End of training
Browse files- README.md +1 -1
- all_results.json +16 -0
- eval_results.json +11 -0
- train_results.json +8 -0
- trainer_state.json +650 -0
README.md
CHANGED
|
@@ -44,7 +44,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 44 |
|
| 45 |
This model is a fine-tuned version of [microsoft/resnet-50](https://huggingface.co/microsoft/resnet-50) on the imagefolder dataset.
|
| 46 |
It achieves the following results on the evaluation set:
|
| 47 |
-
- Loss: 0.
|
| 48 |
- Precision: 0.9909
|
| 49 |
- Recall: 0.9939
|
| 50 |
- F1: 0.9923
|
|
|
|
| 44 |
|
| 45 |
This model is a fine-tuned version of [microsoft/resnet-50](https://huggingface.co/microsoft/resnet-50) on the imagefolder dataset.
|
| 46 |
It achieves the following results on the evaluation set:
|
| 47 |
+
- Loss: 0.0794
|
| 48 |
- Precision: 0.9909
|
| 49 |
- Recall: 0.9939
|
| 50 |
- F1: 0.9923
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.85,
|
| 3 |
+
"eval_accuracy": 0.9906542056074766,
|
| 4 |
+
"eval_f1": 0.9922719141323793,
|
| 5 |
+
"eval_loss": 0.07943902164697647,
|
| 6 |
+
"eval_precision": 0.990909090909091,
|
| 7 |
+
"eval_recall": 0.9939393939393939,
|
| 8 |
+
"eval_runtime": 36.1652,
|
| 9 |
+
"eval_samples_per_second": 2.959,
|
| 10 |
+
"eval_steps_per_second": 0.387,
|
| 11 |
+
"total_flos": 1.7791424076806554e+17,
|
| 12 |
+
"train_loss": 0.16752889706538274,
|
| 13 |
+
"train_runtime": 3044.6611,
|
| 14 |
+
"train_samples_per_second": 2.785,
|
| 15 |
+
"train_steps_per_second": 0.085
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.85,
|
| 3 |
+
"eval_accuracy": 0.9906542056074766,
|
| 4 |
+
"eval_f1": 0.9922719141323793,
|
| 5 |
+
"eval_loss": 0.07943902164697647,
|
| 6 |
+
"eval_precision": 0.990909090909091,
|
| 7 |
+
"eval_recall": 0.9939393939393939,
|
| 8 |
+
"eval_runtime": 36.1652,
|
| 9 |
+
"eval_samples_per_second": 2.959,
|
| 10 |
+
"eval_steps_per_second": 0.387
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.85,
|
| 3 |
+
"total_flos": 1.7791424076806554e+17,
|
| 4 |
+
"train_loss": 0.16752889706538274,
|
| 5 |
+
"train_runtime": 3044.6611,
|
| 6 |
+
"train_samples_per_second": 2.785,
|
| 7 |
+
"train_steps_per_second": 0.085
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,650 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9906542056074766,
|
| 3 |
+
"best_model_checkpoint": "resnet-50-finetuned-FBark\\checkpoint-182",
|
| 4 |
+
"epoch": 19.849056603773583,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 260,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.98,
|
| 13 |
+
"step": 13,
|
| 14 |
+
"train_accuracy": 0.44339622641509435,
|
| 15 |
+
"train_f1": 0.30273443406045597,
|
| 16 |
+
"train_loss": 1.3222538232803345,
|
| 17 |
+
"train_precision": 0.4409161713465874,
|
| 18 |
+
"train_recall": 0.40814882032667876,
|
| 19 |
+
"train_runtime": 131.484,
|
| 20 |
+
"train_samples_per_second": 3.225,
|
| 21 |
+
"train_steps_per_second": 0.403
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.98,
|
| 25 |
+
"grad_norm": 2.9973506927490234,
|
| 26 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 27 |
+
"loss": 1.4155,
|
| 28 |
+
"step": 13
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.98,
|
| 32 |
+
"eval_accuracy": 0.38317757009345793,
|
| 33 |
+
"eval_f1": 0.2502374169040836,
|
| 34 |
+
"eval_loss": 1.2673313617706299,
|
| 35 |
+
"eval_precision": 0.3522222222222222,
|
| 36 |
+
"eval_recall": 0.3991341991341991,
|
| 37 |
+
"eval_runtime": 47.8505,
|
| 38 |
+
"eval_samples_per_second": 2.236,
|
| 39 |
+
"eval_steps_per_second": 0.293,
|
| 40 |
+
"step": 13
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.96,
|
| 44 |
+
"step": 26,
|
| 45 |
+
"train_accuracy": 0.5094339622641509,
|
| 46 |
+
"train_f1": 0.3768708281840832,
|
| 47 |
+
"train_loss": 1.2979044914245605,
|
| 48 |
+
"train_precision": 0.5660861330698287,
|
| 49 |
+
"train_recall": 0.4677431408134941,
|
| 50 |
+
"train_runtime": 148.3187,
|
| 51 |
+
"train_samples_per_second": 2.859,
|
| 52 |
+
"train_steps_per_second": 0.357
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.96,
|
| 56 |
+
"grad_norm": 1.2892597913742065,
|
| 57 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 58 |
+
"loss": 1.389,
|
| 59 |
+
"step": 26
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.96,
|
| 63 |
+
"eval_accuracy": 0.42990654205607476,
|
| 64 |
+
"eval_f1": 0.31675399012689665,
|
| 65 |
+
"eval_loss": 1.2493001222610474,
|
| 66 |
+
"eval_precision": 0.3496434937611408,
|
| 67 |
+
"eval_recall": 0.4454545454545455,
|
| 68 |
+
"eval_runtime": 46.5066,
|
| 69 |
+
"eval_samples_per_second": 2.301,
|
| 70 |
+
"eval_steps_per_second": 0.301,
|
| 71 |
+
"step": 26
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 2.94,
|
| 75 |
+
"step": 39,
|
| 76 |
+
"train_accuracy": 0.4811320754716981,
|
| 77 |
+
"train_f1": 0.3565462103231007,
|
| 78 |
+
"train_loss": 1.2638133764266968,
|
| 79 |
+
"train_precision": 0.35321937321937324,
|
| 80 |
+
"train_recall": 0.44614604462474644,
|
| 81 |
+
"train_runtime": 133.803,
|
| 82 |
+
"train_samples_per_second": 3.169,
|
| 83 |
+
"train_steps_per_second": 0.396
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.94,
|
| 87 |
+
"grad_norm": 1.521682620048523,
|
| 88 |
+
"learning_rate": 8.999999999999999e-05,
|
| 89 |
+
"loss": 1.3742,
|
| 90 |
+
"step": 39
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.94,
|
| 94 |
+
"eval_accuracy": 0.4392523364485981,
|
| 95 |
+
"eval_f1": 0.33554006968641115,
|
| 96 |
+
"eval_loss": 1.223555326461792,
|
| 97 |
+
"eval_precision": 0.3334061930783242,
|
| 98 |
+
"eval_recall": 0.4536796536796537,
|
| 99 |
+
"eval_runtime": 47.089,
|
| 100 |
+
"eval_samples_per_second": 2.272,
|
| 101 |
+
"eval_steps_per_second": 0.297,
|
| 102 |
+
"step": 39
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 4.0,
|
| 106 |
+
"step": 53,
|
| 107 |
+
"train_accuracy": 0.5683962264150944,
|
| 108 |
+
"train_f1": 0.4573118754465906,
|
| 109 |
+
"train_loss": 1.220451831817627,
|
| 110 |
+
"train_precision": 0.6263221242541064,
|
| 111 |
+
"train_recall": 0.5256962525336195,
|
| 112 |
+
"train_runtime": 123.6293,
|
| 113 |
+
"train_samples_per_second": 3.43,
|
| 114 |
+
"train_steps_per_second": 0.429
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 4.0,
|
| 118 |
+
"grad_norm": 1.533220648765564,
|
| 119 |
+
"learning_rate": 0.0001223076923076923,
|
| 120 |
+
"loss": 1.2592,
|
| 121 |
+
"step": 53
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 4.0,
|
| 125 |
+
"eval_accuracy": 0.5233644859813084,
|
| 126 |
+
"eval_f1": 0.459322365844105,
|
| 127 |
+
"eval_loss": 1.1768313646316528,
|
| 128 |
+
"eval_precision": 0.6458215962441315,
|
| 129 |
+
"eval_recall": 0.5359683794466403,
|
| 130 |
+
"eval_runtime": 52.0476,
|
| 131 |
+
"eval_samples_per_second": 2.056,
|
| 132 |
+
"eval_steps_per_second": 0.269,
|
| 133 |
+
"step": 53
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 4.98,
|
| 137 |
+
"step": 66,
|
| 138 |
+
"train_accuracy": 0.5990566037735849,
|
| 139 |
+
"train_f1": 0.49108922737322597,
|
| 140 |
+
"train_loss": 1.1574139595031738,
|
| 141 |
+
"train_precision": 0.6493666135599222,
|
| 142 |
+
"train_recall": 0.55391404130241,
|
| 143 |
+
"train_runtime": 122.643,
|
| 144 |
+
"train_samples_per_second": 3.457,
|
| 145 |
+
"train_steps_per_second": 0.432
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 4.98,
|
| 149 |
+
"grad_norm": 1.2271511554718018,
|
| 150 |
+
"learning_rate": 0.00015230769230769228,
|
| 151 |
+
"loss": 1.287,
|
| 152 |
+
"step": 66
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.98,
|
| 156 |
+
"eval_accuracy": 0.5607476635514018,
|
| 157 |
+
"eval_f1": 0.48464696223316917,
|
| 158 |
+
"eval_loss": 1.1236993074417114,
|
| 159 |
+
"eval_precision": 0.6436363636363637,
|
| 160 |
+
"eval_recall": 0.5723320158102767,
|
| 161 |
+
"eval_runtime": 41.452,
|
| 162 |
+
"eval_samples_per_second": 2.581,
|
| 163 |
+
"eval_steps_per_second": 0.338,
|
| 164 |
+
"step": 66
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 5.96,
|
| 168 |
+
"step": 79,
|
| 169 |
+
"train_accuracy": 0.6226415094339622,
|
| 170 |
+
"train_f1": 0.5420012941762071,
|
| 171 |
+
"train_loss": 1.0438413619995117,
|
| 172 |
+
"train_precision": 0.8243247706573527,
|
| 173 |
+
"train_recall": 0.5806628230445157,
|
| 174 |
+
"train_runtime": 120.0289,
|
| 175 |
+
"train_samples_per_second": 3.532,
|
| 176 |
+
"train_steps_per_second": 0.442
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.96,
|
| 180 |
+
"grad_norm": 1.7175101041793823,
|
| 181 |
+
"learning_rate": 0.00018230769230769228,
|
| 182 |
+
"loss": 1.2337,
|
| 183 |
+
"step": 79
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 5.96,
|
| 187 |
+
"eval_accuracy": 0.5981308411214953,
|
| 188 |
+
"eval_f1": 0.5485250474059656,
|
| 189 |
+
"eval_loss": 1.0242024660110474,
|
| 190 |
+
"eval_precision": 0.865625,
|
| 191 |
+
"eval_recall": 0.6083380387728214,
|
| 192 |
+
"eval_runtime": 45.0579,
|
| 193 |
+
"eval_samples_per_second": 2.375,
|
| 194 |
+
"eval_steps_per_second": 0.311,
|
| 195 |
+
"step": 79
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 6.94,
|
| 199 |
+
"step": 92,
|
| 200 |
+
"train_accuracy": 0.6745283018867925,
|
| 201 |
+
"train_f1": 0.6334819902275806,
|
| 202 |
+
"train_loss": 0.935691237449646,
|
| 203 |
+
"train_precision": 0.8151341081504686,
|
| 204 |
+
"train_recall": 0.6419837529777849,
|
| 205 |
+
"train_runtime": 130.3177,
|
| 206 |
+
"train_samples_per_second": 3.254,
|
| 207 |
+
"train_steps_per_second": 0.407
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 6.94,
|
| 211 |
+
"grad_norm": 1.9299957752227783,
|
| 212 |
+
"learning_rate": 0.0002123076923076923,
|
| 213 |
+
"loss": 1.1327,
|
| 214 |
+
"step": 92
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 6.94,
|
| 218 |
+
"eval_accuracy": 0.7102803738317757,
|
| 219 |
+
"eval_f1": 0.7062065516379095,
|
| 220 |
+
"eval_loss": 0.9071274995803833,
|
| 221 |
+
"eval_precision": 0.8466666666666667,
|
| 222 |
+
"eval_recall": 0.7193111236589498,
|
| 223 |
+
"eval_runtime": 42.5223,
|
| 224 |
+
"eval_samples_per_second": 2.516,
|
| 225 |
+
"eval_steps_per_second": 0.329,
|
| 226 |
+
"step": 92
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"step": 106,
|
| 231 |
+
"train_accuracy": 0.7971698113207547,
|
| 232 |
+
"train_f1": 0.7766198391591397,
|
| 233 |
+
"train_loss": 0.7271670699119568,
|
| 234 |
+
"train_precision": 0.8651733397747755,
|
| 235 |
+
"train_recall": 0.7787121922468249,
|
| 236 |
+
"train_runtime": 114.2154,
|
| 237 |
+
"train_samples_per_second": 3.712,
|
| 238 |
+
"train_steps_per_second": 0.464
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 8.0,
|
| 242 |
+
"grad_norm": 2.2226734161376953,
|
| 243 |
+
"learning_rate": 0.0002446153846153846,
|
| 244 |
+
"loss": 0.9421,
|
| 245 |
+
"step": 106
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 8.0,
|
| 249 |
+
"eval_accuracy": 0.7757009345794392,
|
| 250 |
+
"eval_f1": 0.7750786554065242,
|
| 251 |
+
"eval_loss": 0.7316702008247375,
|
| 252 |
+
"eval_precision": 0.8393822843822842,
|
| 253 |
+
"eval_recall": 0.78223226049313,
|
| 254 |
+
"eval_runtime": 37.8971,
|
| 255 |
+
"eval_samples_per_second": 2.823,
|
| 256 |
+
"eval_steps_per_second": 0.369,
|
| 257 |
+
"step": 106
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 8.98,
|
| 261 |
+
"step": 119,
|
| 262 |
+
"train_accuracy": 0.8867924528301887,
|
| 263 |
+
"train_f1": 0.8861109763599291,
|
| 264 |
+
"train_loss": 0.5811343193054199,
|
| 265 |
+
"train_precision": 0.9043399022853599,
|
| 266 |
+
"train_recall": 0.8790563087733421,
|
| 267 |
+
"train_runtime": 114.7718,
|
| 268 |
+
"train_samples_per_second": 3.694,
|
| 269 |
+
"train_steps_per_second": 0.462
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 8.98,
|
| 273 |
+
"grad_norm": 2.4611976146698,
|
| 274 |
+
"learning_rate": 0.0002746153846153846,
|
| 275 |
+
"loss": 0.8662,
|
| 276 |
+
"step": 119
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 8.98,
|
| 280 |
+
"eval_accuracy": 0.8785046728971962,
|
| 281 |
+
"eval_f1": 0.8814210982815635,
|
| 282 |
+
"eval_loss": 0.5671026110649109,
|
| 283 |
+
"eval_precision": 0.8899470899470898,
|
| 284 |
+
"eval_recall": 0.880764163372859,
|
| 285 |
+
"eval_runtime": 36.298,
|
| 286 |
+
"eval_samples_per_second": 2.948,
|
| 287 |
+
"eval_steps_per_second": 0.386,
|
| 288 |
+
"step": 119
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 9.96,
|
| 292 |
+
"step": 132,
|
| 293 |
+
"train_accuracy": 0.9316037735849056,
|
| 294 |
+
"train_f1": 0.9311136538837689,
|
| 295 |
+
"train_loss": 0.40401148796081543,
|
| 296 |
+
"train_precision": 0.9336201141469944,
|
| 297 |
+
"train_recall": 0.9344224750671604,
|
| 298 |
+
"train_runtime": 117.7535,
|
| 299 |
+
"train_samples_per_second": 3.601,
|
| 300 |
+
"train_steps_per_second": 0.45
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 9.96,
|
| 304 |
+
"grad_norm": 2.956929922103882,
|
| 305 |
+
"learning_rate": 0.0002994871794871795,
|
| 306 |
+
"loss": 0.7054,
|
| 307 |
+
"step": 132
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 9.96,
|
| 311 |
+
"eval_accuracy": 0.9158878504672897,
|
| 312 |
+
"eval_f1": 0.9176957026713124,
|
| 313 |
+
"eval_loss": 0.3391460180282593,
|
| 314 |
+
"eval_precision": 0.922962962962963,
|
| 315 |
+
"eval_recall": 0.9159796725014117,
|
| 316 |
+
"eval_runtime": 36.2906,
|
| 317 |
+
"eval_samples_per_second": 2.948,
|
| 318 |
+
"eval_steps_per_second": 0.386,
|
| 319 |
+
"step": 132
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 10.94,
|
| 323 |
+
"step": 145,
|
| 324 |
+
"train_accuracy": 0.9410377358490566,
|
| 325 |
+
"train_f1": 0.9417046133414193,
|
| 326 |
+
"train_loss": 0.2951917350292206,
|
| 327 |
+
"train_precision": 0.9450684559692762,
|
| 328 |
+
"train_recall": 0.9400955207310657,
|
| 329 |
+
"train_runtime": 119.8783,
|
| 330 |
+
"train_samples_per_second": 3.537,
|
| 331 |
+
"train_steps_per_second": 0.442
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 10.94,
|
| 335 |
+
"grad_norm": 2.339923143386841,
|
| 336 |
+
"learning_rate": 0.00029615384615384616,
|
| 337 |
+
"loss": 0.595,
|
| 338 |
+
"step": 145
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 10.94,
|
| 342 |
+
"eval_accuracy": 0.9626168224299065,
|
| 343 |
+
"eval_f1": 0.9627651607231869,
|
| 344 |
+
"eval_loss": 0.24373753368854523,
|
| 345 |
+
"eval_precision": 0.9635610766045548,
|
| 346 |
+
"eval_recall": 0.9639939770374552,
|
| 347 |
+
"eval_runtime": 38.7327,
|
| 348 |
+
"eval_samples_per_second": 2.763,
|
| 349 |
+
"eval_steps_per_second": 0.361,
|
| 350 |
+
"step": 145
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.0,
|
| 354 |
+
"step": 159,
|
| 355 |
+
"train_accuracy": 0.9127358490566038,
|
| 356 |
+
"train_f1": 0.9180456649072474,
|
| 357 |
+
"train_loss": 0.27252715826034546,
|
| 358 |
+
"train_precision": 0.933049846444602,
|
| 359 |
+
"train_recall": 0.9085866769706422,
|
| 360 |
+
"train_runtime": 116.9422,
|
| 361 |
+
"train_samples_per_second": 3.626,
|
| 362 |
+
"train_steps_per_second": 0.453
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 12.0,
|
| 366 |
+
"grad_norm": 4.72435188293457,
|
| 367 |
+
"learning_rate": 0.00029256410256410254,
|
| 368 |
+
"loss": 0.4509,
|
| 369 |
+
"step": 159
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 12.0,
|
| 373 |
+
"eval_accuracy": 0.9252336448598131,
|
| 374 |
+
"eval_f1": 0.9260400996325433,
|
| 375 |
+
"eval_loss": 0.22485339641571045,
|
| 376 |
+
"eval_precision": 0.9322943722943723,
|
| 377 |
+
"eval_recall": 0.9283832109919066,
|
| 378 |
+
"eval_runtime": 35.9322,
|
| 379 |
+
"eval_samples_per_second": 2.978,
|
| 380 |
+
"eval_steps_per_second": 0.39,
|
| 381 |
+
"step": 159
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 12.98,
|
| 385 |
+
"step": 169,
|
| 386 |
+
"train_accuracy": 0.9363207547169812,
|
| 387 |
+
"train_f1": 0.9364699926202164,
|
| 388 |
+
"train_loss": 0.23557166755199432,
|
| 389 |
+
"train_precision": 0.9409715380405036,
|
| 390 |
+
"train_recall": 0.9348327849415139,
|
| 391 |
+
"train_runtime": 133.051,
|
| 392 |
+
"train_samples_per_second": 3.187,
|
| 393 |
+
"train_steps_per_second": 0.398
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 12.98,
|
| 397 |
+
"grad_norm": 4.02577543258667,
|
| 398 |
+
"learning_rate": 0.00011666666666666665,
|
| 399 |
+
"loss": 0.4954,
|
| 400 |
+
"step": 169
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 12.98,
|
| 404 |
+
"eval_accuracy": 0.9813084112149533,
|
| 405 |
+
"eval_f1": 0.9844101823682084,
|
| 406 |
+
"eval_loss": 0.09772800654172897,
|
| 407 |
+
"eval_precision": 0.9848484848484848,
|
| 408 |
+
"eval_recall": 0.9844155844155844,
|
| 409 |
+
"eval_runtime": 46.2598,
|
| 410 |
+
"eval_samples_per_second": 2.313,
|
| 411 |
+
"eval_steps_per_second": 0.303,
|
| 412 |
+
"step": 169
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 13.96,
|
| 416 |
+
"step": 182,
|
| 417 |
+
"train_accuracy": 0.9693396226415094,
|
| 418 |
+
"train_f1": 0.9713518400105837,
|
| 419 |
+
"train_loss": 0.16475237905979156,
|
| 420 |
+
"train_precision": 0.9745377270610154,
|
| 421 |
+
"train_recall": 0.9693937925419673,
|
| 422 |
+
"train_runtime": 138.0499,
|
| 423 |
+
"train_samples_per_second": 3.071,
|
| 424 |
+
"train_steps_per_second": 0.384
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 13.96,
|
| 428 |
+
"grad_norm": 2.480182647705078,
|
| 429 |
+
"learning_rate": 9.999999999999999e-05,
|
| 430 |
+
"loss": 0.4842,
|
| 431 |
+
"step": 182
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 13.96,
|
| 435 |
+
"eval_accuracy": 0.9906542056074766,
|
| 436 |
+
"eval_f1": 0.9922719141323793,
|
| 437 |
+
"eval_loss": 0.07943902164697647,
|
| 438 |
+
"eval_precision": 0.990909090909091,
|
| 439 |
+
"eval_recall": 0.9939393939393939,
|
| 440 |
+
"eval_runtime": 60.5689,
|
| 441 |
+
"eval_samples_per_second": 1.767,
|
| 442 |
+
"eval_steps_per_second": 0.231,
|
| 443 |
+
"step": 182
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 14.94,
|
| 447 |
+
"step": 195,
|
| 448 |
+
"train_accuracy": 0.9716981132075472,
|
| 449 |
+
"train_f1": 0.9719681475881605,
|
| 450 |
+
"train_loss": 0.1421472579240799,
|
| 451 |
+
"train_precision": 0.9732013799086496,
|
| 452 |
+
"train_recall": 0.9717206659817743,
|
| 453 |
+
"train_runtime": 161.6921,
|
| 454 |
+
"train_samples_per_second": 2.622,
|
| 455 |
+
"train_steps_per_second": 0.328
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 14.94,
|
| 459 |
+
"grad_norm": 3.177818775177002,
|
| 460 |
+
"learning_rate": 8.333333333333333e-05,
|
| 461 |
+
"loss": 0.4194,
|
| 462 |
+
"step": 195
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 14.94,
|
| 466 |
+
"eval_accuracy": 0.9906542056074766,
|
| 467 |
+
"eval_f1": 0.9922719141323793,
|
| 468 |
+
"eval_loss": 0.08730577677488327,
|
| 469 |
+
"eval_precision": 0.990909090909091,
|
| 470 |
+
"eval_recall": 0.9939393939393939,
|
| 471 |
+
"eval_runtime": 47.0651,
|
| 472 |
+
"eval_samples_per_second": 2.273,
|
| 473 |
+
"eval_steps_per_second": 0.297,
|
| 474 |
+
"step": 195
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 16.0,
|
| 478 |
+
"step": 209,
|
| 479 |
+
"train_accuracy": 0.964622641509434,
|
| 480 |
+
"train_f1": 0.9654786147790538,
|
| 481 |
+
"train_loss": 0.170837864279747,
|
| 482 |
+
"train_precision": 0.9662175617420704,
|
| 483 |
+
"train_recall": 0.9652730803864843,
|
| 484 |
+
"train_runtime": 122.9869,
|
| 485 |
+
"train_samples_per_second": 3.448,
|
| 486 |
+
"train_steps_per_second": 0.431
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 16.0,
|
| 490 |
+
"grad_norm": 2.836935520172119,
|
| 491 |
+
"learning_rate": 6.538461538461539e-05,
|
| 492 |
+
"loss": 0.4353,
|
| 493 |
+
"step": 209
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 16.0,
|
| 497 |
+
"eval_accuracy": 0.9906542056074766,
|
| 498 |
+
"eval_f1": 0.9922719141323793,
|
| 499 |
+
"eval_loss": 0.07477138936519623,
|
| 500 |
+
"eval_precision": 0.990909090909091,
|
| 501 |
+
"eval_recall": 0.9939393939393939,
|
| 502 |
+
"eval_runtime": 45.8048,
|
| 503 |
+
"eval_samples_per_second": 2.336,
|
| 504 |
+
"eval_steps_per_second": 0.306,
|
| 505 |
+
"step": 209
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 16.98,
|
| 509 |
+
"step": 222,
|
| 510 |
+
"train_accuracy": 0.9693396226415094,
|
| 511 |
+
"train_f1": 0.9698753879001953,
|
| 512 |
+
"train_loss": 0.13738253712654114,
|
| 513 |
+
"train_precision": 0.9698721656123299,
|
| 514 |
+
"train_recall": 0.9699632431180802,
|
| 515 |
+
"train_runtime": 120.4492,
|
| 516 |
+
"train_samples_per_second": 3.52,
|
| 517 |
+
"train_steps_per_second": 0.44
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 16.98,
|
| 521 |
+
"grad_norm": 2.6075973510742188,
|
| 522 |
+
"learning_rate": 4.871794871794872e-05,
|
| 523 |
+
"loss": 0.3996,
|
| 524 |
+
"step": 222
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 16.98,
|
| 528 |
+
"eval_accuracy": 0.9906542056074766,
|
| 529 |
+
"eval_f1": 0.9922719141323793,
|
| 530 |
+
"eval_loss": 0.06090036779642105,
|
| 531 |
+
"eval_precision": 0.990909090909091,
|
| 532 |
+
"eval_recall": 0.9939393939393939,
|
| 533 |
+
"eval_runtime": 46.0048,
|
| 534 |
+
"eval_samples_per_second": 2.326,
|
| 535 |
+
"eval_steps_per_second": 0.304,
|
| 536 |
+
"step": 222
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 17.96,
|
| 540 |
+
"step": 235,
|
| 541 |
+
"train_accuracy": 0.9669811320754716,
|
| 542 |
+
"train_f1": 0.9669883931291977,
|
| 543 |
+
"train_loss": 0.15114478766918182,
|
| 544 |
+
"train_precision": 0.9689529722559259,
|
| 545 |
+
"train_recall": 0.9654587281320716,
|
| 546 |
+
"train_runtime": 117.4904,
|
| 547 |
+
"train_samples_per_second": 3.609,
|
| 548 |
+
"train_steps_per_second": 0.451
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 17.96,
|
| 552 |
+
"grad_norm": 3.3314599990844727,
|
| 553 |
+
"learning_rate": 3.2051282051282044e-05,
|
| 554 |
+
"loss": 0.4445,
|
| 555 |
+
"step": 235
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 17.96,
|
| 559 |
+
"eval_accuracy": 0.9906542056074766,
|
| 560 |
+
"eval_f1": 0.9922719141323793,
|
| 561 |
+
"eval_loss": 0.06219491362571716,
|
| 562 |
+
"eval_precision": 0.990909090909091,
|
| 563 |
+
"eval_recall": 0.9939393939393939,
|
| 564 |
+
"eval_runtime": 44.1393,
|
| 565 |
+
"eval_samples_per_second": 2.424,
|
| 566 |
+
"eval_steps_per_second": 0.317,
|
| 567 |
+
"step": 235
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 18.94,
|
| 571 |
+
"step": 248,
|
| 572 |
+
"train_accuracy": 0.9740566037735849,
|
| 573 |
+
"train_f1": 0.973824007234904,
|
| 574 |
+
"train_loss": 0.13619276881217957,
|
| 575 |
+
"train_precision": 0.9751913290746111,
|
| 576 |
+
"train_recall": 0.9728311732620997,
|
| 577 |
+
"train_runtime": 116.7219,
|
| 578 |
+
"train_samples_per_second": 3.633,
|
| 579 |
+
"train_steps_per_second": 0.454
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 18.94,
|
| 583 |
+
"grad_norm": 3.210602045059204,
|
| 584 |
+
"learning_rate": 1.5384615384615384e-05,
|
| 585 |
+
"loss": 0.3838,
|
| 586 |
+
"step": 248
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 18.94,
|
| 590 |
+
"eval_accuracy": 0.9906542056074766,
|
| 591 |
+
"eval_f1": 0.9922719141323793,
|
| 592 |
+
"eval_loss": 0.0594821497797966,
|
| 593 |
+
"eval_precision": 0.990909090909091,
|
| 594 |
+
"eval_recall": 0.9939393939393939,
|
| 595 |
+
"eval_runtime": 36.336,
|
| 596 |
+
"eval_samples_per_second": 2.945,
|
| 597 |
+
"eval_steps_per_second": 0.385,
|
| 598 |
+
"step": 248
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 19.85,
|
| 602 |
+
"step": 260,
|
| 603 |
+
"train_accuracy": 0.9599056603773585,
|
| 604 |
+
"train_f1": 0.9607465386599348,
|
| 605 |
+
"train_loss": 0.16524513065814972,
|
| 606 |
+
"train_precision": 0.9620981874857844,
|
| 607 |
+
"train_recall": 0.9597895488701212,
|
| 608 |
+
"train_runtime": 115.5845,
|
| 609 |
+
"train_samples_per_second": 3.668,
|
| 610 |
+
"train_steps_per_second": 0.459
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 19.85,
|
| 614 |
+
"grad_norm": 4.980319976806641,
|
| 615 |
+
"learning_rate": 0.0,
|
| 616 |
+
"loss": 0.4,
|
| 617 |
+
"step": 260
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 19.85,
|
| 621 |
+
"eval_accuracy": 0.9906542056074766,
|
| 622 |
+
"eval_f1": 0.9922719141323793,
|
| 623 |
+
"eval_loss": 0.06939452886581421,
|
| 624 |
+
"eval_precision": 0.990909090909091,
|
| 625 |
+
"eval_recall": 0.9939393939393939,
|
| 626 |
+
"eval_runtime": 35.1775,
|
| 627 |
+
"eval_samples_per_second": 3.042,
|
| 628 |
+
"eval_steps_per_second": 0.398,
|
| 629 |
+
"step": 260
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 19.85,
|
| 633 |
+
"step": 260,
|
| 634 |
+
"total_flos": 1.7791424076806554e+17,
|
| 635 |
+
"train_loss": 0.16752889706538274,
|
| 636 |
+
"train_runtime": 3044.6611,
|
| 637 |
+
"train_samples_per_second": 2.785,
|
| 638 |
+
"train_steps_per_second": 0.085
|
| 639 |
+
}
|
| 640 |
+
],
|
| 641 |
+
"logging_steps": 50,
|
| 642 |
+
"max_steps": 260,
|
| 643 |
+
"num_input_tokens_seen": 0,
|
| 644 |
+
"num_train_epochs": 20,
|
| 645 |
+
"save_steps": 500,
|
| 646 |
+
"total_flos": 1.7791424076806554e+17,
|
| 647 |
+
"train_batch_size": 8,
|
| 648 |
+
"trial_name": null,
|
| 649 |
+
"trial_params": null
|
| 650 |
+
}
|