Aman
commited on
Commit
·
84df304
1
Parent(s):
a0412ea
Upload 7 files
Browse files- all_results.json +14 -11
- config.json +15 -34
- eval_results.json +9 -6
- preprocessor_config.json +2 -7
- train_results.json +6 -6
- trainer_state.json +362 -101
- training_args.bin +2 -2
all_results.json
CHANGED
|
@@ -1,13 +1,16 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"
|
| 9 |
-
"
|
| 10 |
-
"
|
| 11 |
-
"
|
| 12 |
-
"
|
|
|
|
|
|
|
|
|
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9960906958561376,
|
| 4 |
+
"eval_f1": 0.9960906958561376,
|
| 5 |
+
"eval_loss": 0.012254622764885426,
|
| 6 |
+
"eval_precision": 0.9960906958561376,
|
| 7 |
+
"eval_recall": 0.9960906958561376,
|
| 8 |
+
"eval_runtime": 436.7862,
|
| 9 |
+
"eval_samples_per_second": 5.856,
|
| 10 |
+
"eval_steps_per_second": 0.183,
|
| 11 |
+
"total_flos": 5.409667845997019e+18,
|
| 12 |
+
"train_loss": 0.15731672859854168,
|
| 13 |
+
"train_runtime": 52202.335,
|
| 14 |
+
"train_samples_per_second": 1.323,
|
| 15 |
+
"train_steps_per_second": 0.01
|
| 16 |
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "microsoft/swin-
|
| 3 |
"architectures": [
|
| 4 |
"SwinForImageClassification"
|
| 5 |
],
|
|
@@ -7,63 +7,44 @@
|
|
| 7 |
"depths": [
|
| 8 |
2,
|
| 9 |
2,
|
| 10 |
-
|
| 11 |
2
|
| 12 |
],
|
| 13 |
"drop_path_rate": 0.1,
|
| 14 |
-
"embed_dim":
|
| 15 |
"encoder_stride": 32,
|
| 16 |
"hidden_act": "gelu",
|
| 17 |
"hidden_dropout_prob": 0.0,
|
| 18 |
-
"hidden_size":
|
| 19 |
"id2label": {
|
| 20 |
-
"0": "
|
| 21 |
-
"1": "
|
| 22 |
-
"2": "
|
| 23 |
-
"3": "no_hemorrhage",
|
| 24 |
-
"4": "subarachnoid",
|
| 25 |
-
"5": "subdural"
|
| 26 |
},
|
| 27 |
"image_size": 224,
|
| 28 |
"initializer_range": 0.02,
|
| 29 |
"label2id": {
|
| 30 |
-
"
|
| 31 |
-
"
|
| 32 |
-
"
|
| 33 |
-
"no_hemorrhage": 3,
|
| 34 |
-
"subarachnoid": 4,
|
| 35 |
-
"subdural": 5
|
| 36 |
},
|
| 37 |
"layer_norm_eps": 1e-05,
|
| 38 |
"mlp_ratio": 4.0,
|
| 39 |
"model_type": "swin",
|
| 40 |
"num_channels": 3,
|
| 41 |
"num_heads": [
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
],
|
| 47 |
"num_layers": 4,
|
| 48 |
-
"out_features": [
|
| 49 |
-
"stage4"
|
| 50 |
-
],
|
| 51 |
-
"out_indices": [
|
| 52 |
-
4
|
| 53 |
-
],
|
| 54 |
"patch_size": 4,
|
| 55 |
"path_norm": true,
|
| 56 |
"problem_type": "single_label_classification",
|
| 57 |
"qkv_bias": true,
|
| 58 |
-
"stage_names": [
|
| 59 |
-
"stem",
|
| 60 |
-
"stage1",
|
| 61 |
-
"stage2",
|
| 62 |
-
"stage3",
|
| 63 |
-
"stage4"
|
| 64 |
-
],
|
| 65 |
"torch_dtype": "float32",
|
| 66 |
-
"transformers_version": "4.
|
| 67 |
"use_absolute_embeddings": false,
|
| 68 |
"window_size": 7
|
| 69 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "microsoft/swin-base-patch4-window7-224-in22k",
|
| 3 |
"architectures": [
|
| 4 |
"SwinForImageClassification"
|
| 5 |
],
|
|
|
|
| 7 |
"depths": [
|
| 8 |
2,
|
| 9 |
2,
|
| 10 |
+
18,
|
| 11 |
2
|
| 12 |
],
|
| 13 |
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
"encoder_stride": 32,
|
| 16 |
"hidden_act": "gelu",
|
| 17 |
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
"id2label": {
|
| 20 |
+
"0": "Gliomas Tumor",
|
| 21 |
+
"1": "Meningiomas Tumor",
|
| 22 |
+
"2": "Pituitary Tumor"
|
|
|
|
|
|
|
|
|
|
| 23 |
},
|
| 24 |
"image_size": 224,
|
| 25 |
"initializer_range": 0.02,
|
| 26 |
"label2id": {
|
| 27 |
+
"Gliomas Tumor": 0,
|
| 28 |
+
"Meningiomas Tumor": 1,
|
| 29 |
+
"Pituitary Tumor": 2
|
|
|
|
|
|
|
|
|
|
| 30 |
},
|
| 31 |
"layer_norm_eps": 1e-05,
|
| 32 |
"mlp_ratio": 4.0,
|
| 33 |
"model_type": "swin",
|
| 34 |
"num_channels": 3,
|
| 35 |
"num_heads": [
|
| 36 |
+
4,
|
| 37 |
+
8,
|
| 38 |
+
16,
|
| 39 |
+
32
|
| 40 |
],
|
| 41 |
"num_layers": 4,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
"patch_size": 4,
|
| 43 |
"path_norm": true,
|
| 44 |
"problem_type": "single_label_classification",
|
| 45 |
"qkv_bias": true,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
"torch_dtype": "float32",
|
| 47 |
+
"transformers_version": "4.23.1",
|
| 48 |
"use_absolute_embeddings": false,
|
| 49 |
"window_size": 7
|
| 50 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,11 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
|
|
|
|
|
|
|
|
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9960906958561376,
|
| 4 |
+
"eval_f1": 0.9960906958561376,
|
| 5 |
+
"eval_loss": 0.012254622764885426,
|
| 6 |
+
"eval_precision": 0.9960906958561376,
|
| 7 |
+
"eval_recall": 0.9960906958561376,
|
| 8 |
+
"eval_runtime": 436.7862,
|
| 9 |
+
"eval_samples_per_second": 5.856,
|
| 10 |
+
"eval_steps_per_second": 0.183
|
| 11 |
}
|
preprocessor_config.json
CHANGED
|
@@ -1,22 +1,17 @@
|
|
| 1 |
{
|
| 2 |
"do_normalize": true,
|
| 3 |
-
"do_rescale": true,
|
| 4 |
"do_resize": true,
|
|
|
|
| 5 |
"image_mean": [
|
| 6 |
0.485,
|
| 7 |
0.456,
|
| 8 |
0.406
|
| 9 |
],
|
| 10 |
-
"image_processor_type": "ViTImageProcessor",
|
| 11 |
"image_std": [
|
| 12 |
0.229,
|
| 13 |
0.224,
|
| 14 |
0.225
|
| 15 |
],
|
| 16 |
"resample": 3,
|
| 17 |
-
"
|
| 18 |
-
"size": {
|
| 19 |
-
"height": 224,
|
| 20 |
-
"width": 224
|
| 21 |
-
}
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"do_normalize": true,
|
|
|
|
| 3 |
"do_resize": true,
|
| 4 |
+
"feature_extractor_type": "ViTFeatureExtractor",
|
| 5 |
"image_mean": [
|
| 6 |
0.485,
|
| 7 |
0.456,
|
| 8 |
0.406
|
| 9 |
],
|
|
|
|
| 10 |
"image_std": [
|
| 11 |
0.229,
|
| 12 |
0.224,
|
| 13 |
0.225
|
| 14 |
],
|
| 15 |
"resample": 3,
|
| 16 |
+
"size": 224
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 5.409667845997019e+18,
|
| 4 |
+
"train_loss": 0.15731672859854168,
|
| 5 |
+
"train_runtime": 52202.335,
|
| 6 |
+
"train_samples_per_second": 1.323,
|
| 7 |
+
"train_steps_per_second": 0.01
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,124 +1,385 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "
|
| 4 |
-
"epoch":
|
| 5 |
-
"
|
| 6 |
-
"global_step": 10,
|
| 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 |
-
"eval_accuracy": 0.
|
| 14 |
-
"
|
| 15 |
-
"
|
| 16 |
-
"
|
| 17 |
-
"
|
| 18 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
},
|
| 20 |
{
|
| 21 |
"epoch": 2.0,
|
| 22 |
-
"eval_accuracy": 0.
|
| 23 |
-
"
|
| 24 |
-
"
|
| 25 |
-
"
|
| 26 |
-
"
|
| 27 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 3.0,
|
| 31 |
-
"eval_accuracy": 0.2857142857142857,
|
| 32 |
-
"eval_loss": 1.756097435951233,
|
| 33 |
-
"eval_runtime": 0.2838,
|
| 34 |
-
"eval_samples_per_second": 49.327,
|
| 35 |
-
"eval_steps_per_second": 3.523,
|
| 36 |
-
"step": 3
|
| 37 |
-
},
|
| 38 |
-
{
|
| 39 |
-
"epoch": 4.0,
|
| 40 |
-
"eval_accuracy": 0.35714285714285715,
|
| 41 |
-
"eval_loss": 1.7593554258346558,
|
| 42 |
-
"eval_runtime": 0.2252,
|
| 43 |
-
"eval_samples_per_second": 62.159,
|
| 44 |
-
"eval_steps_per_second": 4.44,
|
| 45 |
-
"step": 4
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 5.0,
|
| 49 |
-
"eval_accuracy": 0.35714285714285715,
|
| 50 |
-
"eval_loss": 1.7199592590332031,
|
| 51 |
-
"eval_runtime": 0.2865,
|
| 52 |
-
"eval_samples_per_second": 48.864,
|
| 53 |
-
"eval_steps_per_second": 3.49,
|
| 54 |
-
"step": 5
|
| 55 |
-
},
|
| 56 |
-
{
|
| 57 |
-
"epoch": 6.0,
|
| 58 |
-
"eval_accuracy": 0.35714285714285715,
|
| 59 |
-
"eval_loss": 1.6617456674575806,
|
| 60 |
-
"eval_runtime": 0.1666,
|
| 61 |
-
"eval_samples_per_second": 84.037,
|
| 62 |
-
"eval_steps_per_second": 6.003,
|
| 63 |
-
"step": 6
|
| 64 |
-
},
|
| 65 |
-
{
|
| 66 |
-
"epoch": 7.0,
|
| 67 |
-
"eval_accuracy": 0.35714285714285715,
|
| 68 |
-
"eval_loss": 1.6157296895980835,
|
| 69 |
-
"eval_runtime": 0.1403,
|
| 70 |
-
"eval_samples_per_second": 99.819,
|
| 71 |
-
"eval_steps_per_second": 7.13,
|
| 72 |
-
"step": 7
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 8.0,
|
| 76 |
-
"eval_accuracy": 0.5,
|
| 77 |
-
"eval_loss": 1.5828945636749268,
|
| 78 |
-
"eval_runtime": 0.1732,
|
| 79 |
-
"eval_samples_per_second": 80.834,
|
| 80 |
-
"eval_steps_per_second": 5.774,
|
| 81 |
-
"step": 8
|
| 82 |
-
},
|
| 83 |
-
{
|
| 84 |
-
"epoch": 9.0,
|
| 85 |
-
"eval_accuracy": 0.5,
|
| 86 |
-
"eval_loss": 1.5617578029632568,
|
| 87 |
-
"eval_runtime": 0.1476,
|
| 88 |
-
"eval_samples_per_second": 94.872,
|
| 89 |
-
"eval_steps_per_second": 6.777,
|
| 90 |
-
"step": 9
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"epoch": 10.0,
|
| 94 |
"learning_rate": 0.0,
|
| 95 |
-
"loss":
|
| 96 |
-
"step":
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch":
|
| 100 |
-
"eval_accuracy": 0.
|
| 101 |
-
"
|
| 102 |
-
"
|
| 103 |
-
"
|
| 104 |
-
"
|
| 105 |
-
"
|
|
|
|
|
|
|
|
|
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch":
|
| 109 |
-
"step":
|
| 110 |
-
"total_flos":
|
| 111 |
-
"train_loss":
|
| 112 |
-
"train_runtime":
|
| 113 |
-
"train_samples_per_second":
|
| 114 |
-
"train_steps_per_second": 0.
|
| 115 |
}
|
| 116 |
],
|
| 117 |
-
"
|
| 118 |
-
"
|
| 119 |
-
"
|
| 120 |
-
"save_steps": 500,
|
| 121 |
-
"total_flos": 3.132204658458624e+16,
|
| 122 |
"trial_name": null,
|
| 123 |
"trial_params": null
|
| 124 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9960906958561376,
|
| 3 |
+
"best_model_checkpoint": "Brain_Tumor_Classification_using_swin/checkpoint-540",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 540,
|
|
|
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 9.259259259259259e-06,
|
| 13 |
+
"loss": 1.0953,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.11,
|
| 18 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 19 |
+
"loss": 0.9167,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.17,
|
| 24 |
+
"learning_rate": 2.777777777777778e-05,
|
| 25 |
+
"loss": 0.5877,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.22,
|
| 30 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 31 |
+
"loss": 0.3904,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.28,
|
| 36 |
+
"learning_rate": 4.62962962962963e-05,
|
| 37 |
+
"loss": 0.3268,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.33,
|
| 42 |
+
"learning_rate": 4.938271604938271e-05,
|
| 43 |
+
"loss": 0.3087,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.39,
|
| 48 |
+
"learning_rate": 4.835390946502058e-05,
|
| 49 |
+
"loss": 0.2053,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.44,
|
| 54 |
+
"learning_rate": 4.732510288065844e-05,
|
| 55 |
+
"loss": 0.1831,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.5,
|
| 60 |
+
"learning_rate": 4.62962962962963e-05,
|
| 61 |
+
"loss": 0.1992,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.56,
|
| 66 |
+
"learning_rate": 4.5267489711934157e-05,
|
| 67 |
+
"loss": 0.2218,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.61,
|
| 72 |
+
"learning_rate": 4.423868312757202e-05,
|
| 73 |
+
"loss": 0.2151,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.67,
|
| 78 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 79 |
+
"loss": 0.1468,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.72,
|
| 84 |
+
"learning_rate": 4.2181069958847744e-05,
|
| 85 |
+
"loss": 0.1526,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.78,
|
| 90 |
+
"learning_rate": 4.11522633744856e-05,
|
| 91 |
+
"loss": 0.1269,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.83,
|
| 96 |
+
"learning_rate": 4.012345679012346e-05,
|
| 97 |
+
"loss": 0.1384,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.89,
|
| 102 |
+
"learning_rate": 3.909465020576132e-05,
|
| 103 |
+
"loss": 0.127,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.94,
|
| 108 |
+
"learning_rate": 3.806584362139918e-05,
|
| 109 |
+
"loss": 0.1184,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.0,
|
| 114 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 115 |
+
"loss": 0.1234,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
{
|
| 119 |
"epoch": 1.0,
|
| 120 |
+
"eval_accuracy": 0.9839718530101642,
|
| 121 |
+
"eval_f1": 0.9839718530101642,
|
| 122 |
+
"eval_loss": 0.045045629143714905,
|
| 123 |
+
"eval_precision": 0.9839718530101642,
|
| 124 |
+
"eval_recall": 0.9839718530101642,
|
| 125 |
+
"eval_runtime": 439.4532,
|
| 126 |
+
"eval_samples_per_second": 5.821,
|
| 127 |
+
"eval_steps_per_second": 0.182,
|
| 128 |
+
"step": 180
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.06,
|
| 132 |
+
"learning_rate": 3.60082304526749e-05,
|
| 133 |
+
"loss": 0.1162,
|
| 134 |
+
"step": 190
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.11,
|
| 138 |
+
"learning_rate": 3.497942386831276e-05,
|
| 139 |
+
"loss": 0.1137,
|
| 140 |
+
"step": 200
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.17,
|
| 144 |
+
"learning_rate": 3.395061728395062e-05,
|
| 145 |
+
"loss": 0.1066,
|
| 146 |
+
"step": 210
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.22,
|
| 150 |
+
"learning_rate": 3.292181069958848e-05,
|
| 151 |
+
"loss": 0.0998,
|
| 152 |
+
"step": 220
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.28,
|
| 156 |
+
"learning_rate": 3.1893004115226336e-05,
|
| 157 |
+
"loss": 0.0899,
|
| 158 |
+
"step": 230
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.33,
|
| 162 |
+
"learning_rate": 3.08641975308642e-05,
|
| 163 |
+
"loss": 0.0929,
|
| 164 |
+
"step": 240
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.39,
|
| 168 |
+
"learning_rate": 2.9835390946502057e-05,
|
| 169 |
+
"loss": 0.1265,
|
| 170 |
+
"step": 250
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.44,
|
| 174 |
+
"learning_rate": 2.880658436213992e-05,
|
| 175 |
+
"loss": 0.1177,
|
| 176 |
+
"step": 260
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.5,
|
| 180 |
+
"learning_rate": 2.777777777777778e-05,
|
| 181 |
+
"loss": 0.0866,
|
| 182 |
+
"step": 270
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.56,
|
| 186 |
+
"learning_rate": 2.6748971193415638e-05,
|
| 187 |
+
"loss": 0.1129,
|
| 188 |
+
"step": 280
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.61,
|
| 192 |
+
"learning_rate": 2.5720164609053497e-05,
|
| 193 |
+
"loss": 0.087,
|
| 194 |
+
"step": 290
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.67,
|
| 198 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 199 |
+
"loss": 0.103,
|
| 200 |
+
"step": 300
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.72,
|
| 204 |
+
"learning_rate": 2.366255144032922e-05,
|
| 205 |
+
"loss": 0.0784,
|
| 206 |
+
"step": 310
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.78,
|
| 210 |
+
"learning_rate": 2.2633744855967078e-05,
|
| 211 |
+
"loss": 0.0833,
|
| 212 |
+
"step": 320
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.83,
|
| 216 |
+
"learning_rate": 2.1604938271604937e-05,
|
| 217 |
+
"loss": 0.0847,
|
| 218 |
+
"step": 330
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.89,
|
| 222 |
+
"learning_rate": 2.05761316872428e-05,
|
| 223 |
+
"loss": 0.0738,
|
| 224 |
+
"step": 340
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.94,
|
| 228 |
+
"learning_rate": 1.954732510288066e-05,
|
| 229 |
+
"loss": 0.0844,
|
| 230 |
+
"step": 350
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.0,
|
| 234 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 235 |
+
"loss": 0.0837,
|
| 236 |
+
"step": 360
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 2.0,
|
| 240 |
+
"eval_accuracy": 0.9925723221266615,
|
| 241 |
+
"eval_f1": 0.9925723221266615,
|
| 242 |
+
"eval_loss": 0.019814323633909225,
|
| 243 |
+
"eval_precision": 0.9925723221266615,
|
| 244 |
+
"eval_recall": 0.9925723221266615,
|
| 245 |
+
"eval_runtime": 440.2071,
|
| 246 |
+
"eval_samples_per_second": 5.811,
|
| 247 |
+
"eval_steps_per_second": 0.182,
|
| 248 |
+
"step": 360
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.06,
|
| 252 |
+
"learning_rate": 1.748971193415638e-05,
|
| 253 |
+
"loss": 0.0742,
|
| 254 |
+
"step": 370
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.11,
|
| 258 |
+
"learning_rate": 1.646090534979424e-05,
|
| 259 |
+
"loss": 0.0805,
|
| 260 |
+
"step": 380
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.17,
|
| 264 |
+
"learning_rate": 1.54320987654321e-05,
|
| 265 |
+
"loss": 0.0892,
|
| 266 |
+
"step": 390
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.22,
|
| 270 |
+
"learning_rate": 1.440329218106996e-05,
|
| 271 |
+
"loss": 0.0732,
|
| 272 |
+
"step": 400
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.28,
|
| 276 |
+
"learning_rate": 1.3374485596707819e-05,
|
| 277 |
+
"loss": 0.0823,
|
| 278 |
+
"step": 410
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.33,
|
| 282 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 283 |
+
"loss": 0.0623,
|
| 284 |
+
"step": 420
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.39,
|
| 288 |
+
"learning_rate": 1.1316872427983539e-05,
|
| 289 |
+
"loss": 0.0609,
|
| 290 |
+
"step": 430
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.44,
|
| 294 |
+
"learning_rate": 1.02880658436214e-05,
|
| 295 |
+
"loss": 0.0752,
|
| 296 |
+
"step": 440
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.5,
|
| 300 |
+
"learning_rate": 9.259259259259259e-06,
|
| 301 |
+
"loss": 0.0608,
|
| 302 |
+
"step": 450
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.56,
|
| 306 |
+
"learning_rate": 8.23045267489712e-06,
|
| 307 |
+
"loss": 0.0529,
|
| 308 |
+
"step": 460
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.61,
|
| 312 |
+
"learning_rate": 7.20164609053498e-06,
|
| 313 |
+
"loss": 0.0628,
|
| 314 |
+
"step": 470
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.67,
|
| 318 |
+
"learning_rate": 6.172839506172839e-06,
|
| 319 |
+
"loss": 0.0508,
|
| 320 |
+
"step": 480
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.72,
|
| 324 |
+
"learning_rate": 5.1440329218107e-06,
|
| 325 |
+
"loss": 0.0517,
|
| 326 |
+
"step": 490
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.78,
|
| 330 |
+
"learning_rate": 4.11522633744856e-06,
|
| 331 |
+
"loss": 0.06,
|
| 332 |
+
"step": 500
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.83,
|
| 336 |
+
"learning_rate": 3.0864197530864196e-06,
|
| 337 |
+
"loss": 0.0671,
|
| 338 |
+
"step": 510
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.89,
|
| 342 |
+
"learning_rate": 2.05761316872428e-06,
|
| 343 |
+
"loss": 0.0733,
|
| 344 |
+
"step": 520
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.94,
|
| 348 |
+
"learning_rate": 1.02880658436214e-06,
|
| 349 |
+
"loss": 0.0558,
|
| 350 |
+
"step": 530
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 3.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
"learning_rate": 0.0,
|
| 355 |
+
"loss": 0.0373,
|
| 356 |
+
"step": 540
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 3.0,
|
| 360 |
+
"eval_accuracy": 0.9960906958561376,
|
| 361 |
+
"eval_f1": 0.9960906958561376,
|
| 362 |
+
"eval_loss": 0.012254622764885426,
|
| 363 |
+
"eval_precision": 0.9960906958561376,
|
| 364 |
+
"eval_recall": 0.9960906958561376,
|
| 365 |
+
"eval_runtime": 442.7035,
|
| 366 |
+
"eval_samples_per_second": 5.778,
|
| 367 |
+
"eval_steps_per_second": 0.181,
|
| 368 |
+
"step": 540
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 3.0,
|
| 372 |
+
"step": 540,
|
| 373 |
+
"total_flos": 5.409667845997019e+18,
|
| 374 |
+
"train_loss": 0.15731672859854168,
|
| 375 |
+
"train_runtime": 52202.335,
|
| 376 |
+
"train_samples_per_second": 1.323,
|
| 377 |
+
"train_steps_per_second": 0.01
|
| 378 |
}
|
| 379 |
],
|
| 380 |
+
"max_steps": 540,
|
| 381 |
+
"num_train_epochs": 3,
|
| 382 |
+
"total_flos": 5.409667845997019e+18,
|
|
|
|
|
|
|
| 383 |
"trial_name": null,
|
| 384 |
"trial_params": null
|
| 385 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab8af6b29eb404087346089460439c73ea21969d537e5cddcb8cdf375b09ff52
|
| 3 |
+
size 3451
|