🍻 cheers
Browse files- README.md +7 -6
- all_results.json +16 -0
- eval_results.json +11 -0
- runs/Mar22_11-30-39_80a90583e6d6/events.out.tfevents.1711109882.80a90583e6d6.2864.11 +3 -0
- train_results.json +8 -0
- trainer_state.json +467 -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-epsilon-1e-7
|
| 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-epsilon-1e-7
|
| 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: 0.5348
|
| 25 |
+
- Accuracy: 0.8350
|
| 26 |
+
- Precision: 0.8391
|
| 27 |
+
- Recall: 0.8350
|
| 28 |
+
- F1: 0.8350
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 23.0,
|
| 3 |
+
"eval_accuracy": 0.8349514563106796,
|
| 4 |
+
"eval_f1": 0.8349682338681513,
|
| 5 |
+
"eval_loss": 0.534845769405365,
|
| 6 |
+
"eval_precision": 0.8390873801645569,
|
| 7 |
+
"eval_recall": 0.8349514563106796,
|
| 8 |
+
"eval_runtime": 23.0016,
|
| 9 |
+
"eval_samples_per_second": 125.383,
|
| 10 |
+
"eval_steps_per_second": 15.695,
|
| 11 |
+
"total_flos": 9.140124804986585e+18,
|
| 12 |
+
"train_loss": 0.6221655089083716,
|
| 13 |
+
"train_runtime": 2787.435,
|
| 14 |
+
"train_samples_per_second": 183.968,
|
| 15 |
+
"train_steps_per_second": 11.516
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 23.0,
|
| 3 |
+
"eval_accuracy": 0.8349514563106796,
|
| 4 |
+
"eval_f1": 0.8349682338681513,
|
| 5 |
+
"eval_loss": 0.534845769405365,
|
| 6 |
+
"eval_precision": 0.8390873801645569,
|
| 7 |
+
"eval_recall": 0.8349514563106796,
|
| 8 |
+
"eval_runtime": 23.0016,
|
| 9 |
+
"eval_samples_per_second": 125.383,
|
| 10 |
+
"eval_steps_per_second": 15.695
|
| 11 |
+
}
|
runs/Mar22_11-30-39_80a90583e6d6/events.out.tfevents.1711109882.80a90583e6d6.2864.11
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:516bf625702e14de063670769f0cb3651d9ce0828e739b18b49d95afec6505b9
|
| 3 |
+
size 560
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 23.0,
|
| 3 |
+
"total_flos": 9.140124804986585e+18,
|
| 4 |
+
"train_loss": 0.6221655089083716,
|
| 5 |
+
"train_runtime": 2787.435,
|
| 6 |
+
"train_samples_per_second": 183.968,
|
| 7 |
+
"train_steps_per_second": 11.516
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,467 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.534845769405365,
|
| 3 |
+
"best_model_checkpoint": "./vit-epsilon-1e-7/checkpoint-4173",
|
| 4 |
+
"epoch": 23.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 7383,
|
| 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": 26.270666122436523,
|
| 14 |
+
"learning_rate": 1.8349682631275247e-05,
|
| 15 |
+
"loss": 1.765,
|
| 16 |
+
"step": 321
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.691747572815534,
|
| 21 |
+
"eval_f1": 0.6531054572377663,
|
| 22 |
+
"eval_loss": 0.9570140838623047,
|
| 23 |
+
"eval_precision": 0.6486719468870186,
|
| 24 |
+
"eval_recall": 0.691747572815534,
|
| 25 |
+
"eval_runtime": 22.6395,
|
| 26 |
+
"eval_samples_per_second": 127.388,
|
| 27 |
+
"eval_steps_per_second": 15.946,
|
| 28 |
+
"step": 321
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"grad_norm": 23.715213775634766,
|
| 33 |
+
"learning_rate": 3.6872475476053084e-05,
|
| 34 |
+
"loss": 1.1815,
|
| 35 |
+
"step": 642
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.0,
|
| 39 |
+
"eval_accuracy": 0.7558945908460472,
|
| 40 |
+
"eval_f1": 0.7314126926737505,
|
| 41 |
+
"eval_loss": 0.6948704123497009,
|
| 42 |
+
"eval_precision": 0.7314749711745714,
|
| 43 |
+
"eval_recall": 0.7558945908460472,
|
| 44 |
+
"eval_runtime": 22.8506,
|
| 45 |
+
"eval_samples_per_second": 126.211,
|
| 46 |
+
"eval_steps_per_second": 15.798,
|
| 47 |
+
"step": 642
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"grad_norm": 5.459173202514648,
|
| 52 |
+
"learning_rate": 5.539526832083093e-05,
|
| 53 |
+
"loss": 1.0605,
|
| 54 |
+
"step": 963
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_accuracy": 0.771497919556172,
|
| 59 |
+
"eval_f1": 0.752980330371167,
|
| 60 |
+
"eval_loss": 0.6212790012359619,
|
| 61 |
+
"eval_precision": 0.7648636119210427,
|
| 62 |
+
"eval_recall": 0.771497919556172,
|
| 63 |
+
"eval_runtime": 22.9013,
|
| 64 |
+
"eval_samples_per_second": 125.932,
|
| 65 |
+
"eval_steps_per_second": 15.763,
|
| 66 |
+
"step": 963
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.0,
|
| 70 |
+
"grad_norm": 23.622180938720703,
|
| 71 |
+
"learning_rate": 7.391806116560877e-05,
|
| 72 |
+
"loss": 1.0147,
|
| 73 |
+
"step": 1284
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.0,
|
| 77 |
+
"eval_accuracy": 0.7732316227461858,
|
| 78 |
+
"eval_f1": 0.773441625340043,
|
| 79 |
+
"eval_loss": 0.5767605304718018,
|
| 80 |
+
"eval_precision": 0.7806580682757648,
|
| 81 |
+
"eval_recall": 0.7732316227461858,
|
| 82 |
+
"eval_runtime": 22.9855,
|
| 83 |
+
"eval_samples_per_second": 125.47,
|
| 84 |
+
"eval_steps_per_second": 15.706,
|
| 85 |
+
"step": 1284
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.0,
|
| 89 |
+
"grad_norm": 9.584253311157227,
|
| 90 |
+
"learning_rate": 9.244085401038661e-05,
|
| 91 |
+
"loss": 0.93,
|
| 92 |
+
"step": 1605
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.0,
|
| 96 |
+
"eval_accuracy": 0.7586685159500693,
|
| 97 |
+
"eval_f1": 0.7661547242155936,
|
| 98 |
+
"eval_loss": 0.657228946685791,
|
| 99 |
+
"eval_precision": 0.7939813909551887,
|
| 100 |
+
"eval_recall": 0.7586685159500693,
|
| 101 |
+
"eval_runtime": 22.8929,
|
| 102 |
+
"eval_samples_per_second": 125.978,
|
| 103 |
+
"eval_steps_per_second": 15.769,
|
| 104 |
+
"step": 1605
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.0,
|
| 108 |
+
"grad_norm": 13.967341423034668,
|
| 109 |
+
"learning_rate": 9.984553168214161e-05,
|
| 110 |
+
"loss": 0.9793,
|
| 111 |
+
"step": 1926
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.0,
|
| 115 |
+
"eval_accuracy": 0.7701109570041609,
|
| 116 |
+
"eval_f1": 0.7742181002637638,
|
| 117 |
+
"eval_loss": 0.61651611328125,
|
| 118 |
+
"eval_precision": 0.7940357585259091,
|
| 119 |
+
"eval_recall": 0.7701109570041609,
|
| 120 |
+
"eval_runtime": 22.9308,
|
| 121 |
+
"eval_samples_per_second": 125.77,
|
| 122 |
+
"eval_steps_per_second": 15.743,
|
| 123 |
+
"step": 1926
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 7.0,
|
| 127 |
+
"grad_norm": 12.87781810760498,
|
| 128 |
+
"learning_rate": 9.889061131437471e-05,
|
| 129 |
+
"loss": 0.8662,
|
| 130 |
+
"step": 2247
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 7.0,
|
| 134 |
+
"eval_accuracy": 0.723994452149792,
|
| 135 |
+
"eval_f1": 0.745594196112459,
|
| 136 |
+
"eval_loss": 0.6534591913223267,
|
| 137 |
+
"eval_precision": 0.8097504651538117,
|
| 138 |
+
"eval_recall": 0.723994452149792,
|
| 139 |
+
"eval_runtime": 24.6973,
|
| 140 |
+
"eval_samples_per_second": 116.774,
|
| 141 |
+
"eval_steps_per_second": 14.617,
|
| 142 |
+
"step": 2247
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.0,
|
| 146 |
+
"grad_norm": 3.272010326385498,
|
| 147 |
+
"learning_rate": 9.707265436104638e-05,
|
| 148 |
+
"loss": 0.7767,
|
| 149 |
+
"step": 2568
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 8.0,
|
| 153 |
+
"eval_accuracy": 0.7565880721220527,
|
| 154 |
+
"eval_f1": 0.773329540133459,
|
| 155 |
+
"eval_loss": 0.5812659859657288,
|
| 156 |
+
"eval_precision": 0.8124371295597024,
|
| 157 |
+
"eval_recall": 0.7565880721220527,
|
| 158 |
+
"eval_runtime": 24.6374,
|
| 159 |
+
"eval_samples_per_second": 117.058,
|
| 160 |
+
"eval_steps_per_second": 14.653,
|
| 161 |
+
"step": 2568
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.0,
|
| 165 |
+
"grad_norm": 12.090767860412598,
|
| 166 |
+
"learning_rate": 9.44253127296151e-05,
|
| 167 |
+
"loss": 0.7572,
|
| 168 |
+
"step": 2889
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 9.0,
|
| 172 |
+
"eval_accuracy": 0.8144937586685159,
|
| 173 |
+
"eval_f1": 0.811409191547864,
|
| 174 |
+
"eval_loss": 0.5385076403617859,
|
| 175 |
+
"eval_precision": 0.8131138362731425,
|
| 176 |
+
"eval_recall": 0.8144937586685159,
|
| 177 |
+
"eval_runtime": 23.005,
|
| 178 |
+
"eval_samples_per_second": 125.364,
|
| 179 |
+
"eval_steps_per_second": 15.692,
|
| 180 |
+
"step": 2889
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 10.0,
|
| 184 |
+
"grad_norm": 9.961050987243652,
|
| 185 |
+
"learning_rate": 9.099523058358976e-05,
|
| 186 |
+
"loss": 0.7003,
|
| 187 |
+
"step": 3210
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 10.0,
|
| 191 |
+
"eval_accuracy": 0.8027045769764216,
|
| 192 |
+
"eval_f1": 0.809293314931955,
|
| 193 |
+
"eval_loss": 0.5354512929916382,
|
| 194 |
+
"eval_precision": 0.8276170557646749,
|
| 195 |
+
"eval_recall": 0.8027045769764216,
|
| 196 |
+
"eval_runtime": 23.6688,
|
| 197 |
+
"eval_samples_per_second": 121.848,
|
| 198 |
+
"eval_steps_per_second": 15.252,
|
| 199 |
+
"step": 3210
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 11.0,
|
| 203 |
+
"grad_norm": 25.105077743530273,
|
| 204 |
+
"learning_rate": 8.684284338417735e-05,
|
| 205 |
+
"loss": 0.6316,
|
| 206 |
+
"step": 3531
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 11.0,
|
| 210 |
+
"eval_accuracy": 0.765256588072122,
|
| 211 |
+
"eval_f1": 0.7816439176174772,
|
| 212 |
+
"eval_loss": 0.6284907460212708,
|
| 213 |
+
"eval_precision": 0.8322254365589742,
|
| 214 |
+
"eval_recall": 0.765256588072122,
|
| 215 |
+
"eval_runtime": 23.5967,
|
| 216 |
+
"eval_samples_per_second": 122.22,
|
| 217 |
+
"eval_steps_per_second": 15.299,
|
| 218 |
+
"step": 3531
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 12.0,
|
| 222 |
+
"grad_norm": 10.844632148742676,
|
| 223 |
+
"learning_rate": 8.204131306302357e-05,
|
| 224 |
+
"loss": 0.5723,
|
| 225 |
+
"step": 3852
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 12.0,
|
| 229 |
+
"eval_accuracy": 0.8016643550624133,
|
| 230 |
+
"eval_f1": 0.8104537685542633,
|
| 231 |
+
"eval_loss": 0.5774895548820496,
|
| 232 |
+
"eval_precision": 0.8279128953158427,
|
| 233 |
+
"eval_recall": 0.8016643550624133,
|
| 234 |
+
"eval_runtime": 23.4985,
|
| 235 |
+
"eval_samples_per_second": 122.731,
|
| 236 |
+
"eval_steps_per_second": 15.363,
|
| 237 |
+
"step": 3852
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 13.0,
|
| 241 |
+
"grad_norm": 11.590338706970215,
|
| 242 |
+
"learning_rate": 7.667523896413962e-05,
|
| 243 |
+
"loss": 0.4717,
|
| 244 |
+
"step": 4173
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 13.0,
|
| 248 |
+
"eval_accuracy": 0.8349514563106796,
|
| 249 |
+
"eval_f1": 0.8349682338681513,
|
| 250 |
+
"eval_loss": 0.534845769405365,
|
| 251 |
+
"eval_precision": 0.8390873801645569,
|
| 252 |
+
"eval_recall": 0.8349514563106796,
|
| 253 |
+
"eval_runtime": 23.9213,
|
| 254 |
+
"eval_samples_per_second": 120.562,
|
| 255 |
+
"eval_steps_per_second": 15.091,
|
| 256 |
+
"step": 4173
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 14.0,
|
| 260 |
+
"grad_norm": 3.6801199913024902,
|
| 261 |
+
"learning_rate": 7.083916726724684e-05,
|
| 262 |
+
"loss": 0.4472,
|
| 263 |
+
"step": 4494
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 14.0,
|
| 267 |
+
"eval_accuracy": 0.8238557558945908,
|
| 268 |
+
"eval_f1": 0.8298970188071358,
|
| 269 |
+
"eval_loss": 0.5469394326210022,
|
| 270 |
+
"eval_precision": 0.844241279806222,
|
| 271 |
+
"eval_recall": 0.8238557558945908,
|
| 272 |
+
"eval_runtime": 23.6076,
|
| 273 |
+
"eval_samples_per_second": 122.164,
|
| 274 |
+
"eval_steps_per_second": 15.292,
|
| 275 |
+
"step": 4494
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 15.0,
|
| 279 |
+
"grad_norm": 2.446948528289795,
|
| 280 |
+
"learning_rate": 6.46557084486047e-05,
|
| 281 |
+
"loss": 0.3764,
|
| 282 |
+
"step": 4815
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 15.0,
|
| 286 |
+
"eval_accuracy": 0.8290568654646324,
|
| 287 |
+
"eval_f1": 0.8355171455268711,
|
| 288 |
+
"eval_loss": 0.5737930536270142,
|
| 289 |
+
"eval_precision": 0.8501341497772392,
|
| 290 |
+
"eval_recall": 0.8290568654646324,
|
| 291 |
+
"eval_runtime": 23.7441,
|
| 292 |
+
"eval_samples_per_second": 121.462,
|
| 293 |
+
"eval_steps_per_second": 15.204,
|
| 294 |
+
"step": 4815
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 16.0,
|
| 298 |
+
"grad_norm": 6.043246269226074,
|
| 299 |
+
"learning_rate": 5.8195220793532045e-05,
|
| 300 |
+
"loss": 0.3346,
|
| 301 |
+
"step": 5136
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 16.0,
|
| 305 |
+
"eval_accuracy": 0.8436199722607489,
|
| 306 |
+
"eval_f1": 0.8461452975648701,
|
| 307 |
+
"eval_loss": 0.5367853045463562,
|
| 308 |
+
"eval_precision": 0.851244402940535,
|
| 309 |
+
"eval_recall": 0.8436199722607489,
|
| 310 |
+
"eval_runtime": 23.9285,
|
| 311 |
+
"eval_samples_per_second": 120.526,
|
| 312 |
+
"eval_steps_per_second": 15.087,
|
| 313 |
+
"step": 5136
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 17.0,
|
| 317 |
+
"grad_norm": 15.190231323242188,
|
| 318 |
+
"learning_rate": 5.15903395270923e-05,
|
| 319 |
+
"loss": 0.2826,
|
| 320 |
+
"step": 5457
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 17.0,
|
| 324 |
+
"eval_accuracy": 0.8474341192787794,
|
| 325 |
+
"eval_f1": 0.8467916215502757,
|
| 326 |
+
"eval_loss": 0.5510370135307312,
|
| 327 |
+
"eval_precision": 0.8489075342171026,
|
| 328 |
+
"eval_recall": 0.8474341192787794,
|
| 329 |
+
"eval_runtime": 23.9505,
|
| 330 |
+
"eval_samples_per_second": 120.415,
|
| 331 |
+
"eval_steps_per_second": 15.073,
|
| 332 |
+
"step": 5457
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 18.0,
|
| 336 |
+
"grad_norm": 24.944795608520508,
|
| 337 |
+
"learning_rate": 4.4978023449711664e-05,
|
| 338 |
+
"loss": 0.2659,
|
| 339 |
+
"step": 5778
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 18.0,
|
| 343 |
+
"eval_accuracy": 0.8547156726768377,
|
| 344 |
+
"eval_f1": 0.8549199048922214,
|
| 345 |
+
"eval_loss": 0.546658456325531,
|
| 346 |
+
"eval_precision": 0.8559700691467993,
|
| 347 |
+
"eval_recall": 0.8547156726768377,
|
| 348 |
+
"eval_runtime": 23.4765,
|
| 349 |
+
"eval_samples_per_second": 122.846,
|
| 350 |
+
"eval_steps_per_second": 15.377,
|
| 351 |
+
"step": 5778
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 19.0,
|
| 355 |
+
"grad_norm": 0.11274675279855728,
|
| 356 |
+
"learning_rate": 3.843351058509752e-05,
|
| 357 |
+
"loss": 0.2545,
|
| 358 |
+
"step": 6099
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 19.0,
|
| 362 |
+
"eval_accuracy": 0.8432732316227461,
|
| 363 |
+
"eval_f1": 0.8487015901360936,
|
| 364 |
+
"eval_loss": 0.6155611872673035,
|
| 365 |
+
"eval_precision": 0.8617407571979913,
|
| 366 |
+
"eval_recall": 0.8432732316227461,
|
| 367 |
+
"eval_runtime": 22.4381,
|
| 368 |
+
"eval_samples_per_second": 128.532,
|
| 369 |
+
"eval_steps_per_second": 16.089,
|
| 370 |
+
"step": 6099
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 20.0,
|
| 374 |
+
"grad_norm": 0.5206512808799744,
|
| 375 |
+
"learning_rate": 3.209279054309716e-05,
|
| 376 |
+
"loss": 0.2123,
|
| 377 |
+
"step": 6420
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 20.0,
|
| 381 |
+
"eval_accuracy": 0.8429264909847434,
|
| 382 |
+
"eval_f1": 0.8426622560864744,
|
| 383 |
+
"eval_loss": 0.6871052384376526,
|
| 384 |
+
"eval_precision": 0.8498556529498513,
|
| 385 |
+
"eval_recall": 0.8429264909847434,
|
| 386 |
+
"eval_runtime": 22.4536,
|
| 387 |
+
"eval_samples_per_second": 128.443,
|
| 388 |
+
"eval_steps_per_second": 16.078,
|
| 389 |
+
"step": 6420
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 21.0,
|
| 393 |
+
"grad_norm": 6.716039657592773,
|
| 394 |
+
"learning_rate": 2.6067582030088143e-05,
|
| 395 |
+
"loss": 0.1655,
|
| 396 |
+
"step": 6741
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 21.0,
|
| 400 |
+
"eval_accuracy": 0.8609570041608876,
|
| 401 |
+
"eval_f1": 0.856691150022433,
|
| 402 |
+
"eval_loss": 0.6139292120933533,
|
| 403 |
+
"eval_precision": 0.8551833143301755,
|
| 404 |
+
"eval_recall": 0.8609570041608876,
|
| 405 |
+
"eval_runtime": 22.975,
|
| 406 |
+
"eval_samples_per_second": 125.528,
|
| 407 |
+
"eval_steps_per_second": 15.713,
|
| 408 |
+
"step": 6741
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 22.0,
|
| 412 |
+
"grad_norm": 108.25263977050781,
|
| 413 |
+
"learning_rate": 2.0464044677195966e-05,
|
| 414 |
+
"loss": 0.1246,
|
| 415 |
+
"step": 7062
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 22.0,
|
| 419 |
+
"eval_accuracy": 0.8675450762829403,
|
| 420 |
+
"eval_f1": 0.8676542597333672,
|
| 421 |
+
"eval_loss": 0.6129117012023926,
|
| 422 |
+
"eval_precision": 0.8680620732919173,
|
| 423 |
+
"eval_recall": 0.8675450762829403,
|
| 424 |
+
"eval_runtime": 22.6536,
|
| 425 |
+
"eval_samples_per_second": 127.309,
|
| 426 |
+
"eval_steps_per_second": 15.936,
|
| 427 |
+
"step": 7062
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 23.0,
|
| 431 |
+
"grad_norm": 2.2161059379577637,
|
| 432 |
+
"learning_rate": 1.538090858763548e-05,
|
| 433 |
+
"loss": 0.1394,
|
| 434 |
+
"step": 7383
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 23.0,
|
| 438 |
+
"eval_accuracy": 0.8713592233009708,
|
| 439 |
+
"eval_f1": 0.867730828379127,
|
| 440 |
+
"eval_loss": 0.6522772908210754,
|
| 441 |
+
"eval_precision": 0.8674975891372388,
|
| 442 |
+
"eval_recall": 0.8713592233009708,
|
| 443 |
+
"eval_runtime": 22.7054,
|
| 444 |
+
"eval_samples_per_second": 127.018,
|
| 445 |
+
"eval_steps_per_second": 15.899,
|
| 446 |
+
"step": 7383
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 23.0,
|
| 450 |
+
"step": 7383,
|
| 451 |
+
"total_flos": 9.140124804986585e+18,
|
| 452 |
+
"train_loss": 0.6221655089083716,
|
| 453 |
+
"train_runtime": 2787.435,
|
| 454 |
+
"train_samples_per_second": 183.968,
|
| 455 |
+
"train_steps_per_second": 11.516
|
| 456 |
+
}
|
| 457 |
+
],
|
| 458 |
+
"logging_steps": 500,
|
| 459 |
+
"max_steps": 32100,
|
| 460 |
+
"num_input_tokens_seen": 0,
|
| 461 |
+
"num_train_epochs": 100,
|
| 462 |
+
"save_steps": 500,
|
| 463 |
+
"total_flos": 9.140124804986585e+18,
|
| 464 |
+
"train_batch_size": 16,
|
| 465 |
+
"trial_name": null,
|
| 466 |
+
"trial_params": null
|
| 467 |
+
}
|