🍻 cheers
Browse files- README.md +7 -6
- all_results.json +16 -0
- eval_results.json +11 -0
- runs/Mar21_18-37-45_45242e0ea5c0/events.out.tfevents.1711049708.45242e0ea5c0.20273.28 +3 -0
- train_results.json +8 -0
- trainer_state.json +562 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -18,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
# vit-beta1-0.85
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 0.
|
| 24 |
-
- Accuracy: 0.
|
| 25 |
-
- Precision: 0.
|
| 26 |
-
- Recall: 0.
|
| 27 |
-
- F1: 0.
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 19 |
|
| 20 |
# vit-beta1-0.85
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.5102
|
| 25 |
+
- Accuracy: 0.8558
|
| 26 |
+
- Precision: 0.8568
|
| 27 |
+
- Recall: 0.8558
|
| 28 |
+
- F1: 0.8553
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 28.0,
|
| 3 |
+
"eval_accuracy": 0.855755894590846,
|
| 4 |
+
"eval_f1": 0.8553441136144744,
|
| 5 |
+
"eval_loss": 0.5101504325866699,
|
| 6 |
+
"eval_precision": 0.8568218210572909,
|
| 7 |
+
"eval_recall": 0.855755894590846,
|
| 8 |
+
"eval_runtime": 23.4289,
|
| 9 |
+
"eval_samples_per_second": 123.096,
|
| 10 |
+
"eval_steps_per_second": 15.408,
|
| 11 |
+
"total_flos": 1.1127108458244538e+19,
|
| 12 |
+
"train_loss": 0.5253127112513815,
|
| 13 |
+
"train_runtime": 3374.2787,
|
| 14 |
+
"train_samples_per_second": 151.973,
|
| 15 |
+
"train_steps_per_second": 9.513
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 28.0,
|
| 3 |
+
"eval_accuracy": 0.855755894590846,
|
| 4 |
+
"eval_f1": 0.8553441136144744,
|
| 5 |
+
"eval_loss": 0.5101504325866699,
|
| 6 |
+
"eval_precision": 0.8568218210572909,
|
| 7 |
+
"eval_recall": 0.855755894590846,
|
| 8 |
+
"eval_runtime": 23.4289,
|
| 9 |
+
"eval_samples_per_second": 123.096,
|
| 10 |
+
"eval_steps_per_second": 15.408
|
| 11 |
+
}
|
runs/Mar21_18-37-45_45242e0ea5c0/events.out.tfevents.1711049708.45242e0ea5c0.20273.28
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4549719e01b3acde1fc37e105922335bfcafae8032476e2e4042bd632cc3fa2
|
| 3 |
+
size 560
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 28.0,
|
| 3 |
+
"total_flos": 1.1127108458244538e+19,
|
| 4 |
+
"train_loss": 0.5253127112513815,
|
| 5 |
+
"train_runtime": 3374.2787,
|
| 6 |
+
"train_samples_per_second": 151.973,
|
| 7 |
+
"train_steps_per_second": 9.513
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,562 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5101504325866699,
|
| 3 |
+
"best_model_checkpoint": "./vit-beta1-0.85/checkpoint-5778",
|
| 4 |
+
"epoch": 28.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8988,
|
| 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 |
+
"grad_norm": 23.742420196533203,
|
| 14 |
+
"learning_rate": 1.8291979226774382e-05,
|
| 15 |
+
"loss": 1.763,
|
| 16 |
+
"step": 321
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.6952149791955617,
|
| 21 |
+
"eval_f1": 0.6201672138002952,
|
| 22 |
+
"eval_loss": 0.9504804015159607,
|
| 23 |
+
"eval_precision": 0.6345772764957709,
|
| 24 |
+
"eval_recall": 0.6952149791955617,
|
| 25 |
+
"eval_runtime": 23.0549,
|
| 26 |
+
"eval_samples_per_second": 125.093,
|
| 27 |
+
"eval_steps_per_second": 15.658,
|
| 28 |
+
"step": 321
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"grad_norm": 20.97748374938965,
|
| 33 |
+
"learning_rate": 3.681477207155222e-05,
|
| 34 |
+
"loss": 1.149,
|
| 35 |
+
"step": 642
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.0,
|
| 39 |
+
"eval_accuracy": 0.7444521497919556,
|
| 40 |
+
"eval_f1": 0.7229775407881395,
|
| 41 |
+
"eval_loss": 0.714677631855011,
|
| 42 |
+
"eval_precision": 0.74566856205471,
|
| 43 |
+
"eval_recall": 0.7444521497919556,
|
| 44 |
+
"eval_runtime": 23.0018,
|
| 45 |
+
"eval_samples_per_second": 125.381,
|
| 46 |
+
"eval_steps_per_second": 15.694,
|
| 47 |
+
"step": 642
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"grad_norm": 8.279248237609863,
|
| 52 |
+
"learning_rate": 5.5337564916330066e-05,
|
| 53 |
+
"loss": 1.0452,
|
| 54 |
+
"step": 963
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_accuracy": 0.7572815533980582,
|
| 59 |
+
"eval_f1": 0.7320657787389255,
|
| 60 |
+
"eval_loss": 0.6250461935997009,
|
| 61 |
+
"eval_precision": 0.7591347407203577,
|
| 62 |
+
"eval_recall": 0.7572815533980582,
|
| 63 |
+
"eval_runtime": 23.1015,
|
| 64 |
+
"eval_samples_per_second": 124.841,
|
| 65 |
+
"eval_steps_per_second": 15.627,
|
| 66 |
+
"step": 963
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.0,
|
| 70 |
+
"grad_norm": 41.02663803100586,
|
| 71 |
+
"learning_rate": 7.386035776110792e-05,
|
| 72 |
+
"loss": 1.0048,
|
| 73 |
+
"step": 1284
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.0,
|
| 77 |
+
"eval_accuracy": 0.7784327323162274,
|
| 78 |
+
"eval_f1": 0.7737493702586475,
|
| 79 |
+
"eval_loss": 0.5614432692527771,
|
| 80 |
+
"eval_precision": 0.779177150378666,
|
| 81 |
+
"eval_recall": 0.7784327323162274,
|
| 82 |
+
"eval_runtime": 23.126,
|
| 83 |
+
"eval_samples_per_second": 124.708,
|
| 84 |
+
"eval_steps_per_second": 15.61,
|
| 85 |
+
"step": 1284
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.0,
|
| 89 |
+
"grad_norm": 7.507669448852539,
|
| 90 |
+
"learning_rate": 9.238315060588575e-05,
|
| 91 |
+
"loss": 0.931,
|
| 92 |
+
"step": 1605
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.0,
|
| 96 |
+
"eval_accuracy": 0.7739251040221914,
|
| 97 |
+
"eval_f1": 0.7822626472987246,
|
| 98 |
+
"eval_loss": 0.6081557869911194,
|
| 99 |
+
"eval_precision": 0.8019901444431083,
|
| 100 |
+
"eval_recall": 0.7739251040221914,
|
| 101 |
+
"eval_runtime": 22.3445,
|
| 102 |
+
"eval_samples_per_second": 129.07,
|
| 103 |
+
"eval_steps_per_second": 16.156,
|
| 104 |
+
"step": 1605
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.0,
|
| 108 |
+
"grad_norm": 8.154308319091797,
|
| 109 |
+
"learning_rate": 9.984715255878176e-05,
|
| 110 |
+
"loss": 0.9808,
|
| 111 |
+
"step": 1926
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.0,
|
| 115 |
+
"eval_accuracy": 0.7981969486823856,
|
| 116 |
+
"eval_f1": 0.7950945949242283,
|
| 117 |
+
"eval_loss": 0.554189145565033,
|
| 118 |
+
"eval_precision": 0.7983952215842433,
|
| 119 |
+
"eval_recall": 0.7981969486823856,
|
| 120 |
+
"eval_runtime": 22.8586,
|
| 121 |
+
"eval_samples_per_second": 126.167,
|
| 122 |
+
"eval_steps_per_second": 15.793,
|
| 123 |
+
"step": 1926
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 7.0,
|
| 127 |
+
"grad_norm": 11.704543113708496,
|
| 128 |
+
"learning_rate": 9.889061131437471e-05,
|
| 129 |
+
"loss": 0.8908,
|
| 130 |
+
"step": 2247
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 7.0,
|
| 134 |
+
"eval_accuracy": 0.7545076282940361,
|
| 135 |
+
"eval_f1": 0.7708612112738275,
|
| 136 |
+
"eval_loss": 0.5956947207450867,
|
| 137 |
+
"eval_precision": 0.820205289282276,
|
| 138 |
+
"eval_recall": 0.7545076282940361,
|
| 139 |
+
"eval_runtime": 23.7447,
|
| 140 |
+
"eval_samples_per_second": 121.459,
|
| 141 |
+
"eval_steps_per_second": 15.203,
|
| 142 |
+
"step": 2247
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.0,
|
| 146 |
+
"grad_norm": 3.4443271160125732,
|
| 147 |
+
"learning_rate": 9.707265436104638e-05,
|
| 148 |
+
"loss": 0.7747,
|
| 149 |
+
"step": 2568
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 8.0,
|
| 153 |
+
"eval_accuracy": 0.7694174757281553,
|
| 154 |
+
"eval_f1": 0.7836405012298657,
|
| 155 |
+
"eval_loss": 0.5766288638114929,
|
| 156 |
+
"eval_precision": 0.8155227380188221,
|
| 157 |
+
"eval_recall": 0.7694174757281553,
|
| 158 |
+
"eval_runtime": 23.1168,
|
| 159 |
+
"eval_samples_per_second": 124.758,
|
| 160 |
+
"eval_steps_per_second": 15.616,
|
| 161 |
+
"step": 2568
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.0,
|
| 165 |
+
"grad_norm": 7.433899879455566,
|
| 166 |
+
"learning_rate": 9.44253127296151e-05,
|
| 167 |
+
"loss": 0.741,
|
| 168 |
+
"step": 2889
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 9.0,
|
| 172 |
+
"eval_accuracy": 0.7995839112343966,
|
| 173 |
+
"eval_f1": 0.804684275131234,
|
| 174 |
+
"eval_loss": 0.5430790185928345,
|
| 175 |
+
"eval_precision": 0.8189836827516774,
|
| 176 |
+
"eval_recall": 0.7995839112343966,
|
| 177 |
+
"eval_runtime": 23.7114,
|
| 178 |
+
"eval_samples_per_second": 121.629,
|
| 179 |
+
"eval_steps_per_second": 15.225,
|
| 180 |
+
"step": 2889
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 10.0,
|
| 184 |
+
"grad_norm": 12.33841609954834,
|
| 185 |
+
"learning_rate": 9.099523058358976e-05,
|
| 186 |
+
"loss": 0.7179,
|
| 187 |
+
"step": 3210
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 10.0,
|
| 191 |
+
"eval_accuracy": 0.7773925104022191,
|
| 192 |
+
"eval_f1": 0.7904480536083208,
|
| 193 |
+
"eval_loss": 0.5864874720573425,
|
| 194 |
+
"eval_precision": 0.831331916729736,
|
| 195 |
+
"eval_recall": 0.7773925104022191,
|
| 196 |
+
"eval_runtime": 23.035,
|
| 197 |
+
"eval_samples_per_second": 125.201,
|
| 198 |
+
"eval_steps_per_second": 15.672,
|
| 199 |
+
"step": 3210
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 11.0,
|
| 203 |
+
"grad_norm": 21.430646896362305,
|
| 204 |
+
"learning_rate": 8.684284338417735e-05,
|
| 205 |
+
"loss": 0.6102,
|
| 206 |
+
"step": 3531
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 11.0,
|
| 210 |
+
"eval_accuracy": 0.8096393897364771,
|
| 211 |
+
"eval_f1": 0.8180305563212311,
|
| 212 |
+
"eval_loss": 0.5288417935371399,
|
| 213 |
+
"eval_precision": 0.8361295244369513,
|
| 214 |
+
"eval_recall": 0.8096393897364771,
|
| 215 |
+
"eval_runtime": 23.7358,
|
| 216 |
+
"eval_samples_per_second": 121.504,
|
| 217 |
+
"eval_steps_per_second": 15.209,
|
| 218 |
+
"step": 3531
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 12.0,
|
| 222 |
+
"grad_norm": 75.45726013183594,
|
| 223 |
+
"learning_rate": 8.204131306302357e-05,
|
| 224 |
+
"loss": 0.574,
|
| 225 |
+
"step": 3852
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 12.0,
|
| 229 |
+
"eval_accuracy": 0.7995839112343966,
|
| 230 |
+
"eval_f1": 0.8096217027832384,
|
| 231 |
+
"eval_loss": 0.599109947681427,
|
| 232 |
+
"eval_precision": 0.8332319256993149,
|
| 233 |
+
"eval_recall": 0.7995839112343966,
|
| 234 |
+
"eval_runtime": 23.369,
|
| 235 |
+
"eval_samples_per_second": 123.411,
|
| 236 |
+
"eval_steps_per_second": 15.448,
|
| 237 |
+
"step": 3852
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 13.0,
|
| 241 |
+
"grad_norm": 3.8113694190979004,
|
| 242 |
+
"learning_rate": 7.667523896413962e-05,
|
| 243 |
+
"loss": 0.4515,
|
| 244 |
+
"step": 4173
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 13.0,
|
| 248 |
+
"eval_accuracy": 0.8370319001386962,
|
| 249 |
+
"eval_f1": 0.8292572001875775,
|
| 250 |
+
"eval_loss": 0.5889743566513062,
|
| 251 |
+
"eval_precision": 0.8333558129412982,
|
| 252 |
+
"eval_recall": 0.8370319001386962,
|
| 253 |
+
"eval_runtime": 23.6944,
|
| 254 |
+
"eval_samples_per_second": 121.717,
|
| 255 |
+
"eval_steps_per_second": 15.236,
|
| 256 |
+
"step": 4173
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 14.0,
|
| 260 |
+
"grad_norm": 5.727066516876221,
|
| 261 |
+
"learning_rate": 7.083916726724684e-05,
|
| 262 |
+
"loss": 0.4629,
|
| 263 |
+
"step": 4494
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 14.0,
|
| 267 |
+
"eval_accuracy": 0.8120665742024965,
|
| 268 |
+
"eval_f1": 0.8205347890470436,
|
| 269 |
+
"eval_loss": 0.5573027729988098,
|
| 270 |
+
"eval_precision": 0.8462670503631797,
|
| 271 |
+
"eval_recall": 0.8120665742024965,
|
| 272 |
+
"eval_runtime": 24.8143,
|
| 273 |
+
"eval_samples_per_second": 116.223,
|
| 274 |
+
"eval_steps_per_second": 14.548,
|
| 275 |
+
"step": 4494
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 15.0,
|
| 279 |
+
"grad_norm": 4.605546474456787,
|
| 280 |
+
"learning_rate": 6.463592515537568e-05,
|
| 281 |
+
"loss": 0.3927,
|
| 282 |
+
"step": 4815
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 15.0,
|
| 286 |
+
"eval_accuracy": 0.8332177531206657,
|
| 287 |
+
"eval_f1": 0.835722282090202,
|
| 288 |
+
"eval_loss": 0.5279321670532227,
|
| 289 |
+
"eval_precision": 0.8506448765217454,
|
| 290 |
+
"eval_recall": 0.8332177531206657,
|
| 291 |
+
"eval_runtime": 25.6733,
|
| 292 |
+
"eval_samples_per_second": 112.335,
|
| 293 |
+
"eval_steps_per_second": 14.061,
|
| 294 |
+
"step": 4815
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 16.0,
|
| 298 |
+
"grad_norm": 5.051961898803711,
|
| 299 |
+
"learning_rate": 5.8174809077430184e-05,
|
| 300 |
+
"loss": 0.3535,
|
| 301 |
+
"step": 5136
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 16.0,
|
| 305 |
+
"eval_accuracy": 0.8356449375866851,
|
| 306 |
+
"eval_f1": 0.8404866875551775,
|
| 307 |
+
"eval_loss": 0.5363761782646179,
|
| 308 |
+
"eval_precision": 0.8494425873765888,
|
| 309 |
+
"eval_recall": 0.8356449375866851,
|
| 310 |
+
"eval_runtime": 23.6271,
|
| 311 |
+
"eval_samples_per_second": 122.063,
|
| 312 |
+
"eval_steps_per_second": 15.279,
|
| 313 |
+
"step": 5136
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 17.0,
|
| 317 |
+
"grad_norm": 28.548500061035156,
|
| 318 |
+
"learning_rate": 5.156965902716534e-05,
|
| 319 |
+
"loss": 0.2635,
|
| 320 |
+
"step": 5457
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 17.0,
|
| 324 |
+
"eval_accuracy": 0.8547156726768377,
|
| 325 |
+
"eval_f1": 0.853209116290087,
|
| 326 |
+
"eval_loss": 0.5475490093231201,
|
| 327 |
+
"eval_precision": 0.8625841938606562,
|
| 328 |
+
"eval_recall": 0.8547156726768377,
|
| 329 |
+
"eval_runtime": 25.2986,
|
| 330 |
+
"eval_samples_per_second": 113.999,
|
| 331 |
+
"eval_steps_per_second": 14.27,
|
| 332 |
+
"step": 5457
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 18.0,
|
| 336 |
+
"grad_norm": 16.80175018310547,
|
| 337 |
+
"learning_rate": 4.493685276832998e-05,
|
| 338 |
+
"loss": 0.2493,
|
| 339 |
+
"step": 5778
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 18.0,
|
| 343 |
+
"eval_accuracy": 0.855755894590846,
|
| 344 |
+
"eval_f1": 0.8553441136144744,
|
| 345 |
+
"eval_loss": 0.5101504325866699,
|
| 346 |
+
"eval_precision": 0.8568218210572909,
|
| 347 |
+
"eval_recall": 0.855755894590846,
|
| 348 |
+
"eval_runtime": 23.831,
|
| 349 |
+
"eval_samples_per_second": 121.019,
|
| 350 |
+
"eval_steps_per_second": 15.148,
|
| 351 |
+
"step": 5778
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 19.0,
|
| 355 |
+
"grad_norm": 0.11270666122436523,
|
| 356 |
+
"learning_rate": 3.839325534621579e-05,
|
| 357 |
+
"loss": 0.2125,
|
| 358 |
+
"step": 6099
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 19.0,
|
| 362 |
+
"eval_accuracy": 0.8328710124826629,
|
| 363 |
+
"eval_f1": 0.8418273768641612,
|
| 364 |
+
"eval_loss": 0.6120281219482422,
|
| 365 |
+
"eval_precision": 0.8622992078235902,
|
| 366 |
+
"eval_recall": 0.8328710124826629,
|
| 367 |
+
"eval_runtime": 23.1564,
|
| 368 |
+
"eval_samples_per_second": 124.544,
|
| 369 |
+
"eval_steps_per_second": 15.59,
|
| 370 |
+
"step": 6099
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 20.0,
|
| 374 |
+
"grad_norm": 0.06091728433966637,
|
| 375 |
+
"learning_rate": 3.205416001367289e-05,
|
| 376 |
+
"loss": 0.2179,
|
| 377 |
+
"step": 6420
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 20.0,
|
| 381 |
+
"eval_accuracy": 0.8567961165048543,
|
| 382 |
+
"eval_f1": 0.85627072801028,
|
| 383 |
+
"eval_loss": 0.5720838904380798,
|
| 384 |
+
"eval_precision": 0.8563063857199564,
|
| 385 |
+
"eval_recall": 0.8567961165048543,
|
| 386 |
+
"eval_runtime": 23.124,
|
| 387 |
+
"eval_samples_per_second": 124.719,
|
| 388 |
+
"eval_steps_per_second": 15.611,
|
| 389 |
+
"step": 6420
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 21.0,
|
| 393 |
+
"grad_norm": 2.7066216468811035,
|
| 394 |
+
"learning_rate": 2.6067582030088143e-05,
|
| 395 |
+
"loss": 0.1598,
|
| 396 |
+
"step": 6741
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 21.0,
|
| 400 |
+
"eval_accuracy": 0.8651178918169209,
|
| 401 |
+
"eval_f1": 0.8632510833215378,
|
| 402 |
+
"eval_loss": 0.5503374338150024,
|
| 403 |
+
"eval_precision": 0.8623213572654986,
|
| 404 |
+
"eval_recall": 0.8651178918169209,
|
| 405 |
+
"eval_runtime": 23.2584,
|
| 406 |
+
"eval_samples_per_second": 123.998,
|
| 407 |
+
"eval_steps_per_second": 15.521,
|
| 408 |
+
"step": 6741
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 22.0,
|
| 412 |
+
"grad_norm": 37.88532257080078,
|
| 413 |
+
"learning_rate": 2.0464044677195966e-05,
|
| 414 |
+
"loss": 0.1194,
|
| 415 |
+
"step": 7062
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 22.0,
|
| 419 |
+
"eval_accuracy": 0.8678918169209431,
|
| 420 |
+
"eval_f1": 0.8668972179399497,
|
| 421 |
+
"eval_loss": 0.5829476714134216,
|
| 422 |
+
"eval_precision": 0.8672496533540811,
|
| 423 |
+
"eval_recall": 0.8678918169209431,
|
| 424 |
+
"eval_runtime": 23.2515,
|
| 425 |
+
"eval_samples_per_second": 124.035,
|
| 426 |
+
"eval_steps_per_second": 15.526,
|
| 427 |
+
"step": 7062
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 23.0,
|
| 431 |
+
"grad_norm": 0.11761299520730972,
|
| 432 |
+
"learning_rate": 1.538090858763548e-05,
|
| 433 |
+
"loss": 0.1245,
|
| 434 |
+
"step": 7383
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 23.0,
|
| 438 |
+
"eval_accuracy": 0.8682385575589459,
|
| 439 |
+
"eval_f1": 0.8629371653200789,
|
| 440 |
+
"eval_loss": 0.6137722134590149,
|
| 441 |
+
"eval_precision": 0.8631764321501026,
|
| 442 |
+
"eval_recall": 0.8682385575589459,
|
| 443 |
+
"eval_runtime": 24.6364,
|
| 444 |
+
"eval_samples_per_second": 117.063,
|
| 445 |
+
"eval_steps_per_second": 14.653,
|
| 446 |
+
"step": 7383
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 24.0,
|
| 450 |
+
"grad_norm": 0.2003249078989029,
|
| 451 |
+
"learning_rate": 1.0907734786732799e-05,
|
| 452 |
+
"loss": 0.1239,
|
| 453 |
+
"step": 7704
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 24.0,
|
| 457 |
+
"eval_accuracy": 0.8730929264909847,
|
| 458 |
+
"eval_f1": 0.8694766569722234,
|
| 459 |
+
"eval_loss": 0.6135957837104797,
|
| 460 |
+
"eval_precision": 0.8702946096941098,
|
| 461 |
+
"eval_recall": 0.8730929264909847,
|
| 462 |
+
"eval_runtime": 23.1398,
|
| 463 |
+
"eval_samples_per_second": 124.633,
|
| 464 |
+
"eval_steps_per_second": 15.601,
|
| 465 |
+
"step": 7704
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 25.0,
|
| 469 |
+
"grad_norm": 0.018545789644122124,
|
| 470 |
+
"learning_rate": 7.123337224185811e-06,
|
| 471 |
+
"loss": 0.1159,
|
| 472 |
+
"step": 8025
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 25.0,
|
| 476 |
+
"eval_accuracy": 0.8751733703190014,
|
| 477 |
+
"eval_f1": 0.8726121524469985,
|
| 478 |
+
"eval_loss": 0.5931476354598999,
|
| 479 |
+
"eval_precision": 0.8724487380217695,
|
| 480 |
+
"eval_recall": 0.8751733703190014,
|
| 481 |
+
"eval_runtime": 23.8052,
|
| 482 |
+
"eval_samples_per_second": 121.15,
|
| 483 |
+
"eval_steps_per_second": 15.165,
|
| 484 |
+
"step": 8025
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 26.0,
|
| 488 |
+
"grad_norm": 13.244468688964844,
|
| 489 |
+
"learning_rate": 4.094394131694684e-06,
|
| 490 |
+
"loss": 0.089,
|
| 491 |
+
"step": 8346
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 26.0,
|
| 495 |
+
"eval_accuracy": 0.8776005547850209,
|
| 496 |
+
"eval_f1": 0.8749752761914638,
|
| 497 |
+
"eval_loss": 0.5846670866012573,
|
| 498 |
+
"eval_precision": 0.8742768994280256,
|
| 499 |
+
"eval_recall": 0.8776005547850209,
|
| 500 |
+
"eval_runtime": 23.0803,
|
| 501 |
+
"eval_samples_per_second": 124.955,
|
| 502 |
+
"eval_steps_per_second": 15.641,
|
| 503 |
+
"step": 8346
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 27.0,
|
| 507 |
+
"grad_norm": 0.4510207176208496,
|
| 508 |
+
"learning_rate": 1.8742732027931087e-06,
|
| 509 |
+
"loss": 0.1123,
|
| 510 |
+
"step": 8667
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 27.0,
|
| 514 |
+
"eval_accuracy": 0.8751733703190014,
|
| 515 |
+
"eval_f1": 0.8718954079920697,
|
| 516 |
+
"eval_loss": 0.594070553779602,
|
| 517 |
+
"eval_precision": 0.8710238593528318,
|
| 518 |
+
"eval_recall": 0.8751733703190014,
|
| 519 |
+
"eval_runtime": 22.9979,
|
| 520 |
+
"eval_samples_per_second": 125.403,
|
| 521 |
+
"eval_steps_per_second": 15.697,
|
| 522 |
+
"step": 8667
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 28.0,
|
| 526 |
+
"grad_norm": 0.05787363648414612,
|
| 527 |
+
"learning_rate": 5.020912943263345e-07,
|
| 528 |
+
"loss": 0.0779,
|
| 529 |
+
"step": 8988
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 28.0,
|
| 533 |
+
"eval_accuracy": 0.8765603328710125,
|
| 534 |
+
"eval_f1": 0.8729435556323918,
|
| 535 |
+
"eval_loss": 0.6037870049476624,
|
| 536 |
+
"eval_precision": 0.872156471246016,
|
| 537 |
+
"eval_recall": 0.8765603328710125,
|
| 538 |
+
"eval_runtime": 24.3339,
|
| 539 |
+
"eval_samples_per_second": 118.518,
|
| 540 |
+
"eval_steps_per_second": 14.835,
|
| 541 |
+
"step": 8988
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 28.0,
|
| 545 |
+
"step": 8988,
|
| 546 |
+
"total_flos": 1.1127108458244538e+19,
|
| 547 |
+
"train_loss": 0.5253127112513815,
|
| 548 |
+
"train_runtime": 3374.2787,
|
| 549 |
+
"train_samples_per_second": 151.973,
|
| 550 |
+
"train_steps_per_second": 9.513
|
| 551 |
+
}
|
| 552 |
+
],
|
| 553 |
+
"logging_steps": 500,
|
| 554 |
+
"max_steps": 32100,
|
| 555 |
+
"num_input_tokens_seen": 0,
|
| 556 |
+
"num_train_epochs": 100,
|
| 557 |
+
"save_steps": 500,
|
| 558 |
+
"total_flos": 1.1127108458244538e+19,
|
| 559 |
+
"train_batch_size": 16,
|
| 560 |
+
"trial_name": null,
|
| 561 |
+
"trial_params": null
|
| 562 |
+
}
|