826086,
|
| 317 |
+
"eval_recall": 0.9578947368421052,
|
| 318 |
+
"eval_runtime": 1.0329,
|
| 319 |
+
"eval_samples_per_second": 172.331,
|
| 320 |
+
"eval_steps_per_second": 5.809,
|
| 321 |
+
"step": 200
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 16.8,
|
| 325 |
+
"learning_rate": 2.314814814814815e-05,
|
| 326 |
+
"loss": 0.2383,
|
| 327 |
+
"step": 210
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 16.96,
|
| 331 |
+
"eval_accuracy": 0.9719101123595506,
|
| 332 |
+
"eval_f1_score": 0.9732620320855614,
|
| 333 |
+
"eval_loss": 0.10420462489128113,
|
| 334 |
+
"eval_precision": 0.9891304347826086,
|
| 335 |
+
"eval_recall": 0.9578947368421052,
|
| 336 |
+
"eval_runtime": 1.0218,
|
| 337 |
+
"eval_samples_per_second": 174.204,
|
| 338 |
+
"eval_steps_per_second": 5.872,
|
| 339 |
+
"step": 212
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 17.6,
|
| 343 |
+
"learning_rate": 2.1604938271604937e-05,
|
| 344 |
+
"loss": 0.2552,
|
| 345 |
+
"step": 220
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 18.0,
|
| 349 |
+
"eval_accuracy": 0.9719101123595506,
|
| 350 |
+
"eval_f1_score": 0.9732620320855614,
|
| 351 |
+
"eval_loss": 0.10807426273822784,
|
| 352 |
+
"eval_precision": 0.9891304347826086,
|
| 353 |
+
"eval_recall": 0.9578947368421052,
|
| 354 |
+
"eval_runtime": 1.0141,
|
| 355 |
+
"eval_samples_per_second": 175.532,
|
| 356 |
+
"eval_steps_per_second": 5.917,
|
| 357 |
+
"step": 225
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 18.4,
|
| 361 |
+
"learning_rate": 2.006172839506173e-05,
|
| 362 |
+
"loss": 0.2223,
|
| 363 |
+
"step": 230
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 18.96,
|
| 367 |
+
"eval_accuracy": 0.9662921348314607,
|
| 368 |
+
"eval_f1_score": 0.9680851063829788,
|
| 369 |
+
"eval_loss": 0.11498517543077469,
|
| 370 |
+
"eval_precision": 0.978494623655914,
|
| 371 |
+
"eval_recall": 0.9578947368421052,
|
| 372 |
+
"eval_runtime": 1.0007,
|
| 373 |
+
"eval_samples_per_second": 177.868,
|
| 374 |
+
"eval_steps_per_second": 5.996,
|
| 375 |
+
"step": 237
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 19.2,
|
| 379 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 380 |
+
"loss": 0.2364,
|
| 381 |
+
"step": 240
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 20.0,
|
| 385 |
+
"learning_rate": 1.697530864197531e-05,
|
| 386 |
+
"loss": 0.2561,
|
| 387 |
+
"step": 250
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 20.0,
|
| 391 |
+
"eval_accuracy": 0.9550561797752809,
|
| 392 |
+
"eval_f1_score": 0.9574468085106383,
|
| 393 |
+
"eval_loss": 0.12337563186883926,
|
| 394 |
+
"eval_precision": 0.967741935483871,
|
| 395 |
+
"eval_recall": 0.9473684210526315,
|
| 396 |
+
"eval_runtime": 1.025,
|
| 397 |
+
"eval_samples_per_second": 173.653,
|
| 398 |
+
"eval_steps_per_second": 5.853,
|
| 399 |
+
"step": 250
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 20.8,
|
| 403 |
+
"learning_rate": 1.54320987654321e-05,
|
| 404 |
+
"loss": 0.2462,
|
| 405 |
+
"step": 260
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 20.96,
|
| 409 |
+
"eval_accuracy": 0.9606741573033708,
|
| 410 |
+
"eval_f1_score": 0.962962962962963,
|
| 411 |
+
"eval_loss": 0.11782826483249664,
|
| 412 |
+
"eval_precision": 0.9680851063829787,
|
| 413 |
+
"eval_recall": 0.9578947368421052,
|
| 414 |
+
"eval_runtime": 1.277,
|
| 415 |
+
"eval_samples_per_second": 139.389,
|
| 416 |
+
"eval_steps_per_second": 4.698,
|
| 417 |
+
"step": 262
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 21.6,
|
| 421 |
+
"learning_rate": 1.388888888888889e-05,
|
| 422 |
+
"loss": 0.2294,
|
| 423 |
+
"step": 270
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 22.0,
|
| 427 |
+
"eval_accuracy": 0.9382022471910112,
|
| 428 |
+
"eval_f1_score": 0.9430051813471502,
|
| 429 |
+
"eval_loss": 0.12620893120765686,
|
| 430 |
+
"eval_precision": 0.9285714285714286,
|
| 431 |
+
"eval_recall": 0.9578947368421052,
|
| 432 |
+
"eval_runtime": 1.0278,
|
| 433 |
+
"eval_samples_per_second": 173.184,
|
| 434 |
+
"eval_steps_per_second": 5.838,
|
| 435 |
+
"step": 275
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 22.4,
|
| 439 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 440 |
+
"loss": 0.2296,
|
| 441 |
+
"step": 280
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 22.96,
|
| 445 |
"eval_accuracy": 0.9438202247191011,
|
| 446 |
+
"eval_f1_score": 0.9479166666666666,
|
| 447 |
+
"eval_loss": 0.12900537252426147,
|
| 448 |
+
"eval_precision": 0.9381443298969072,
|
| 449 |
+
"eval_recall": 0.9578947368421052,
|
| 450 |
+
"eval_runtime": 1.0136,
|
| 451 |
+
"eval_samples_per_second": 175.614,
|
| 452 |
+
"eval_steps_per_second": 5.92,
|
| 453 |
+
"step": 287
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 23.2,
|
| 457 |
+
"learning_rate": 1.0802469135802469e-05,
|
| 458 |
+
"loss": 0.2351,
|
| 459 |
+
"step": 290
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 24.0,
|
| 463 |
+
"learning_rate": 9.259259259259259e-06,
|
| 464 |
+
"loss": 0.2224,
|
| 465 |
+
"step": 300
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 24.0,
|
| 469 |
+
"eval_accuracy": 0.949438202247191,
|
| 470 |
+
"eval_f1_score": 0.9528795811518324,
|
| 471 |
+
"eval_loss": 0.1152968481183052,
|
| 472 |
+
"eval_precision": 0.9479166666666666,
|
| 473 |
+
"eval_recall": 0.9578947368421052,
|
| 474 |
+
"eval_runtime": 1.0182,
|
| 475 |
+
"eval_samples_per_second": 174.822,
|
| 476 |
+
"eval_steps_per_second": 5.893,
|
| 477 |
+
"step": 300
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 24.8,
|
| 481 |
+
"learning_rate": 7.71604938271605e-06,
|
| 482 |
+
"loss": 0.2205,
|
| 483 |
+
"step": 310
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 24.96,
|
| 487 |
+
"eval_accuracy": 0.949438202247191,
|
| 488 |
+
"eval_f1_score": 0.9528795811518324,
|
| 489 |
+
"eval_loss": 0.11498025059700012,
|
| 490 |
+
"eval_precision": 0.9479166666666666,
|
| 491 |
+
"eval_recall": 0.9578947368421052,
|
| 492 |
+
"eval_runtime": 1.0221,
|
| 493 |
+
"eval_samples_per_second": 174.15,
|
| 494 |
+
"eval_steps_per_second": 5.87,
|
| 495 |
+
"step": 312
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 25.6,
|
| 499 |
+
"learning_rate": 6.172839506172839e-06,
|
| 500 |
+
"loss": 0.2169,
|
| 501 |
+
"step": 320
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 26.0,
|
| 505 |
+
"eval_accuracy": 0.9550561797752809,
|
| 506 |
+
"eval_f1_score": 0.9574468085106383,
|
| 507 |
+
"eval_loss": 0.1120525524020195,
|
| 508 |
+
"eval_precision": 0.967741935483871,
|
| 509 |
+
"eval_recall": 0.9473684210526315,
|
| 510 |
+
"eval_runtime": 1.0268,
|
| 511 |
+
"eval_samples_per_second": 173.35,
|
| 512 |
+
"eval_steps_per_second": 5.843,
|
| 513 |
+
"step": 325
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 26.4,
|
| 517 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 518 |
+
"loss": 0.2212,
|
| 519 |
+
"step": 330
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 26.96,
|
| 523 |
+
"eval_accuracy": 0.949438202247191,
|
| 524 |
+
"eval_f1_score": 0.9528795811518324,
|
| 525 |
+
"eval_loss": 0.11452833563089371,
|
| 526 |
+
"eval_precision": 0.9479166666666666,
|
| 527 |
+
"eval_recall": 0.9578947368421052,
|
| 528 |
+
"eval_runtime": 1.951,
|
| 529 |
+
"eval_samples_per_second": 91.236,
|
| 530 |
+
"eval_steps_per_second": 3.075,
|
| 531 |
+
"step": 337
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 27.2,
|
| 535 |
+
"learning_rate": 3.0864197530864196e-06,
|
| 536 |
+
"loss": 0.2265,
|
| 537 |
+
"step": 340
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 28.0,
|
| 541 |
+
"learning_rate": 1.5432098765432098e-06,
|
| 542 |
+
"loss": 0.2188,
|
| 543 |
+
"step": 350
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 28.0,
|
| 547 |
+
"eval_accuracy": 0.949438202247191,
|
| 548 |
+
"eval_f1_score": 0.9523809523809523,
|
| 549 |
+
"eval_loss": 0.11309711635112762,
|
| 550 |
+
"eval_precision": 0.9574468085106383,
|
| 551 |
+
"eval_recall": 0.9473684210526315,
|
| 552 |
+
"eval_runtime": 1.0255,
|
| 553 |
+
"eval_samples_per_second": 173.568,
|
| 554 |
+
"eval_steps_per_second": 5.851,
|
| 555 |
+
"step": 350
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 28.8,
|
| 559 |
+
"learning_rate": 0.0,
|
| 560 |
+
"loss": 0.2015,
|
| 561 |
+
"step": 360
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 28.8,
|
| 565 |
+
"eval_accuracy": 0.949438202247191,
|
| 566 |
+
"eval_f1_score": 0.9523809523809523,
|
| 567 |
+
"eval_loss": 0.11295311897993088,
|
| 568 |
+
"eval_precision": 0.9574468085106383,
|
| 569 |
+
"eval_recall": 0.9473684210526315,
|
| 570 |
+
"eval_runtime": 1.3178,
|
| 571 |
+
"eval_samples_per_second": 135.073,
|
| 572 |
+
"eval_steps_per_second": 4.553,
|
| 573 |
+
"step": 360
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 28.8,
|
| 577 |
+
"step": 360,
|
| 578 |
+
"total_flos": 2.0837941191062323e+17,
|
| 579 |
+
"train_loss": 0.3229101174407535,
|
| 580 |
+
"train_runtime": 715.5601,
|
| 581 |
+
"train_samples_per_second": 66.787,
|
| 582 |
+
"train_steps_per_second": 0.503
|
| 583 |
}
|
| 584 |
],
|
| 585 |
"logging_steps": 10,
|
| 586 |
+
"max_steps": 360,
|
| 587 |
"num_input_tokens_seen": 0,
|
| 588 |
+
"num_train_epochs": 30,
|
| 589 |
"save_steps": 500,
|
| 590 |
+
"total_flos": 2.0837941191062323e+17,
|
| 591 |
"train_batch_size": 32,
|
| 592 |
"trial_name": null,
|
| 593 |
"trial_params": null
|