🍻 cheers
Browse files- README.md +7 -6
- all_results.json +16 -0
- eval_results.json +11 -0
- runs/Mar21_13-11-52_45242e0ea5c0/events.out.tfevents.1711030007.45242e0ea5c0.20273.19 +3 -0
- train_results.json +8 -0
- trainer_state.json +543 -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-weight-decay-1e-2
|
| 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-weight-decay-1e-2
|
| 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: 0.4994
|
| 25 |
+
- Accuracy: 0.8603
|
| 26 |
+
- Precision: 0.8618
|
| 27 |
+
- Recall: 0.8603
|
| 28 |
+
- F1: 0.8600
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"eval_accuracy": 0.8602635228848821,
|
| 4 |
+
"eval_f1": 0.8599959330449201,
|
| 5 |
+
"eval_loss": 0.4994313716888428,
|
| 6 |
+
"eval_precision": 0.8617536928422816,
|
| 7 |
+
"eval_recall": 0.8602635228848821,
|
| 8 |
+
"eval_runtime": 25.4942,
|
| 9 |
+
"eval_samples_per_second": 113.124,
|
| 10 |
+
"eval_steps_per_second": 14.16,
|
| 11 |
+
"total_flos": 1.0729711727592948e+19,
|
| 12 |
+
"train_loss": 0.5187536703752126,
|
| 13 |
+
"train_runtime": 3224.0652,
|
| 14 |
+
"train_samples_per_second": 159.054,
|
| 15 |
+
"train_steps_per_second": 9.956
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"eval_accuracy": 0.8602635228848821,
|
| 4 |
+
"eval_f1": 0.8599959330449201,
|
| 5 |
+
"eval_loss": 0.4994313716888428,
|
| 6 |
+
"eval_precision": 0.8617536928422816,
|
| 7 |
+
"eval_recall": 0.8602635228848821,
|
| 8 |
+
"eval_runtime": 25.4942,
|
| 9 |
+
"eval_samples_per_second": 113.124,
|
| 10 |
+
"eval_steps_per_second": 14.16
|
| 11 |
+
}
|
runs/Mar21_13-11-52_45242e0ea5c0/events.out.tfevents.1711030007.45242e0ea5c0.20273.19
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e8b6eca79b4fd0e80e56f42320fe4313898a5c8aaa25a580dd54df95e1bf592
|
| 3 |
+
size 560
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"total_flos": 1.0729711727592948e+19,
|
| 4 |
+
"train_loss": 0.5187536703752126,
|
| 5 |
+
"train_runtime": 3224.0652,
|
| 6 |
+
"train_samples_per_second": 159.054,
|
| 7 |
+
"train_steps_per_second": 9.956
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,543 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4994313716888428,
|
| 3 |
+
"best_model_checkpoint": "./vit-weight-decay-1e-2/checkpoint-5457",
|
| 4 |
+
"epoch": 27.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8667,
|
| 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": 27.05959129333496,
|
| 14 |
+
"learning_rate": 2.6004922067268256e-05,
|
| 15 |
+
"loss": 1.7124,
|
| 16 |
+
"step": 321
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.6924410540915396,
|
| 21 |
+
"eval_f1": 0.603022144115533,
|
| 22 |
+
"eval_loss": 0.8697461485862732,
|
| 23 |
+
"eval_precision": 0.6656411815509768,
|
| 24 |
+
"eval_recall": 0.6924410540915396,
|
| 25 |
+
"eval_runtime": 24.2005,
|
| 26 |
+
"eval_samples_per_second": 119.171,
|
| 27 |
+
"eval_steps_per_second": 14.917,
|
| 28 |
+
"step": 321
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"grad_norm": 31.123794555664062,
|
| 33 |
+
"learning_rate": 5.233798195242002e-05,
|
| 34 |
+
"loss": 1.1476,
|
| 35 |
+
"step": 642
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.0,
|
| 39 |
+
"eval_accuracy": 0.6990291262135923,
|
| 40 |
+
"eval_f1": 0.714853056600681,
|
| 41 |
+
"eval_loss": 0.7271208763122559,
|
| 42 |
+
"eval_precision": 0.7684437503161472,
|
| 43 |
+
"eval_recall": 0.6990291262135923,
|
| 44 |
+
"eval_runtime": 23.1919,
|
| 45 |
+
"eval_samples_per_second": 124.354,
|
| 46 |
+
"eval_steps_per_second": 15.566,
|
| 47 |
+
"step": 642
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"grad_norm": 4.652539253234863,
|
| 52 |
+
"learning_rate": 7.867104183757178e-05,
|
| 53 |
+
"loss": 1.0734,
|
| 54 |
+
"step": 963
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_accuracy": 0.7687239944521498,
|
| 59 |
+
"eval_f1": 0.7417221561170182,
|
| 60 |
+
"eval_loss": 0.6440889239311218,
|
| 61 |
+
"eval_precision": 0.7568176957187778,
|
| 62 |
+
"eval_recall": 0.7687239944521498,
|
| 63 |
+
"eval_runtime": 23.9098,
|
| 64 |
+
"eval_samples_per_second": 120.62,
|
| 65 |
+
"eval_steps_per_second": 15.098,
|
| 66 |
+
"step": 963
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.0,
|
| 70 |
+
"grad_norm": 26.467485427856445,
|
| 71 |
+
"learning_rate": 9.99845966779335e-05,
|
| 72 |
+
"loss": 1.0271,
|
| 73 |
+
"step": 1284
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.0,
|
| 77 |
+
"eval_accuracy": 0.7773925104022191,
|
| 78 |
+
"eval_f1": 0.7814315937772335,
|
| 79 |
+
"eval_loss": 0.5854852199554443,
|
| 80 |
+
"eval_precision": 0.78834249113171,
|
| 81 |
+
"eval_recall": 0.7773925104022191,
|
| 82 |
+
"eval_runtime": 23.2363,
|
| 83 |
+
"eval_samples_per_second": 124.116,
|
| 84 |
+
"eval_steps_per_second": 15.536,
|
| 85 |
+
"step": 1284
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.0,
|
| 89 |
+
"grad_norm": 9.265125274658203,
|
| 90 |
+
"learning_rate": 9.93971225198763e-05,
|
| 91 |
+
"loss": 0.9158,
|
| 92 |
+
"step": 1605
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.0,
|
| 96 |
+
"eval_accuracy": 0.7635228848821082,
|
| 97 |
+
"eval_f1": 0.7661961115296833,
|
| 98 |
+
"eval_loss": 0.700226366519928,
|
| 99 |
+
"eval_precision": 0.7929662231425049,
|
| 100 |
+
"eval_recall": 0.7635228848821082,
|
| 101 |
+
"eval_runtime": 23.9812,
|
| 102 |
+
"eval_samples_per_second": 120.261,
|
| 103 |
+
"eval_steps_per_second": 15.053,
|
| 104 |
+
"step": 1605
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.0,
|
| 108 |
+
"grad_norm": 7.993426322937012,
|
| 109 |
+
"learning_rate": 9.796799913911281e-05,
|
| 110 |
+
"loss": 0.9167,
|
| 111 |
+
"step": 1926
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.0,
|
| 115 |
+
"eval_accuracy": 0.7812066574202496,
|
| 116 |
+
"eval_f1": 0.7900429974573581,
|
| 117 |
+
"eval_loss": 0.5867117047309875,
|
| 118 |
+
"eval_precision": 0.806466933291998,
|
| 119 |
+
"eval_recall": 0.7812066574202496,
|
| 120 |
+
"eval_runtime": 23.5885,
|
| 121 |
+
"eval_samples_per_second": 122.263,
|
| 122 |
+
"eval_steps_per_second": 15.304,
|
| 123 |
+
"step": 1926
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 7.0,
|
| 127 |
+
"grad_norm": 8.043280601501465,
|
| 128 |
+
"learning_rate": 9.572157654878572e-05,
|
| 129 |
+
"loss": 0.786,
|
| 130 |
+
"step": 2247
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 7.0,
|
| 134 |
+
"eval_accuracy": 0.7340499306518724,
|
| 135 |
+
"eval_f1": 0.751486114744647,
|
| 136 |
+
"eval_loss": 0.6516677737236023,
|
| 137 |
+
"eval_precision": 0.8047120902571854,
|
| 138 |
+
"eval_recall": 0.7340499306518724,
|
| 139 |
+
"eval_runtime": 24.7074,
|
| 140 |
+
"eval_samples_per_second": 116.726,
|
| 141 |
+
"eval_steps_per_second": 14.611,
|
| 142 |
+
"step": 2247
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.0,
|
| 146 |
+
"grad_norm": 3.0447545051574707,
|
| 147 |
+
"learning_rate": 9.26961302542397e-05,
|
| 148 |
+
"loss": 0.7406,
|
| 149 |
+
"step": 2568
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 8.0,
|
| 153 |
+
"eval_accuracy": 0.7066574202496533,
|
| 154 |
+
"eval_f1": 0.7330088250984578,
|
| 155 |
+
"eval_loss": 0.6647158265113831,
|
| 156 |
+
"eval_precision": 0.8133714017605812,
|
| 157 |
+
"eval_recall": 0.7066574202496533,
|
| 158 |
+
"eval_runtime": 24.3728,
|
| 159 |
+
"eval_samples_per_second": 118.329,
|
| 160 |
+
"eval_steps_per_second": 14.812,
|
| 161 |
+
"step": 2568
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.0,
|
| 165 |
+
"grad_norm": 3.4238717555999756,
|
| 166 |
+
"learning_rate": 8.89432090986511e-05,
|
| 167 |
+
"loss": 0.682,
|
| 168 |
+
"step": 2889
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 9.0,
|
| 172 |
+
"eval_accuracy": 0.8228155339805825,
|
| 173 |
+
"eval_f1": 0.8207044848679155,
|
| 174 |
+
"eval_loss": 0.510608434677124,
|
| 175 |
+
"eval_precision": 0.8230905088203688,
|
| 176 |
+
"eval_recall": 0.8228155339805825,
|
| 177 |
+
"eval_runtime": 24.4807,
|
| 178 |
+
"eval_samples_per_second": 117.807,
|
| 179 |
+
"eval_steps_per_second": 14.746,
|
| 180 |
+
"step": 2889
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 10.0,
|
| 184 |
+
"grad_norm": 11.989164352416992,
|
| 185 |
+
"learning_rate": 8.45267569518721e-05,
|
| 186 |
+
"loss": 0.6427,
|
| 187 |
+
"step": 3210
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 10.0,
|
| 191 |
+
"eval_accuracy": 0.8165742024965326,
|
| 192 |
+
"eval_f1": 0.8221884365803362,
|
| 193 |
+
"eval_loss": 0.5032415390014648,
|
| 194 |
+
"eval_precision": 0.835376334433007,
|
| 195 |
+
"eval_recall": 0.8165742024965326,
|
| 196 |
+
"eval_runtime": 23.2354,
|
| 197 |
+
"eval_samples_per_second": 124.121,
|
| 198 |
+
"eval_steps_per_second": 15.537,
|
| 199 |
+
"step": 3210
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 11.0,
|
| 203 |
+
"grad_norm": 34.21263885498047,
|
| 204 |
+
"learning_rate": 7.952202320752798e-05,
|
| 205 |
+
"loss": 0.5663,
|
| 206 |
+
"step": 3531
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 11.0,
|
| 210 |
+
"eval_accuracy": 0.8151872399445215,
|
| 211 |
+
"eval_f1": 0.8216396746544562,
|
| 212 |
+
"eval_loss": 0.5357819199562073,
|
| 213 |
+
"eval_precision": 0.8325502393616235,
|
| 214 |
+
"eval_recall": 0.8151872399445215,
|
| 215 |
+
"eval_runtime": 24.4554,
|
| 216 |
+
"eval_samples_per_second": 117.929,
|
| 217 |
+
"eval_steps_per_second": 14.762,
|
| 218 |
+
"step": 3531
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 12.0,
|
| 222 |
+
"grad_norm": 44.03895568847656,
|
| 223 |
+
"learning_rate": 7.401428065178325e-05,
|
| 224 |
+
"loss": 0.5395,
|
| 225 |
+
"step": 3852
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 12.0,
|
| 229 |
+
"eval_accuracy": 0.8248959778085991,
|
| 230 |
+
"eval_f1": 0.8298561156988427,
|
| 231 |
+
"eval_loss": 0.5487632155418396,
|
| 232 |
+
"eval_precision": 0.8391637957530821,
|
| 233 |
+
"eval_recall": 0.8248959778085991,
|
| 234 |
+
"eval_runtime": 23.3681,
|
| 235 |
+
"eval_samples_per_second": 123.416,
|
| 236 |
+
"eval_steps_per_second": 15.448,
|
| 237 |
+
"step": 3852
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 13.0,
|
| 241 |
+
"grad_norm": 27.093250274658203,
|
| 242 |
+
"learning_rate": 6.80973725492743e-05,
|
| 243 |
+
"loss": 0.4468,
|
| 244 |
+
"step": 4173
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 13.0,
|
| 248 |
+
"eval_accuracy": 0.8231622746185853,
|
| 249 |
+
"eval_f1": 0.8260339295474823,
|
| 250 |
+
"eval_loss": 0.578988254070282,
|
| 251 |
+
"eval_precision": 0.8397229815779756,
|
| 252 |
+
"eval_recall": 0.8231622746185853,
|
| 253 |
+
"eval_runtime": 24.8132,
|
| 254 |
+
"eval_samples_per_second": 116.229,
|
| 255 |
+
"eval_steps_per_second": 14.549,
|
| 256 |
+
"step": 4173
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 14.0,
|
| 260 |
+
"grad_norm": 5.494964122772217,
|
| 261 |
+
"learning_rate": 6.187211370157784e-05,
|
| 262 |
+
"loss": 0.4247,
|
| 263 |
+
"step": 4494
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 14.0,
|
| 267 |
+
"eval_accuracy": 0.8415395284327323,
|
| 268 |
+
"eval_f1": 0.8448790763195109,
|
| 269 |
+
"eval_loss": 0.5437958240509033,
|
| 270 |
+
"eval_precision": 0.8570099067237934,
|
| 271 |
+
"eval_recall": 0.8415395284327323,
|
| 272 |
+
"eval_runtime": 23.4933,
|
| 273 |
+
"eval_samples_per_second": 122.758,
|
| 274 |
+
"eval_steps_per_second": 15.366,
|
| 275 |
+
"step": 4494
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 15.0,
|
| 279 |
+
"grad_norm": 5.022720813751221,
|
| 280 |
+
"learning_rate": 5.544457272166217e-05,
|
| 281 |
+
"loss": 0.3495,
|
| 282 |
+
"step": 4815
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 15.0,
|
| 286 |
+
"eval_accuracy": 0.8453536754507628,
|
| 287 |
+
"eval_f1": 0.846725089255697,
|
| 288 |
+
"eval_loss": 0.5135474801063538,
|
| 289 |
+
"eval_precision": 0.8518606648557052,
|
| 290 |
+
"eval_recall": 0.8453536754507628,
|
| 291 |
+
"eval_runtime": 22.9349,
|
| 292 |
+
"eval_samples_per_second": 125.747,
|
| 293 |
+
"eval_steps_per_second": 15.74,
|
| 294 |
+
"step": 4815
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 16.0,
|
| 298 |
+
"grad_norm": 21.08028221130371,
|
| 299 |
+
"learning_rate": 4.894460661440583e-05,
|
| 300 |
+
"loss": 0.3039,
|
| 301 |
+
"step": 5136
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 16.0,
|
| 305 |
+
"eval_accuracy": 0.8408460471567267,
|
| 306 |
+
"eval_f1": 0.8447631505343497,
|
| 307 |
+
"eval_loss": 0.5631198287010193,
|
| 308 |
+
"eval_precision": 0.8520329480153485,
|
| 309 |
+
"eval_recall": 0.8408460471567267,
|
| 310 |
+
"eval_runtime": 22.9052,
|
| 311 |
+
"eval_samples_per_second": 125.91,
|
| 312 |
+
"eval_steps_per_second": 15.761,
|
| 313 |
+
"step": 5136
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 17.0,
|
| 317 |
+
"grad_norm": 0.3130456805229187,
|
| 318 |
+
"learning_rate": 4.244239774409037e-05,
|
| 319 |
+
"loss": 0.2602,
|
| 320 |
+
"step": 5457
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 17.0,
|
| 324 |
+
"eval_accuracy": 0.8602635228848821,
|
| 325 |
+
"eval_f1": 0.8599959330449201,
|
| 326 |
+
"eval_loss": 0.4994313716888428,
|
| 327 |
+
"eval_precision": 0.8617536928422816,
|
| 328 |
+
"eval_recall": 0.8602635228848821,
|
| 329 |
+
"eval_runtime": 24.0474,
|
| 330 |
+
"eval_samples_per_second": 119.93,
|
| 331 |
+
"eval_steps_per_second": 15.012,
|
| 332 |
+
"step": 5457
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 18.0,
|
| 336 |
+
"grad_norm": 13.50483226776123,
|
| 337 |
+
"learning_rate": 3.606895852147351e-05,
|
| 338 |
+
"loss": 0.2616,
|
| 339 |
+
"step": 5778
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 18.0,
|
| 343 |
+
"eval_accuracy": 0.8564493758668515,
|
| 344 |
+
"eval_f1": 0.8585397395861224,
|
| 345 |
+
"eval_loss": 0.5405685901641846,
|
| 346 |
+
"eval_precision": 0.8621724878350435,
|
| 347 |
+
"eval_recall": 0.8564493758668515,
|
| 348 |
+
"eval_runtime": 23.9163,
|
| 349 |
+
"eval_samples_per_second": 120.587,
|
| 350 |
+
"eval_steps_per_second": 15.094,
|
| 351 |
+
"step": 5778
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 19.0,
|
| 355 |
+
"grad_norm": 0.11348175257444382,
|
| 356 |
+
"learning_rate": 2.9932882319894417e-05,
|
| 357 |
+
"loss": 0.1876,
|
| 358 |
+
"step": 6099
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 19.0,
|
| 362 |
+
"eval_accuracy": 0.848127600554785,
|
| 363 |
+
"eval_f1": 0.8525273952669322,
|
| 364 |
+
"eval_loss": 0.5612274408340454,
|
| 365 |
+
"eval_precision": 0.8629290477513907,
|
| 366 |
+
"eval_recall": 0.848127600554785,
|
| 367 |
+
"eval_runtime": 24.2059,
|
| 368 |
+
"eval_samples_per_second": 119.145,
|
| 369 |
+
"eval_steps_per_second": 14.914,
|
| 370 |
+
"step": 6099
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 20.0,
|
| 374 |
+
"grad_norm": 1.7572256326675415,
|
| 375 |
+
"learning_rate": 2.4138718220394167e-05,
|
| 376 |
+
"loss": 0.2052,
|
| 377 |
+
"step": 6420
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 20.0,
|
| 381 |
+
"eval_accuracy": 0.8429264909847434,
|
| 382 |
+
"eval_f1": 0.8427565344137952,
|
| 383 |
+
"eval_loss": 0.6802518367767334,
|
| 384 |
+
"eval_precision": 0.8502279276035353,
|
| 385 |
+
"eval_recall": 0.8429264909847434,
|
| 386 |
+
"eval_runtime": 23.6557,
|
| 387 |
+
"eval_samples_per_second": 121.916,
|
| 388 |
+
"eval_steps_per_second": 15.261,
|
| 389 |
+
"step": 6420
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 21.0,
|
| 393 |
+
"grad_norm": 11.348849296569824,
|
| 394 |
+
"learning_rate": 1.8785189659922232e-05,
|
| 395 |
+
"loss": 0.1533,
|
| 396 |
+
"step": 6741
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 21.0,
|
| 400 |
+
"eval_accuracy": 0.8734396671289875,
|
| 401 |
+
"eval_f1": 0.870883732217841,
|
| 402 |
+
"eval_loss": 0.546351432800293,
|
| 403 |
+
"eval_precision": 0.8698420453848273,
|
| 404 |
+
"eval_recall": 0.8734396671289875,
|
| 405 |
+
"eval_runtime": 23.8505,
|
| 406 |
+
"eval_samples_per_second": 120.92,
|
| 407 |
+
"eval_steps_per_second": 15.136,
|
| 408 |
+
"step": 6741
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 22.0,
|
| 412 |
+
"grad_norm": 54.87451171875,
|
| 413 |
+
"learning_rate": 1.396351233934956e-05,
|
| 414 |
+
"loss": 0.1175,
|
| 415 |
+
"step": 7062
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 22.0,
|
| 419 |
+
"eval_accuracy": 0.8685852981969486,
|
| 420 |
+
"eval_f1": 0.8672514399266257,
|
| 421 |
+
"eval_loss": 0.5572792291641235,
|
| 422 |
+
"eval_precision": 0.8667361486336195,
|
| 423 |
+
"eval_recall": 0.8685852981969486,
|
| 424 |
+
"eval_runtime": 23.4628,
|
| 425 |
+
"eval_samples_per_second": 122.918,
|
| 426 |
+
"eval_steps_per_second": 15.386,
|
| 427 |
+
"step": 7062
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 23.0,
|
| 431 |
+
"grad_norm": 0.37277576327323914,
|
| 432 |
+
"learning_rate": 9.755840051487997e-06,
|
| 433 |
+
"loss": 0.1218,
|
| 434 |
+
"step": 7383
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 23.0,
|
| 438 |
+
"eval_accuracy": 0.8703190013869625,
|
| 439 |
+
"eval_f1": 0.8669113876377326,
|
| 440 |
+
"eval_loss": 0.6043308973312378,
|
| 441 |
+
"eval_precision": 0.8680623901339135,
|
| 442 |
+
"eval_recall": 0.8703190013869625,
|
| 443 |
+
"eval_runtime": 23.6979,
|
| 444 |
+
"eval_samples_per_second": 121.698,
|
| 445 |
+
"eval_steps_per_second": 15.233,
|
| 446 |
+
"step": 7383
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 24.0,
|
| 450 |
+
"grad_norm": 1.1696586608886719,
|
| 451 |
+
"learning_rate": 6.233864909760889e-06,
|
| 452 |
+
"loss": 0.114,
|
| 453 |
+
"step": 7704
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 24.0,
|
| 457 |
+
"eval_accuracy": 0.871012482662968,
|
| 458 |
+
"eval_f1": 0.8692580136075093,
|
| 459 |
+
"eval_loss": 0.5944604277610779,
|
| 460 |
+
"eval_precision": 0.8705845000186183,
|
| 461 |
+
"eval_recall": 0.871012482662968,
|
| 462 |
+
"eval_runtime": 22.9273,
|
| 463 |
+
"eval_samples_per_second": 125.789,
|
| 464 |
+
"eval_steps_per_second": 15.745,
|
| 465 |
+
"step": 7704
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 25.0,
|
| 469 |
+
"grad_norm": 0.24613961577415466,
|
| 470 |
+
"learning_rate": 3.457595827424931e-06,
|
| 471 |
+
"loss": 0.104,
|
| 472 |
+
"step": 8025
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 25.0,
|
| 476 |
+
"eval_accuracy": 0.8765603328710125,
|
| 477 |
+
"eval_f1": 0.8751858009450587,
|
| 478 |
+
"eval_loss": 0.5850355625152588,
|
| 479 |
+
"eval_precision": 0.8753430843268125,
|
| 480 |
+
"eval_recall": 0.8765603328710125,
|
| 481 |
+
"eval_runtime": 23.185,
|
| 482 |
+
"eval_samples_per_second": 124.391,
|
| 483 |
+
"eval_steps_per_second": 15.57,
|
| 484 |
+
"step": 8025
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 26.0,
|
| 488 |
+
"grad_norm": 15.901623725891113,
|
| 489 |
+
"learning_rate": 1.4743360601349622e-06,
|
| 490 |
+
"loss": 0.0752,
|
| 491 |
+
"step": 8346
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 26.0,
|
| 495 |
+
"eval_accuracy": 0.8782940360610264,
|
| 496 |
+
"eval_f1": 0.8757212451946566,
|
| 497 |
+
"eval_loss": 0.5867504477500916,
|
| 498 |
+
"eval_precision": 0.8747328684004348,
|
| 499 |
+
"eval_recall": 0.8782940360610264,
|
| 500 |
+
"eval_runtime": 24.8645,
|
| 501 |
+
"eval_samples_per_second": 115.989,
|
| 502 |
+
"eval_steps_per_second": 14.519,
|
| 503 |
+
"step": 8346
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 27.0,
|
| 507 |
+
"grad_norm": 0.1058092936873436,
|
| 508 |
+
"learning_rate": 3.1787723291717977e-07,
|
| 509 |
+
"loss": 0.1309,
|
| 510 |
+
"step": 8667
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 27.0,
|
| 514 |
+
"eval_accuracy": 0.8786407766990292,
|
| 515 |
+
"eval_f1": 0.876119808222944,
|
| 516 |
+
"eval_loss": 0.5839141011238098,
|
| 517 |
+
"eval_precision": 0.875311708483151,
|
| 518 |
+
"eval_recall": 0.8786407766990292,
|
| 519 |
+
"eval_runtime": 24.8375,
|
| 520 |
+
"eval_samples_per_second": 116.115,
|
| 521 |
+
"eval_steps_per_second": 14.534,
|
| 522 |
+
"step": 8667
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 27.0,
|
| 526 |
+
"step": 8667,
|
| 527 |
+
"total_flos": 1.0729711727592948e+19,
|
| 528 |
+
"train_loss": 0.5187536703752126,
|
| 529 |
+
"train_runtime": 3224.0652,
|
| 530 |
+
"train_samples_per_second": 159.054,
|
| 531 |
+
"train_steps_per_second": 9.956
|
| 532 |
+
}
|
| 533 |
+
],
|
| 534 |
+
"logging_steps": 500,
|
| 535 |
+
"max_steps": 32100,
|
| 536 |
+
"num_input_tokens_seen": 0,
|
| 537 |
+
"num_train_epochs": 100,
|
| 538 |
+
"save_steps": 500,
|
| 539 |
+
"total_flos": 1.0729711727592948e+19,
|
| 540 |
+
"train_batch_size": 16,
|
| 541 |
+
"trial_name": null,
|
| 542 |
+
"trial_params": null
|
| 543 |
+
}
|