Commit
•
40095a9
1
Parent(s):
e1acb96
Upload 9 files
Browse files- config.json +36 -0
- optimizer.pt +3 -0
- preprocessor_config.json +22 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +616 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "adenocarcinoma_left.lower.lobe_T2_N0_M0_Ib",
|
| 13 |
+
"1": "large.cell.carcinoma_left.hilum_T2_N2_M0_IIIa",
|
| 14 |
+
"2": "normal",
|
| 15 |
+
"3": "squamous.cell.carcinoma_left.hilum_T1_N2_M0_IIIa"
|
| 16 |
+
},
|
| 17 |
+
"image_size": 224,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"adenocarcinoma_left.lower.lobe_T2_N0_M0_Ib": "0",
|
| 22 |
+
"large.cell.carcinoma_left.hilum_T2_N2_M0_IIIa": "1",
|
| 23 |
+
"normal": "2",
|
| 24 |
+
"squamous.cell.carcinoma_left.hilum_T1_N2_M0_IIIa": "3"
|
| 25 |
+
},
|
| 26 |
+
"layer_norm_eps": 1e-12,
|
| 27 |
+
"model_type": "vit",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_channels": 3,
|
| 30 |
+
"num_hidden_layers": 12,
|
| 31 |
+
"patch_size": 16,
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"qkv_bias": true,
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.26.1"
|
| 36 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd5920de0a177b07700e30f0293dd88ee9784d3f0cd3190805097d7a47eea3aa
|
| 3 |
+
size 686531205
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f445cac5ea8bb06e72af02526a56bc6b5dc82560fc2d28b2c79623f3812f4b38
|
| 3 |
+
size 343274861
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:90e6f0eeba8b9894240a6cd020cfd0c5f4fb28efecdfe8ced27fec5df5f9c79a
|
| 3 |
+
size 14575
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eedd043df6cf348e21712f6001f51ee5b59434655346c4271bdeabe13333a8a9
|
| 3 |
+
size 557
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1dbf481eba4f2d73b9de4b77467dbd66e6e88c22703fc527864f8e746df4392
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,616 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9590163934426229,
|
| 3 |
+
"best_model_checkpoint": "./vit-LungCancerViT/checkpoint-350",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 700,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.29,
|
| 12 |
+
"learning_rate": 2.857142857142857e-05,
|
| 13 |
+
"loss": 1.3951,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.57,
|
| 18 |
+
"learning_rate": 5.714285714285714e-05,
|
| 19 |
+
"loss": 0.9527,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.86,
|
| 24 |
+
"learning_rate": 8.571428571428571e-05,
|
| 25 |
+
"loss": 0.7463,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"eval_accuracy": 0.7213114754098361,
|
| 31 |
+
"eval_loss": 0.7026457190513611,
|
| 32 |
+
"eval_runtime": 52.9268,
|
| 33 |
+
"eval_samples_per_second": 2.305,
|
| 34 |
+
"eval_steps_per_second": 0.302,
|
| 35 |
+
"step": 35
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.14,
|
| 39 |
+
"learning_rate": 0.00011428571428571428,
|
| 40 |
+
"loss": 0.5309,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.43,
|
| 45 |
+
"learning_rate": 0.00014285714285714287,
|
| 46 |
+
"loss": 0.4553,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.71,
|
| 51 |
+
"learning_rate": 0.00017142857142857143,
|
| 52 |
+
"loss": 0.3735,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.0,
|
| 57 |
+
"learning_rate": 0.0002,
|
| 58 |
+
"loss": 0.2276,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_accuracy": 0.8360655737704918,
|
| 64 |
+
"eval_loss": 0.3694429099559784,
|
| 65 |
+
"eval_runtime": 1.4376,
|
| 66 |
+
"eval_samples_per_second": 84.864,
|
| 67 |
+
"eval_steps_per_second": 11.13,
|
| 68 |
+
"step": 70
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.29,
|
| 72 |
+
"learning_rate": 0.00019682539682539682,
|
| 73 |
+
"loss": 0.321,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.57,
|
| 78 |
+
"learning_rate": 0.00019365079365079365,
|
| 79 |
+
"loss": 0.1708,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.86,
|
| 84 |
+
"learning_rate": 0.00019047619047619048,
|
| 85 |
+
"loss": 0.3134,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.0,
|
| 90 |
+
"eval_accuracy": 0.860655737704918,
|
| 91 |
+
"eval_loss": 0.3415001332759857,
|
| 92 |
+
"eval_runtime": 1.2015,
|
| 93 |
+
"eval_samples_per_second": 101.542,
|
| 94 |
+
"eval_steps_per_second": 13.317,
|
| 95 |
+
"step": 105
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 3.14,
|
| 99 |
+
"learning_rate": 0.00018730158730158731,
|
| 100 |
+
"loss": 0.1296,
|
| 101 |
+
"step": 110
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.43,
|
| 105 |
+
"learning_rate": 0.00018412698412698412,
|
| 106 |
+
"loss": 0.1161,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.71,
|
| 111 |
+
"learning_rate": 0.00018095238095238095,
|
| 112 |
+
"loss": 0.0696,
|
| 113 |
+
"step": 130
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.0,
|
| 117 |
+
"learning_rate": 0.00017777777777777779,
|
| 118 |
+
"loss": 0.1487,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 4.0,
|
| 123 |
+
"eval_accuracy": 0.860655737704918,
|
| 124 |
+
"eval_loss": 0.4990597069263458,
|
| 125 |
+
"eval_runtime": 1.1086,
|
| 126 |
+
"eval_samples_per_second": 110.045,
|
| 127 |
+
"eval_steps_per_second": 14.432,
|
| 128 |
+
"step": 140
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.29,
|
| 132 |
+
"learning_rate": 0.00017460317460317462,
|
| 133 |
+
"loss": 0.0783,
|
| 134 |
+
"step": 150
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.57,
|
| 138 |
+
"learning_rate": 0.00017142857142857143,
|
| 139 |
+
"loss": 0.1082,
|
| 140 |
+
"step": 160
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.86,
|
| 144 |
+
"learning_rate": 0.00016825396825396826,
|
| 145 |
+
"loss": 0.0301,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.0,
|
| 150 |
+
"eval_accuracy": 0.8934426229508197,
|
| 151 |
+
"eval_loss": 0.2610397934913635,
|
| 152 |
+
"eval_runtime": 1.1421,
|
| 153 |
+
"eval_samples_per_second": 106.821,
|
| 154 |
+
"eval_steps_per_second": 14.009,
|
| 155 |
+
"step": 175
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 5.14,
|
| 159 |
+
"learning_rate": 0.0001650793650793651,
|
| 160 |
+
"loss": 0.0994,
|
| 161 |
+
"step": 180
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.43,
|
| 165 |
+
"learning_rate": 0.00016190476190476192,
|
| 166 |
+
"loss": 0.1404,
|
| 167 |
+
"step": 190
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 5.71,
|
| 171 |
+
"learning_rate": 0.00015873015873015873,
|
| 172 |
+
"loss": 0.0216,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 6.0,
|
| 177 |
+
"learning_rate": 0.00015555555555555556,
|
| 178 |
+
"loss": 0.0567,
|
| 179 |
+
"step": 210
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 6.0,
|
| 183 |
+
"eval_accuracy": 0.9098360655737705,
|
| 184 |
+
"eval_loss": 0.2997801601886749,
|
| 185 |
+
"eval_runtime": 1.3611,
|
| 186 |
+
"eval_samples_per_second": 89.633,
|
| 187 |
+
"eval_steps_per_second": 11.755,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 6.29,
|
| 192 |
+
"learning_rate": 0.00015238095238095237,
|
| 193 |
+
"loss": 0.0529,
|
| 194 |
+
"step": 220
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 6.57,
|
| 198 |
+
"learning_rate": 0.00014920634920634923,
|
| 199 |
+
"loss": 0.0305,
|
| 200 |
+
"step": 230
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 6.86,
|
| 204 |
+
"learning_rate": 0.00014603174603174603,
|
| 205 |
+
"loss": 0.0084,
|
| 206 |
+
"step": 240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.0,
|
| 210 |
+
"eval_accuracy": 0.9016393442622951,
|
| 211 |
+
"eval_loss": 0.4632587730884552,
|
| 212 |
+
"eval_runtime": 1.1837,
|
| 213 |
+
"eval_samples_per_second": 103.068,
|
| 214 |
+
"eval_steps_per_second": 13.517,
|
| 215 |
+
"step": 245
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 7.14,
|
| 219 |
+
"learning_rate": 0.00014285714285714287,
|
| 220 |
+
"loss": 0.0174,
|
| 221 |
+
"step": 250
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 7.43,
|
| 225 |
+
"learning_rate": 0.00013968253968253967,
|
| 226 |
+
"loss": 0.0027,
|
| 227 |
+
"step": 260
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 7.71,
|
| 231 |
+
"learning_rate": 0.0001365079365079365,
|
| 232 |
+
"loss": 0.0245,
|
| 233 |
+
"step": 270
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 8.0,
|
| 237 |
+
"learning_rate": 0.00013333333333333334,
|
| 238 |
+
"loss": 0.0071,
|
| 239 |
+
"step": 280
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 8.0,
|
| 243 |
+
"eval_accuracy": 0.9508196721311475,
|
| 244 |
+
"eval_loss": 0.21911580860614777,
|
| 245 |
+
"eval_runtime": 1.1253,
|
| 246 |
+
"eval_samples_per_second": 108.412,
|
| 247 |
+
"eval_steps_per_second": 14.218,
|
| 248 |
+
"step": 280
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.29,
|
| 252 |
+
"learning_rate": 0.00013015873015873017,
|
| 253 |
+
"loss": 0.0169,
|
| 254 |
+
"step": 290
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 8.57,
|
| 258 |
+
"learning_rate": 0.00012698412698412698,
|
| 259 |
+
"loss": 0.014,
|
| 260 |
+
"step": 300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 8.86,
|
| 264 |
+
"learning_rate": 0.0001238095238095238,
|
| 265 |
+
"loss": 0.0231,
|
| 266 |
+
"step": 310
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 9.0,
|
| 270 |
+
"eval_accuracy": 0.9344262295081968,
|
| 271 |
+
"eval_loss": 0.3097061216831207,
|
| 272 |
+
"eval_runtime": 1.1316,
|
| 273 |
+
"eval_samples_per_second": 107.81,
|
| 274 |
+
"eval_steps_per_second": 14.139,
|
| 275 |
+
"step": 315
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 9.14,
|
| 279 |
+
"learning_rate": 0.00012063492063492063,
|
| 280 |
+
"loss": 0.0003,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 9.43,
|
| 285 |
+
"learning_rate": 0.00011746031746031746,
|
| 286 |
+
"loss": 0.0195,
|
| 287 |
+
"step": 330
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 9.71,
|
| 291 |
+
"learning_rate": 0.00011428571428571428,
|
| 292 |
+
"loss": 0.0062,
|
| 293 |
+
"step": 340
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 10.0,
|
| 297 |
+
"learning_rate": 0.00011111111111111112,
|
| 298 |
+
"loss": 0.0003,
|
| 299 |
+
"step": 350
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.0,
|
| 303 |
+
"eval_accuracy": 0.9590163934426229,
|
| 304 |
+
"eval_loss": 0.2627410292625427,
|
| 305 |
+
"eval_runtime": 1.345,
|
| 306 |
+
"eval_samples_per_second": 90.707,
|
| 307 |
+
"eval_steps_per_second": 11.896,
|
| 308 |
+
"step": 350
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 10.29,
|
| 312 |
+
"learning_rate": 0.00010793650793650794,
|
| 313 |
+
"loss": 0.0004,
|
| 314 |
+
"step": 360
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 10.57,
|
| 318 |
+
"learning_rate": 0.00010476190476190477,
|
| 319 |
+
"loss": 0.003,
|
| 320 |
+
"step": 370
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 10.86,
|
| 324 |
+
"learning_rate": 0.00010158730158730159,
|
| 325 |
+
"loss": 0.0002,
|
| 326 |
+
"step": 380
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 11.0,
|
| 330 |
+
"eval_accuracy": 0.9508196721311475,
|
| 331 |
+
"eval_loss": 0.284305602312088,
|
| 332 |
+
"eval_runtime": 1.3379,
|
| 333 |
+
"eval_samples_per_second": 91.19,
|
| 334 |
+
"eval_steps_per_second": 11.959,
|
| 335 |
+
"step": 385
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 11.14,
|
| 339 |
+
"learning_rate": 9.841269841269841e-05,
|
| 340 |
+
"loss": 0.0051,
|
| 341 |
+
"step": 390
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 11.43,
|
| 345 |
+
"learning_rate": 9.523809523809524e-05,
|
| 346 |
+
"loss": 0.0001,
|
| 347 |
+
"step": 400
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 11.71,
|
| 351 |
+
"learning_rate": 9.206349206349206e-05,
|
| 352 |
+
"loss": 0.0016,
|
| 353 |
+
"step": 410
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 12.0,
|
| 357 |
+
"learning_rate": 8.888888888888889e-05,
|
| 358 |
+
"loss": 0.005,
|
| 359 |
+
"step": 420
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 12.0,
|
| 363 |
+
"eval_accuracy": 0.9508196721311475,
|
| 364 |
+
"eval_loss": 0.28199827671051025,
|
| 365 |
+
"eval_runtime": 1.1031,
|
| 366 |
+
"eval_samples_per_second": 110.597,
|
| 367 |
+
"eval_steps_per_second": 14.505,
|
| 368 |
+
"step": 420
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 12.29,
|
| 372 |
+
"learning_rate": 8.571428571428571e-05,
|
| 373 |
+
"loss": 0.0001,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 12.57,
|
| 378 |
+
"learning_rate": 8.253968253968255e-05,
|
| 379 |
+
"loss": 0.0001,
|
| 380 |
+
"step": 440
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 12.86,
|
| 384 |
+
"learning_rate": 7.936507936507937e-05,
|
| 385 |
+
"loss": 0.0054,
|
| 386 |
+
"step": 450
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 13.0,
|
| 390 |
+
"eval_accuracy": 0.9508196721311475,
|
| 391 |
+
"eval_loss": 0.2708677649497986,
|
| 392 |
+
"eval_runtime": 1.117,
|
| 393 |
+
"eval_samples_per_second": 109.217,
|
| 394 |
+
"eval_steps_per_second": 14.324,
|
| 395 |
+
"step": 455
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 13.14,
|
| 399 |
+
"learning_rate": 7.619047619047618e-05,
|
| 400 |
+
"loss": 0.0001,
|
| 401 |
+
"step": 460
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 13.43,
|
| 405 |
+
"learning_rate": 7.301587301587302e-05,
|
| 406 |
+
"loss": 0.0001,
|
| 407 |
+
"step": 470
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 13.71,
|
| 411 |
+
"learning_rate": 6.984126984126984e-05,
|
| 412 |
+
"loss": 0.0027,
|
| 413 |
+
"step": 480
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 14.0,
|
| 417 |
+
"learning_rate": 6.666666666666667e-05,
|
| 418 |
+
"loss": 0.0026,
|
| 419 |
+
"step": 490
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 14.0,
|
| 423 |
+
"eval_accuracy": 0.9508196721311475,
|
| 424 |
+
"eval_loss": 0.2664688229560852,
|
| 425 |
+
"eval_runtime": 1.2394,
|
| 426 |
+
"eval_samples_per_second": 98.438,
|
| 427 |
+
"eval_steps_per_second": 12.91,
|
| 428 |
+
"step": 490
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 14.29,
|
| 432 |
+
"learning_rate": 6.349206349206349e-05,
|
| 433 |
+
"loss": 0.0023,
|
| 434 |
+
"step": 500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 14.57,
|
| 438 |
+
"learning_rate": 6.0317460317460316e-05,
|
| 439 |
+
"loss": 0.0038,
|
| 440 |
+
"step": 510
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 14.86,
|
| 444 |
+
"learning_rate": 5.714285714285714e-05,
|
| 445 |
+
"loss": 0.0001,
|
| 446 |
+
"step": 520
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 15.0,
|
| 450 |
+
"eval_accuracy": 0.9590163934426229,
|
| 451 |
+
"eval_loss": 0.2634933888912201,
|
| 452 |
+
"eval_runtime": 1.3242,
|
| 453 |
+
"eval_samples_per_second": 92.133,
|
| 454 |
+
"eval_steps_per_second": 12.083,
|
| 455 |
+
"step": 525
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 15.14,
|
| 459 |
+
"learning_rate": 5.396825396825397e-05,
|
| 460 |
+
"loss": 0.0023,
|
| 461 |
+
"step": 530
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 15.43,
|
| 465 |
+
"learning_rate": 5.0793650793650794e-05,
|
| 466 |
+
"loss": 0.0028,
|
| 467 |
+
"step": 540
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 15.71,
|
| 471 |
+
"learning_rate": 4.761904761904762e-05,
|
| 472 |
+
"loss": 0.0001,
|
| 473 |
+
"step": 550
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 16.0,
|
| 477 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 478 |
+
"loss": 0.0001,
|
| 479 |
+
"step": 560
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 16.0,
|
| 483 |
+
"eval_accuracy": 0.9590163934426229,
|
| 484 |
+
"eval_loss": 0.26172590255737305,
|
| 485 |
+
"eval_runtime": 1.134,
|
| 486 |
+
"eval_samples_per_second": 107.579,
|
| 487 |
+
"eval_steps_per_second": 14.109,
|
| 488 |
+
"step": 560
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 16.29,
|
| 492 |
+
"learning_rate": 4.126984126984127e-05,
|
| 493 |
+
"loss": 0.0025,
|
| 494 |
+
"step": 570
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 16.57,
|
| 498 |
+
"learning_rate": 3.809523809523809e-05,
|
| 499 |
+
"loss": 0.0026,
|
| 500 |
+
"step": 580
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 16.86,
|
| 504 |
+
"learning_rate": 3.492063492063492e-05,
|
| 505 |
+
"loss": 0.0001,
|
| 506 |
+
"step": 590
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 17.0,
|
| 510 |
+
"eval_accuracy": 0.9590163934426229,
|
| 511 |
+
"eval_loss": 0.26094985008239746,
|
| 512 |
+
"eval_runtime": 1.1056,
|
| 513 |
+
"eval_samples_per_second": 110.343,
|
| 514 |
+
"eval_steps_per_second": 14.471,
|
| 515 |
+
"step": 595
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 17.14,
|
| 519 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 520 |
+
"loss": 0.0001,
|
| 521 |
+
"step": 600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 17.43,
|
| 525 |
+
"learning_rate": 2.857142857142857e-05,
|
| 526 |
+
"loss": 0.0025,
|
| 527 |
+
"step": 610
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 17.71,
|
| 531 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 532 |
+
"loss": 0.0001,
|
| 533 |
+
"step": 620
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 18.0,
|
| 537 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 538 |
+
"loss": 0.0026,
|
| 539 |
+
"step": 630
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 18.0,
|
| 543 |
+
"eval_accuracy": 0.9590163934426229,
|
| 544 |
+
"eval_loss": 0.2608335614204407,
|
| 545 |
+
"eval_runtime": 1.0804,
|
| 546 |
+
"eval_samples_per_second": 112.918,
|
| 547 |
+
"eval_steps_per_second": 14.809,
|
| 548 |
+
"step": 630
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 18.29,
|
| 552 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 553 |
+
"loss": 0.0023,
|
| 554 |
+
"step": 640
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 18.57,
|
| 558 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 559 |
+
"loss": 0.0001,
|
| 560 |
+
"step": 650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 18.86,
|
| 564 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 565 |
+
"loss": 0.0028,
|
| 566 |
+
"step": 660
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 19.0,
|
| 570 |
+
"eval_accuracy": 0.9590163934426229,
|
| 571 |
+
"eval_loss": 0.2601701617240906,
|
| 572 |
+
"eval_runtime": 1.2677,
|
| 573 |
+
"eval_samples_per_second": 96.24,
|
| 574 |
+
"eval_steps_per_second": 12.622,
|
| 575 |
+
"step": 665
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 19.14,
|
| 579 |
+
"learning_rate": 9.523809523809523e-06,
|
| 580 |
+
"loss": 0.0001,
|
| 581 |
+
"step": 670
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 19.43,
|
| 585 |
+
"learning_rate": 6.349206349206349e-06,
|
| 586 |
+
"loss": 0.0026,
|
| 587 |
+
"step": 680
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 19.71,
|
| 591 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 592 |
+
"loss": 0.0001,
|
| 593 |
+
"step": 690
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 20.0,
|
| 597 |
+
"learning_rate": 0.0,
|
| 598 |
+
"loss": 0.0022,
|
| 599 |
+
"step": 700
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 20.0,
|
| 603 |
+
"eval_accuracy": 0.9590163934426229,
|
| 604 |
+
"eval_loss": 0.2600506544113159,
|
| 605 |
+
"eval_runtime": 1.1031,
|
| 606 |
+
"eval_samples_per_second": 110.597,
|
| 607 |
+
"eval_steps_per_second": 14.504,
|
| 608 |
+
"step": 700
|
| 609 |
+
}
|
| 610 |
+
],
|
| 611 |
+
"max_steps": 700,
|
| 612 |
+
"num_train_epochs": 20,
|
| 613 |
+
"total_flos": 1.6924553905235558e+18,
|
| 614 |
+
"trial_name": null,
|
| 615 |
+
"trial_params": null
|
| 616 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c77464eaea89e575ceac1fe7045582109dd5ff51ceb9b4381179e8b068eea8f9
|
| 3 |
+
size 3515
|