VIT_SEIZURE_231112
Browse files- README.md +5 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Nov13_00-09-36_ed98c7365e8b/events.out.tfevents.1699843803.ed98c7365e8b +3 -0
- train_results.json +8 -0
- trainer_state.json +703 -0
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
model-index:
|
| 7 |
- name: seizure_vit_jlb_231112_fft_raw_combo
|
|
@@ -13,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# seizure_vit_jlb_231112_fft_raw_combo
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
-
- Roc Auc: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- seizure-detection
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: seizure_vit_jlb_231112_fft_raw_combo
|
|
|
|
| 15 |
|
| 16 |
# seizure_vit_jlb_231112_fft_raw_combo
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the JLB-JLB/seizure_detection_224x224_raw_frequency dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.4822
|
| 21 |
+
- Roc Auc: 0.7667
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.4821653366088867,
|
| 4 |
+
"eval_roc_auc": 0.7666592334506762,
|
| 5 |
+
"eval_runtime": 520.8761,
|
| 6 |
+
"eval_samples_per_second": 60.252,
|
| 7 |
+
"eval_steps_per_second": 7.532,
|
| 8 |
+
"total_flos": 2.165002202648317e+19,
|
| 9 |
+
"train_loss": 0.34656610874380017,
|
| 10 |
+
"train_runtime": 7351.286,
|
| 11 |
+
"train_samples_per_second": 38.005,
|
| 12 |
+
"train_steps_per_second": 1.188
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.4821653366088867,
|
| 4 |
+
"eval_roc_auc": 0.7666592334506762,
|
| 5 |
+
"eval_runtime": 520.8761,
|
| 6 |
+
"eval_samples_per_second": 60.252,
|
| 7 |
+
"eval_steps_per_second": 7.532
|
| 8 |
+
}
|
runs/Nov13_00-09-36_ed98c7365e8b/events.out.tfevents.1699843803.ed98c7365e8b
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ea2f1600351faedb02434e4cfc73d805464902ec0c9f70c63e15530095002fd
|
| 3 |
+
size 40
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.165002202648317e+19,
|
| 4 |
+
"train_loss": 0.34656610874380017,
|
| 5 |
+
"train_runtime": 7351.286,
|
| 6 |
+
"train_samples_per_second": 38.005,
|
| 7 |
+
"train_steps_per_second": 1.188
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,703 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5043315291404724,
|
| 3 |
+
"best_model_checkpoint": "seizure_vit/seizure_vit_jlb_231112_fft_raw_combo/checkpoint-2000",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8733,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 1.977098362532921e-06,
|
| 14 |
+
"loss": 0.661,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 1.954196725065842e-06,
|
| 20 |
+
"loss": 0.5647,
|
| 21 |
+
"step": 200
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.1,
|
| 25 |
+
"learning_rate": 1.931295087598763e-06,
|
| 26 |
+
"loss": 0.5265,
|
| 27 |
+
"step": 300
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.14,
|
| 31 |
+
"learning_rate": 1.9083934501316844e-06,
|
| 32 |
+
"loss": 0.5186,
|
| 33 |
+
"step": 400
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.17,
|
| 37 |
+
"learning_rate": 1.8854918126646053e-06,
|
| 38 |
+
"loss": 0.4777,
|
| 39 |
+
"step": 500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.17,
|
| 43 |
+
"eval_loss": 0.5236575603485107,
|
| 44 |
+
"eval_roc_auc": 0.7454533473906167,
|
| 45 |
+
"eval_runtime": 252.0313,
|
| 46 |
+
"eval_samples_per_second": 60.215,
|
| 47 |
+
"eval_steps_per_second": 7.527,
|
| 48 |
+
"step": 500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.21,
|
| 52 |
+
"learning_rate": 1.8625901751975266e-06,
|
| 53 |
+
"loss": 0.4781,
|
| 54 |
+
"step": 600
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.24,
|
| 58 |
+
"learning_rate": 1.8396885377304477e-06,
|
| 59 |
+
"loss": 0.4512,
|
| 60 |
+
"step": 700
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.27,
|
| 64 |
+
"learning_rate": 1.8167869002633686e-06,
|
| 65 |
+
"loss": 0.4484,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.31,
|
| 70 |
+
"learning_rate": 1.79388526279629e-06,
|
| 71 |
+
"loss": 0.4431,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.34,
|
| 76 |
+
"learning_rate": 1.770983625329211e-06,
|
| 77 |
+
"loss": 0.4469,
|
| 78 |
+
"step": 1000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.34,
|
| 82 |
+
"eval_loss": 0.5114009380340576,
|
| 83 |
+
"eval_roc_auc": 0.7541512915129152,
|
| 84 |
+
"eval_runtime": 254.0044,
|
| 85 |
+
"eval_samples_per_second": 59.747,
|
| 86 |
+
"eval_steps_per_second": 7.468,
|
| 87 |
+
"step": 1000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.38,
|
| 91 |
+
"learning_rate": 1.748081987862132e-06,
|
| 92 |
+
"loss": 0.4332,
|
| 93 |
+
"step": 1100
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.41,
|
| 97 |
+
"learning_rate": 1.7251803503950533e-06,
|
| 98 |
+
"loss": 0.4353,
|
| 99 |
+
"step": 1200
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.45,
|
| 103 |
+
"learning_rate": 1.7022787129279744e-06,
|
| 104 |
+
"loss": 0.4137,
|
| 105 |
+
"step": 1300
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.48,
|
| 109 |
+
"learning_rate": 1.6793770754608953e-06,
|
| 110 |
+
"loss": 0.4306,
|
| 111 |
+
"step": 1400
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.52,
|
| 115 |
+
"learning_rate": 1.6564754379938164e-06,
|
| 116 |
+
"loss": 0.4122,
|
| 117 |
+
"step": 1500
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.52,
|
| 121 |
+
"eval_loss": 0.5084041953086853,
|
| 122 |
+
"eval_roc_auc": 0.7566552451238797,
|
| 123 |
+
"eval_runtime": 251.639,
|
| 124 |
+
"eval_samples_per_second": 60.309,
|
| 125 |
+
"eval_steps_per_second": 7.539,
|
| 126 |
+
"step": 1500
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.55,
|
| 130 |
+
"learning_rate": 1.6335738005267376e-06,
|
| 131 |
+
"loss": 0.3944,
|
| 132 |
+
"step": 1600
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.58,
|
| 136 |
+
"learning_rate": 1.6106721630596587e-06,
|
| 137 |
+
"loss": 0.4001,
|
| 138 |
+
"step": 1700
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.62,
|
| 142 |
+
"learning_rate": 1.5877705255925798e-06,
|
| 143 |
+
"loss": 0.3981,
|
| 144 |
+
"step": 1800
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.65,
|
| 148 |
+
"learning_rate": 1.564868888125501e-06,
|
| 149 |
+
"loss": 0.3807,
|
| 150 |
+
"step": 1900
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.69,
|
| 154 |
+
"learning_rate": 1.541967250658422e-06,
|
| 155 |
+
"loss": 0.3904,
|
| 156 |
+
"step": 2000
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.69,
|
| 160 |
+
"eval_loss": 0.5043315291404724,
|
| 161 |
+
"eval_roc_auc": 0.7610701107011071,
|
| 162 |
+
"eval_runtime": 253.2239,
|
| 163 |
+
"eval_samples_per_second": 59.931,
|
| 164 |
+
"eval_steps_per_second": 7.491,
|
| 165 |
+
"step": 2000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.72,
|
| 169 |
+
"learning_rate": 1.5190656131913431e-06,
|
| 170 |
+
"loss": 0.3841,
|
| 171 |
+
"step": 2100
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.76,
|
| 175 |
+
"learning_rate": 1.496163975724264e-06,
|
| 176 |
+
"loss": 0.367,
|
| 177 |
+
"step": 2200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.79,
|
| 181 |
+
"learning_rate": 1.4732623382571854e-06,
|
| 182 |
+
"loss": 0.3748,
|
| 183 |
+
"step": 2300
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.82,
|
| 187 |
+
"learning_rate": 1.4503607007901065e-06,
|
| 188 |
+
"loss": 0.3775,
|
| 189 |
+
"step": 2400
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.86,
|
| 193 |
+
"learning_rate": 1.4274590633230274e-06,
|
| 194 |
+
"loss": 0.3619,
|
| 195 |
+
"step": 2500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.86,
|
| 199 |
+
"eval_loss": 0.5282618999481201,
|
| 200 |
+
"eval_roc_auc": 0.7608724301528729,
|
| 201 |
+
"eval_runtime": 248.745,
|
| 202 |
+
"eval_samples_per_second": 61.01,
|
| 203 |
+
"eval_steps_per_second": 7.626,
|
| 204 |
+
"step": 2500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.89,
|
| 208 |
+
"learning_rate": 1.4045574258559487e-06,
|
| 209 |
+
"loss": 0.3557,
|
| 210 |
+
"step": 2600
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.93,
|
| 214 |
+
"learning_rate": 1.3816557883888698e-06,
|
| 215 |
+
"loss": 0.3551,
|
| 216 |
+
"step": 2700
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.96,
|
| 220 |
+
"learning_rate": 1.3587541509217907e-06,
|
| 221 |
+
"loss": 0.3539,
|
| 222 |
+
"step": 2800
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.0,
|
| 226 |
+
"learning_rate": 1.335852513454712e-06,
|
| 227 |
+
"loss": 0.3686,
|
| 228 |
+
"step": 2900
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.03,
|
| 232 |
+
"learning_rate": 1.3129508759876332e-06,
|
| 233 |
+
"loss": 0.3528,
|
| 234 |
+
"step": 3000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.03,
|
| 238 |
+
"eval_loss": 0.5351711511611938,
|
| 239 |
+
"eval_roc_auc": 0.7517132314180285,
|
| 240 |
+
"eval_runtime": 247.7427,
|
| 241 |
+
"eval_samples_per_second": 61.257,
|
| 242 |
+
"eval_steps_per_second": 7.657,
|
| 243 |
+
"step": 3000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.06,
|
| 247 |
+
"learning_rate": 1.290049238520554e-06,
|
| 248 |
+
"loss": 0.3515,
|
| 249 |
+
"step": 3100
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.1,
|
| 253 |
+
"learning_rate": 1.2671476010534752e-06,
|
| 254 |
+
"loss": 0.3481,
|
| 255 |
+
"step": 3200
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.13,
|
| 259 |
+
"learning_rate": 1.2442459635863965e-06,
|
| 260 |
+
"loss": 0.3509,
|
| 261 |
+
"step": 3300
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.17,
|
| 265 |
+
"learning_rate": 1.2213443261193174e-06,
|
| 266 |
+
"loss": 0.3486,
|
| 267 |
+
"step": 3400
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.2,
|
| 271 |
+
"learning_rate": 1.1984426886522385e-06,
|
| 272 |
+
"loss": 0.3445,
|
| 273 |
+
"step": 3500
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.2,
|
| 277 |
+
"eval_loss": 0.5338488221168518,
|
| 278 |
+
"eval_roc_auc": 0.757248286768582,
|
| 279 |
+
"eval_runtime": 258.0066,
|
| 280 |
+
"eval_samples_per_second": 58.82,
|
| 281 |
+
"eval_steps_per_second": 7.353,
|
| 282 |
+
"step": 3500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.24,
|
| 286 |
+
"learning_rate": 1.1755410511851599e-06,
|
| 287 |
+
"loss": 0.3215,
|
| 288 |
+
"step": 3600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.27,
|
| 292 |
+
"learning_rate": 1.1526394137180808e-06,
|
| 293 |
+
"loss": 0.3344,
|
| 294 |
+
"step": 3700
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.31,
|
| 298 |
+
"learning_rate": 1.1297377762510019e-06,
|
| 299 |
+
"loss": 0.3263,
|
| 300 |
+
"step": 3800
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.34,
|
| 304 |
+
"learning_rate": 1.106836138783923e-06,
|
| 305 |
+
"loss": 0.3255,
|
| 306 |
+
"step": 3900
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.37,
|
| 310 |
+
"learning_rate": 1.0839345013168441e-06,
|
| 311 |
+
"loss": 0.3221,
|
| 312 |
+
"step": 4000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.37,
|
| 316 |
+
"eval_loss": 0.5387703776359558,
|
| 317 |
+
"eval_roc_auc": 0.7508566157090143,
|
| 318 |
+
"eval_runtime": 249.4365,
|
| 319 |
+
"eval_samples_per_second": 60.841,
|
| 320 |
+
"eval_steps_per_second": 7.605,
|
| 321 |
+
"step": 4000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.41,
|
| 325 |
+
"learning_rate": 1.0610328638497652e-06,
|
| 326 |
+
"loss": 0.3388,
|
| 327 |
+
"step": 4100
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.44,
|
| 331 |
+
"learning_rate": 1.0381312263826861e-06,
|
| 332 |
+
"loss": 0.3365,
|
| 333 |
+
"step": 4200
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.48,
|
| 337 |
+
"learning_rate": 1.0152295889156075e-06,
|
| 338 |
+
"loss": 0.3256,
|
| 339 |
+
"step": 4300
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.51,
|
| 343 |
+
"learning_rate": 9.923279514485286e-07,
|
| 344 |
+
"loss": 0.3057,
|
| 345 |
+
"step": 4400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.55,
|
| 349 |
+
"learning_rate": 9.694263139814495e-07,
|
| 350 |
+
"loss": 0.3109,
|
| 351 |
+
"step": 4500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.55,
|
| 355 |
+
"eval_loss": 0.5640743374824524,
|
| 356 |
+
"eval_roc_auc": 0.7457828149710068,
|
| 357 |
+
"eval_runtime": 248.3252,
|
| 358 |
+
"eval_samples_per_second": 61.113,
|
| 359 |
+
"eval_steps_per_second": 7.639,
|
| 360 |
+
"step": 4500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.58,
|
| 364 |
+
"learning_rate": 9.465246765143707e-07,
|
| 365 |
+
"loss": 0.3301,
|
| 366 |
+
"step": 4600
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.61,
|
| 370 |
+
"learning_rate": 9.236230390472919e-07,
|
| 371 |
+
"loss": 0.309,
|
| 372 |
+
"step": 4700
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.65,
|
| 376 |
+
"learning_rate": 9.007214015802129e-07,
|
| 377 |
+
"loss": 0.3168,
|
| 378 |
+
"step": 4800
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.68,
|
| 382 |
+
"learning_rate": 8.778197641131341e-07,
|
| 383 |
+
"loss": 0.3337,
|
| 384 |
+
"step": 4900
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.72,
|
| 388 |
+
"learning_rate": 8.549181266460551e-07,
|
| 389 |
+
"loss": 0.3203,
|
| 390 |
+
"step": 5000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.72,
|
| 394 |
+
"eval_loss": 0.5403934717178345,
|
| 395 |
+
"eval_roc_auc": 0.757445967316816,
|
| 396 |
+
"eval_runtime": 249.3451,
|
| 397 |
+
"eval_samples_per_second": 60.863,
|
| 398 |
+
"eval_steps_per_second": 7.608,
|
| 399 |
+
"step": 5000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.75,
|
| 403 |
+
"learning_rate": 8.320164891789763e-07,
|
| 404 |
+
"loss": 0.3144,
|
| 405 |
+
"step": 5100
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.79,
|
| 409 |
+
"learning_rate": 8.091148517118974e-07,
|
| 410 |
+
"loss": 0.3166,
|
| 411 |
+
"step": 5200
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.82,
|
| 415 |
+
"learning_rate": 7.862132142448184e-07,
|
| 416 |
+
"loss": 0.3091,
|
| 417 |
+
"step": 5300
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.86,
|
| 421 |
+
"learning_rate": 7.633115767777396e-07,
|
| 422 |
+
"loss": 0.3097,
|
| 423 |
+
"step": 5400
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.89,
|
| 427 |
+
"learning_rate": 7.404099393106607e-07,
|
| 428 |
+
"loss": 0.294,
|
| 429 |
+
"step": 5500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.89,
|
| 433 |
+
"eval_loss": 0.5420622229576111,
|
| 434 |
+
"eval_roc_auc": 0.7563916710595677,
|
| 435 |
+
"eval_runtime": 247.7049,
|
| 436 |
+
"eval_samples_per_second": 61.266,
|
| 437 |
+
"eval_steps_per_second": 7.658,
|
| 438 |
+
"step": 5500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.92,
|
| 442 |
+
"learning_rate": 7.175083018435818e-07,
|
| 443 |
+
"loss": 0.2897,
|
| 444 |
+
"step": 5600
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.96,
|
| 448 |
+
"learning_rate": 6.946066643765029e-07,
|
| 449 |
+
"loss": 0.3126,
|
| 450 |
+
"step": 5700
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.99,
|
| 454 |
+
"learning_rate": 6.71705026909424e-07,
|
| 455 |
+
"loss": 0.2998,
|
| 456 |
+
"step": 5800
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.03,
|
| 460 |
+
"learning_rate": 6.488033894423451e-07,
|
| 461 |
+
"loss": 0.2941,
|
| 462 |
+
"step": 5900
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.06,
|
| 466 |
+
"learning_rate": 6.259017519752662e-07,
|
| 467 |
+
"loss": 0.2964,
|
| 468 |
+
"step": 6000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.06,
|
| 472 |
+
"eval_loss": 0.5581579208374023,
|
| 473 |
+
"eval_roc_auc": 0.7493410648392198,
|
| 474 |
+
"eval_runtime": 253.738,
|
| 475 |
+
"eval_samples_per_second": 59.81,
|
| 476 |
+
"eval_steps_per_second": 7.476,
|
| 477 |
+
"step": 6000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.1,
|
| 481 |
+
"learning_rate": 6.030001145081873e-07,
|
| 482 |
+
"loss": 0.296,
|
| 483 |
+
"step": 6100
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.13,
|
| 487 |
+
"learning_rate": 5.800984770411084e-07,
|
| 488 |
+
"loss": 0.3026,
|
| 489 |
+
"step": 6200
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.16,
|
| 493 |
+
"learning_rate": 5.571968395740295e-07,
|
| 494 |
+
"loss": 0.2958,
|
| 495 |
+
"step": 6300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.2,
|
| 499 |
+
"learning_rate": 5.342952021069507e-07,
|
| 500 |
+
"loss": 0.2778,
|
| 501 |
+
"step": 6400
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.23,
|
| 505 |
+
"learning_rate": 5.113935646398717e-07,
|
| 506 |
+
"loss": 0.292,
|
| 507 |
+
"step": 6500
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 2.23,
|
| 511 |
+
"eval_loss": 0.5513432621955872,
|
| 512 |
+
"eval_roc_auc": 0.7560622034791777,
|
| 513 |
+
"eval_runtime": 244.5312,
|
| 514 |
+
"eval_samples_per_second": 62.062,
|
| 515 |
+
"eval_steps_per_second": 7.758,
|
| 516 |
+
"step": 6500
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 2.27,
|
| 520 |
+
"learning_rate": 4.884919271727928e-07,
|
| 521 |
+
"loss": 0.294,
|
| 522 |
+
"step": 6600
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.3,
|
| 526 |
+
"learning_rate": 4.6559028970571394e-07,
|
| 527 |
+
"loss": 0.295,
|
| 528 |
+
"step": 6700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.34,
|
| 532 |
+
"learning_rate": 4.4268865223863505e-07,
|
| 533 |
+
"loss": 0.2914,
|
| 534 |
+
"step": 6800
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.37,
|
| 538 |
+
"learning_rate": 4.1978701477155617e-07,
|
| 539 |
+
"loss": 0.2771,
|
| 540 |
+
"step": 6900
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.4,
|
| 544 |
+
"learning_rate": 3.9688537730447723e-07,
|
| 545 |
+
"loss": 0.2838,
|
| 546 |
+
"step": 7000
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.4,
|
| 550 |
+
"eval_loss": 0.5557020306587219,
|
| 551 |
+
"eval_roc_auc": 0.7597522403795467,
|
| 552 |
+
"eval_runtime": 252.0114,
|
| 553 |
+
"eval_samples_per_second": 60.219,
|
| 554 |
+
"eval_steps_per_second": 7.527,
|
| 555 |
+
"step": 7000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.44,
|
| 559 |
+
"learning_rate": 3.7398373983739835e-07,
|
| 560 |
+
"loss": 0.3156,
|
| 561 |
+
"step": 7100
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.47,
|
| 565 |
+
"learning_rate": 3.5108210237031946e-07,
|
| 566 |
+
"loss": 0.2875,
|
| 567 |
+
"step": 7200
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.51,
|
| 571 |
+
"learning_rate": 3.281804649032406e-07,
|
| 572 |
+
"loss": 0.2779,
|
| 573 |
+
"step": 7300
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.54,
|
| 577 |
+
"learning_rate": 3.052788274361617e-07,
|
| 578 |
+
"loss": 0.2959,
|
| 579 |
+
"step": 7400
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.58,
|
| 583 |
+
"learning_rate": 2.8237718996908276e-07,
|
| 584 |
+
"loss": 0.2736,
|
| 585 |
+
"step": 7500
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.58,
|
| 589 |
+
"eval_loss": 0.5514435172080994,
|
| 590 |
+
"eval_roc_auc": 0.760608856088561,
|
| 591 |
+
"eval_runtime": 250.5016,
|
| 592 |
+
"eval_samples_per_second": 60.582,
|
| 593 |
+
"eval_steps_per_second": 7.573,
|
| 594 |
+
"step": 7500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.61,
|
| 598 |
+
"learning_rate": 2.5947555250200387e-07,
|
| 599 |
+
"loss": 0.2989,
|
| 600 |
+
"step": 7600
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.65,
|
| 604 |
+
"learning_rate": 2.36573915034925e-07,
|
| 605 |
+
"loss": 0.2889,
|
| 606 |
+
"step": 7700
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.68,
|
| 610 |
+
"learning_rate": 2.1367227756784608e-07,
|
| 611 |
+
"loss": 0.2814,
|
| 612 |
+
"step": 7800
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.71,
|
| 616 |
+
"learning_rate": 1.907706401007672e-07,
|
| 617 |
+
"loss": 0.2917,
|
| 618 |
+
"step": 7900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.75,
|
| 622 |
+
"learning_rate": 1.678690026336883e-07,
|
| 623 |
+
"loss": 0.2922,
|
| 624 |
+
"step": 8000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.75,
|
| 628 |
+
"eval_loss": 0.550301730632782,
|
| 629 |
+
"eval_roc_auc": 0.7537559304164471,
|
| 630 |
+
"eval_runtime": 244.8445,
|
| 631 |
+
"eval_samples_per_second": 61.982,
|
| 632 |
+
"eval_steps_per_second": 7.748,
|
| 633 |
+
"step": 8000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 2.78,
|
| 637 |
+
"learning_rate": 1.449673651666094e-07,
|
| 638 |
+
"loss": 0.2588,
|
| 639 |
+
"step": 8100
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.82,
|
| 643 |
+
"learning_rate": 1.2206572769953052e-07,
|
| 644 |
+
"loss": 0.2947,
|
| 645 |
+
"step": 8200
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 2.85,
|
| 649 |
+
"learning_rate": 9.916409023245162e-08,
|
| 650 |
+
"loss": 0.2901,
|
| 651 |
+
"step": 8300
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 2.89,
|
| 655 |
+
"learning_rate": 7.626245276537272e-08,
|
| 656 |
+
"loss": 0.2827,
|
| 657 |
+
"step": 8400
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 2.92,
|
| 661 |
+
"learning_rate": 5.3360815298293824e-08,
|
| 662 |
+
"loss": 0.2699,
|
| 663 |
+
"step": 8500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.92,
|
| 667 |
+
"eval_loss": 0.553462564945221,
|
| 668 |
+
"eval_roc_auc": 0.757841328413284,
|
| 669 |
+
"eval_runtime": 246.3262,
|
| 670 |
+
"eval_samples_per_second": 61.609,
|
| 671 |
+
"eval_steps_per_second": 7.701,
|
| 672 |
+
"step": 8500
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.95,
|
| 676 |
+
"learning_rate": 3.0459177831214926e-08,
|
| 677 |
+
"loss": 0.2763,
|
| 678 |
+
"step": 8600
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.99,
|
| 682 |
+
"learning_rate": 7.557540364136036e-09,
|
| 683 |
+
"loss": 0.2881,
|
| 684 |
+
"step": 8700
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.0,
|
| 688 |
+
"step": 8733,
|
| 689 |
+
"total_flos": 2.165002202648317e+19,
|
| 690 |
+
"train_loss": 0.34656610874380017,
|
| 691 |
+
"train_runtime": 7351.286,
|
| 692 |
+
"train_samples_per_second": 38.005,
|
| 693 |
+
"train_steps_per_second": 1.188
|
| 694 |
+
}
|
| 695 |
+
],
|
| 696 |
+
"logging_steps": 100,
|
| 697 |
+
"max_steps": 8733,
|
| 698 |
+
"num_train_epochs": 3,
|
| 699 |
+
"save_steps": 500,
|
| 700 |
+
"total_flos": 2.165002202648317e+19,
|
| 701 |
+
"trial_name": null,
|
| 702 |
+
"trial_params": null
|
| 703 |
+
}
|