Upload 8 files
Browse files- config.json +61 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +22 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1399 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/swin-base-patch4-window7-224-in22k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 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": "healthy",
|
| 21 |
+
"1": "parkinson"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"healthy": 0,
|
| 27 |
+
"parkinson": 1
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"mlp_ratio": 4.0,
|
| 31 |
+
"model_type": "swin",
|
| 32 |
+
"num_channels": 3,
|
| 33 |
+
"num_heads": [
|
| 34 |
+
4,
|
| 35 |
+
8,
|
| 36 |
+
16,
|
| 37 |
+
32
|
| 38 |
+
],
|
| 39 |
+
"num_layers": 4,
|
| 40 |
+
"out_features": [
|
| 41 |
+
"stage4"
|
| 42 |
+
],
|
| 43 |
+
"out_indices": [
|
| 44 |
+
4
|
| 45 |
+
],
|
| 46 |
+
"patch_size": 4,
|
| 47 |
+
"path_norm": true,
|
| 48 |
+
"problem_type": "single_label_classification",
|
| 49 |
+
"qkv_bias": true,
|
| 50 |
+
"stage_names": [
|
| 51 |
+
"stem",
|
| 52 |
+
"stage1",
|
| 53 |
+
"stage2",
|
| 54 |
+
"stage3",
|
| 55 |
+
"stage4"
|
| 56 |
+
],
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.35.2",
|
| 59 |
+
"use_absolute_embeddings": false,
|
| 60 |
+
"window_size": 7
|
| 61 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc7ef455b8e803897db390fd97567a4ed977d7ea083d00eb07fd8802be877f2a
|
| 3 |
+
size 347498816
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44377e7248f4e67dff8538c9f20790813c480baf9168156ce67c0932e013c0f5
|
| 3 |
+
size 694317645
|
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 |
+
0.485,
|
| 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 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06f031f5270dbad7eb3b2e2354f13d6429f8510d692e24c1e33cacef94887c30
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb9311f5b96fda8aa83e66d73e74f4201f18e813a34360e4da010cd2c4bfeab2
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1399 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "./swin-soiral/checkpoint-72",
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 400,
|
| 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": 0.5,
|
| 14 |
+
"eval_loss": 0.7685546875,
|
| 15 |
+
"eval_runtime": 0.2552,
|
| 16 |
+
"eval_samples_per_second": 47.013,
|
| 17 |
+
"eval_steps_per_second": 7.835,
|
| 18 |
+
"step": 4
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.25,
|
| 22 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 23 |
+
"loss": 0.7776,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.0,
|
| 28 |
+
"eval_accuracy": 0.5,
|
| 29 |
+
"eval_loss": 0.7459920048713684,
|
| 30 |
+
"eval_runtime": 0.2163,
|
| 31 |
+
"eval_samples_per_second": 55.485,
|
| 32 |
+
"eval_steps_per_second": 9.247,
|
| 33 |
+
"step": 8
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.5,
|
| 37 |
+
"learning_rate": 2.25e-06,
|
| 38 |
+
"loss": 0.7839,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 3.0,
|
| 43 |
+
"eval_accuracy": 0.5,
|
| 44 |
+
"eval_loss": 0.71142578125,
|
| 45 |
+
"eval_runtime": 0.2223,
|
| 46 |
+
"eval_samples_per_second": 53.982,
|
| 47 |
+
"eval_steps_per_second": 8.997,
|
| 48 |
+
"step": 12
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.75,
|
| 52 |
+
"learning_rate": 3.5e-06,
|
| 53 |
+
"loss": 0.6965,
|
| 54 |
+
"step": 15
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 4.0,
|
| 58 |
+
"eval_accuracy": 0.6666666666666666,
|
| 59 |
+
"eval_loss": 0.6708170771598816,
|
| 60 |
+
"eval_runtime": 0.1987,
|
| 61 |
+
"eval_samples_per_second": 60.404,
|
| 62 |
+
"eval_steps_per_second": 10.067,
|
| 63 |
+
"step": 16
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 5.0,
|
| 67 |
+
"learning_rate": 4.75e-06,
|
| 68 |
+
"loss": 0.6778,
|
| 69 |
+
"step": 20
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 5.0,
|
| 73 |
+
"eval_accuracy": 0.6666666666666666,
|
| 74 |
+
"eval_loss": 0.62982177734375,
|
| 75 |
+
"eval_runtime": 0.1998,
|
| 76 |
+
"eval_samples_per_second": 60.045,
|
| 77 |
+
"eval_steps_per_second": 10.008,
|
| 78 |
+
"step": 20
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 6.0,
|
| 82 |
+
"eval_accuracy": 0.6666666666666666,
|
| 83 |
+
"eval_loss": 0.5903117060661316,
|
| 84 |
+
"eval_runtime": 0.2941,
|
| 85 |
+
"eval_samples_per_second": 40.798,
|
| 86 |
+
"eval_steps_per_second": 6.8,
|
| 87 |
+
"step": 24
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 6.25,
|
| 91 |
+
"learning_rate": 6e-06,
|
| 92 |
+
"loss": 0.6297,
|
| 93 |
+
"step": 25
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 7.0,
|
| 97 |
+
"eval_accuracy": 0.75,
|
| 98 |
+
"eval_loss": 0.5383097529411316,
|
| 99 |
+
"eval_runtime": 0.2066,
|
| 100 |
+
"eval_samples_per_second": 58.089,
|
| 101 |
+
"eval_steps_per_second": 9.682,
|
| 102 |
+
"step": 28
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 7.5,
|
| 106 |
+
"learning_rate": 7.25e-06,
|
| 107 |
+
"loss": 0.5812,
|
| 108 |
+
"step": 30
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 8.0,
|
| 112 |
+
"eval_accuracy": 0.75,
|
| 113 |
+
"eval_loss": 0.4876708984375,
|
| 114 |
+
"eval_runtime": 0.2154,
|
| 115 |
+
"eval_samples_per_second": 55.709,
|
| 116 |
+
"eval_steps_per_second": 9.285,
|
| 117 |
+
"step": 32
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 8.75,
|
| 121 |
+
"learning_rate": 8.5e-06,
|
| 122 |
+
"loss": 0.5141,
|
| 123 |
+
"step": 35
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 9.0,
|
| 127 |
+
"eval_accuracy": 0.9166666666666666,
|
| 128 |
+
"eval_loss": 0.42340087890625,
|
| 129 |
+
"eval_runtime": 0.2045,
|
| 130 |
+
"eval_samples_per_second": 58.685,
|
| 131 |
+
"eval_steps_per_second": 9.781,
|
| 132 |
+
"step": 36
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 10.0,
|
| 136 |
+
"learning_rate": 9.75e-06,
|
| 137 |
+
"loss": 0.4186,
|
| 138 |
+
"step": 40
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 10.0,
|
| 142 |
+
"eval_accuracy": 0.9166666666666666,
|
| 143 |
+
"eval_loss": 0.3495737612247467,
|
| 144 |
+
"eval_runtime": 0.2143,
|
| 145 |
+
"eval_samples_per_second": 55.994,
|
| 146 |
+
"eval_steps_per_second": 9.332,
|
| 147 |
+
"step": 40
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 11.0,
|
| 151 |
+
"eval_accuracy": 0.8333333333333334,
|
| 152 |
+
"eval_loss": 0.2956085205078125,
|
| 153 |
+
"eval_runtime": 0.204,
|
| 154 |
+
"eval_samples_per_second": 58.832,
|
| 155 |
+
"eval_steps_per_second": 9.805,
|
| 156 |
+
"step": 44
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 11.25,
|
| 160 |
+
"learning_rate": 9.88888888888889e-06,
|
| 161 |
+
"loss": 0.3791,
|
| 162 |
+
"step": 45
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 12.0,
|
| 166 |
+
"eval_accuracy": 0.9166666666666666,
|
| 167 |
+
"eval_loss": 0.26324209570884705,
|
| 168 |
+
"eval_runtime": 0.2276,
|
| 169 |
+
"eval_samples_per_second": 52.73,
|
| 170 |
+
"eval_steps_per_second": 8.788,
|
| 171 |
+
"step": 48
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 12.5,
|
| 175 |
+
"learning_rate": 9.75e-06,
|
| 176 |
+
"loss": 0.2906,
|
| 177 |
+
"step": 50
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 13.0,
|
| 181 |
+
"eval_accuracy": 0.9166666666666666,
|
| 182 |
+
"eval_loss": 0.21923191845417023,
|
| 183 |
+
"eval_runtime": 0.2866,
|
| 184 |
+
"eval_samples_per_second": 41.866,
|
| 185 |
+
"eval_steps_per_second": 6.978,
|
| 186 |
+
"step": 52
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 13.75,
|
| 190 |
+
"learning_rate": 9.611111111111112e-06,
|
| 191 |
+
"loss": 0.2247,
|
| 192 |
+
"step": 55
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 14.0,
|
| 196 |
+
"eval_accuracy": 0.9166666666666666,
|
| 197 |
+
"eval_loss": 0.18271827697753906,
|
| 198 |
+
"eval_runtime": 0.2073,
|
| 199 |
+
"eval_samples_per_second": 57.879,
|
| 200 |
+
"eval_steps_per_second": 9.647,
|
| 201 |
+
"step": 56
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 15.0,
|
| 205 |
+
"learning_rate": 9.472222222222223e-06,
|
| 206 |
+
"loss": 0.1724,
|
| 207 |
+
"step": 60
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.0,
|
| 211 |
+
"eval_accuracy": 0.9166666666666666,
|
| 212 |
+
"eval_loss": 0.1536579132080078,
|
| 213 |
+
"eval_runtime": 0.3096,
|
| 214 |
+
"eval_samples_per_second": 38.763,
|
| 215 |
+
"eval_steps_per_second": 6.461,
|
| 216 |
+
"step": 60
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 16.0,
|
| 220 |
+
"eval_accuracy": 0.9166666666666666,
|
| 221 |
+
"eval_loss": 0.11898771673440933,
|
| 222 |
+
"eval_runtime": 0.207,
|
| 223 |
+
"eval_samples_per_second": 57.977,
|
| 224 |
+
"eval_steps_per_second": 9.663,
|
| 225 |
+
"step": 64
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 16.25,
|
| 229 |
+
"learning_rate": 9.333333333333334e-06,
|
| 230 |
+
"loss": 0.1077,
|
| 231 |
+
"step": 65
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 17.0,
|
| 235 |
+
"eval_accuracy": 0.9166666666666666,
|
| 236 |
+
"eval_loss": 0.12817128002643585,
|
| 237 |
+
"eval_runtime": 0.2101,
|
| 238 |
+
"eval_samples_per_second": 57.105,
|
| 239 |
+
"eval_steps_per_second": 9.517,
|
| 240 |
+
"step": 68
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 17.5,
|
| 244 |
+
"learning_rate": 9.194444444444445e-06,
|
| 245 |
+
"loss": 0.1336,
|
| 246 |
+
"step": 70
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 18.0,
|
| 250 |
+
"eval_accuracy": 1.0,
|
| 251 |
+
"eval_loss": 0.0798807144165039,
|
| 252 |
+
"eval_runtime": 0.3334,
|
| 253 |
+
"eval_samples_per_second": 35.996,
|
| 254 |
+
"eval_steps_per_second": 5.999,
|
| 255 |
+
"step": 72
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 18.75,
|
| 259 |
+
"learning_rate": 9.083333333333333e-06,
|
| 260 |
+
"loss": 0.0656,
|
| 261 |
+
"step": 75
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 19.0,
|
| 265 |
+
"eval_accuracy": 1.0,
|
| 266 |
+
"eval_loss": 0.05580584332346916,
|
| 267 |
+
"eval_runtime": 0.206,
|
| 268 |
+
"eval_samples_per_second": 58.24,
|
| 269 |
+
"eval_steps_per_second": 9.707,
|
| 270 |
+
"step": 76
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 20.0,
|
| 274 |
+
"learning_rate": 8.944444444444446e-06,
|
| 275 |
+
"loss": 0.0564,
|
| 276 |
+
"step": 80
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 20.0,
|
| 280 |
+
"eval_accuracy": 1.0,
|
| 281 |
+
"eval_loss": 0.046213943511247635,
|
| 282 |
+
"eval_runtime": 0.3162,
|
| 283 |
+
"eval_samples_per_second": 37.95,
|
| 284 |
+
"eval_steps_per_second": 6.325,
|
| 285 |
+
"step": 80
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 21.0,
|
| 289 |
+
"eval_accuracy": 1.0,
|
| 290 |
+
"eval_loss": 0.05256287381052971,
|
| 291 |
+
"eval_runtime": 0.2134,
|
| 292 |
+
"eval_samples_per_second": 56.23,
|
| 293 |
+
"eval_steps_per_second": 9.372,
|
| 294 |
+
"step": 84
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 21.25,
|
| 298 |
+
"learning_rate": 8.805555555555557e-06,
|
| 299 |
+
"loss": 0.0703,
|
| 300 |
+
"step": 85
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 22.0,
|
| 304 |
+
"eval_accuracy": 1.0,
|
| 305 |
+
"eval_loss": 0.03809436038136482,
|
| 306 |
+
"eval_runtime": 0.2,
|
| 307 |
+
"eval_samples_per_second": 60.004,
|
| 308 |
+
"eval_steps_per_second": 10.001,
|
| 309 |
+
"step": 88
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 22.5,
|
| 313 |
+
"learning_rate": 8.666666666666668e-06,
|
| 314 |
+
"loss": 0.044,
|
| 315 |
+
"step": 90
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 23.0,
|
| 319 |
+
"eval_accuracy": 1.0,
|
| 320 |
+
"eval_loss": 0.01835465431213379,
|
| 321 |
+
"eval_runtime": 0.1962,
|
| 322 |
+
"eval_samples_per_second": 61.174,
|
| 323 |
+
"eval_steps_per_second": 10.196,
|
| 324 |
+
"step": 92
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 23.75,
|
| 328 |
+
"learning_rate": 8.527777777777779e-06,
|
| 329 |
+
"loss": 0.0239,
|
| 330 |
+
"step": 95
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.0,
|
| 334 |
+
"eval_accuracy": 1.0,
|
| 335 |
+
"eval_loss": 0.020169338211417198,
|
| 336 |
+
"eval_runtime": 0.2151,
|
| 337 |
+
"eval_samples_per_second": 55.779,
|
| 338 |
+
"eval_steps_per_second": 9.297,
|
| 339 |
+
"step": 96
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 25.0,
|
| 343 |
+
"learning_rate": 8.38888888888889e-06,
|
| 344 |
+
"loss": 0.0214,
|
| 345 |
+
"step": 100
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 25.0,
|
| 349 |
+
"eval_accuracy": 1.0,
|
| 350 |
+
"eval_loss": 0.024330079555511475,
|
| 351 |
+
"eval_runtime": 0.2292,
|
| 352 |
+
"eval_samples_per_second": 52.363,
|
| 353 |
+
"eval_steps_per_second": 8.727,
|
| 354 |
+
"step": 100
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 26.0,
|
| 358 |
+
"eval_accuracy": 1.0,
|
| 359 |
+
"eval_loss": 0.011685073375701904,
|
| 360 |
+
"eval_runtime": 0.2099,
|
| 361 |
+
"eval_samples_per_second": 57.169,
|
| 362 |
+
"eval_steps_per_second": 9.528,
|
| 363 |
+
"step": 104
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 26.25,
|
| 367 |
+
"learning_rate": 8.25e-06,
|
| 368 |
+
"loss": 0.031,
|
| 369 |
+
"step": 105
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 27.0,
|
| 373 |
+
"eval_accuracy": 1.0,
|
| 374 |
+
"eval_loss": 0.008951862342655659,
|
| 375 |
+
"eval_runtime": 0.2024,
|
| 376 |
+
"eval_samples_per_second": 59.277,
|
| 377 |
+
"eval_steps_per_second": 9.88,
|
| 378 |
+
"step": 108
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 27.5,
|
| 382 |
+
"learning_rate": 8.111111111111112e-06,
|
| 383 |
+
"loss": 0.0334,
|
| 384 |
+
"step": 110
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 28.0,
|
| 388 |
+
"eval_accuracy": 1.0,
|
| 389 |
+
"eval_loss": 0.03853602334856987,
|
| 390 |
+
"eval_runtime": 0.2061,
|
| 391 |
+
"eval_samples_per_second": 58.221,
|
| 392 |
+
"eval_steps_per_second": 9.703,
|
| 393 |
+
"step": 112
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 28.75,
|
| 397 |
+
"learning_rate": 7.972222222222224e-06,
|
| 398 |
+
"loss": 0.0046,
|
| 399 |
+
"step": 115
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 29.0,
|
| 403 |
+
"eval_accuracy": 1.0,
|
| 404 |
+
"eval_loss": 0.03924691677093506,
|
| 405 |
+
"eval_runtime": 0.3171,
|
| 406 |
+
"eval_samples_per_second": 37.847,
|
| 407 |
+
"eval_steps_per_second": 6.308,
|
| 408 |
+
"step": 116
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 30.0,
|
| 412 |
+
"learning_rate": 7.833333333333333e-06,
|
| 413 |
+
"loss": 0.0051,
|
| 414 |
+
"step": 120
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 30.0,
|
| 418 |
+
"eval_accuracy": 1.0,
|
| 419 |
+
"eval_loss": 0.0031117696780711412,
|
| 420 |
+
"eval_runtime": 0.217,
|
| 421 |
+
"eval_samples_per_second": 55.303,
|
| 422 |
+
"eval_steps_per_second": 9.217,
|
| 423 |
+
"step": 120
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 31.0,
|
| 427 |
+
"eval_accuracy": 1.0,
|
| 428 |
+
"eval_loss": 0.0025899012107402086,
|
| 429 |
+
"eval_runtime": 0.318,
|
| 430 |
+
"eval_samples_per_second": 37.734,
|
| 431 |
+
"eval_steps_per_second": 6.289,
|
| 432 |
+
"step": 124
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 31.25,
|
| 436 |
+
"learning_rate": 7.694444444444446e-06,
|
| 437 |
+
"loss": 0.0045,
|
| 438 |
+
"step": 125
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 32.0,
|
| 442 |
+
"eval_accuracy": 1.0,
|
| 443 |
+
"eval_loss": 0.002510249614715576,
|
| 444 |
+
"eval_runtime": 0.2034,
|
| 445 |
+
"eval_samples_per_second": 58.986,
|
| 446 |
+
"eval_steps_per_second": 9.831,
|
| 447 |
+
"step": 128
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 32.5,
|
| 451 |
+
"learning_rate": 7.555555555555556e-06,
|
| 452 |
+
"loss": 0.0133,
|
| 453 |
+
"step": 130
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 33.0,
|
| 457 |
+
"eval_accuracy": 0.9166666666666666,
|
| 458 |
+
"eval_loss": 0.07476559281349182,
|
| 459 |
+
"eval_runtime": 0.2934,
|
| 460 |
+
"eval_samples_per_second": 40.893,
|
| 461 |
+
"eval_steps_per_second": 6.816,
|
| 462 |
+
"step": 132
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 33.75,
|
| 466 |
+
"learning_rate": 7.416666666666668e-06,
|
| 467 |
+
"loss": 0.0014,
|
| 468 |
+
"step": 135
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 34.0,
|
| 472 |
+
"eval_accuracy": 0.9166666666666666,
|
| 473 |
+
"eval_loss": 0.16315531730651855,
|
| 474 |
+
"eval_runtime": 0.2053,
|
| 475 |
+
"eval_samples_per_second": 58.463,
|
| 476 |
+
"eval_steps_per_second": 9.744,
|
| 477 |
+
"step": 136
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 35.0,
|
| 481 |
+
"learning_rate": 7.277777777777778e-06,
|
| 482 |
+
"loss": 0.0134,
|
| 483 |
+
"step": 140
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 35.0,
|
| 487 |
+
"eval_accuracy": 1.0,
|
| 488 |
+
"eval_loss": 0.006152550224214792,
|
| 489 |
+
"eval_runtime": 0.2135,
|
| 490 |
+
"eval_samples_per_second": 56.219,
|
| 491 |
+
"eval_steps_per_second": 9.37,
|
| 492 |
+
"step": 140
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 36.0,
|
| 496 |
+
"eval_accuracy": 1.0,
|
| 497 |
+
"eval_loss": 0.005291670560836792,
|
| 498 |
+
"eval_runtime": 0.2054,
|
| 499 |
+
"eval_samples_per_second": 58.422,
|
| 500 |
+
"eval_steps_per_second": 9.737,
|
| 501 |
+
"step": 144
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.25,
|
| 505 |
+
"learning_rate": 7.13888888888889e-06,
|
| 506 |
+
"loss": 0.0365,
|
| 507 |
+
"step": 145
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 37.0,
|
| 511 |
+
"eval_accuracy": 1.0,
|
| 512 |
+
"eval_loss": 0.0365130789577961,
|
| 513 |
+
"eval_runtime": 0.2268,
|
| 514 |
+
"eval_samples_per_second": 52.911,
|
| 515 |
+
"eval_steps_per_second": 8.819,
|
| 516 |
+
"step": 148
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 37.5,
|
| 520 |
+
"learning_rate": 7e-06,
|
| 521 |
+
"loss": 0.001,
|
| 522 |
+
"step": 150
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 38.0,
|
| 526 |
+
"eval_accuracy": 1.0,
|
| 527 |
+
"eval_loss": 0.0017459988594055176,
|
| 528 |
+
"eval_runtime": 0.2143,
|
| 529 |
+
"eval_samples_per_second": 56.005,
|
| 530 |
+
"eval_steps_per_second": 9.334,
|
| 531 |
+
"step": 152
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 38.75,
|
| 535 |
+
"learning_rate": 6.88888888888889e-06,
|
| 536 |
+
"loss": 0.0503,
|
| 537 |
+
"step": 155
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 39.0,
|
| 541 |
+
"eval_accuracy": 1.0,
|
| 542 |
+
"eval_loss": 0.05090367794036865,
|
| 543 |
+
"eval_runtime": 0.208,
|
| 544 |
+
"eval_samples_per_second": 57.695,
|
| 545 |
+
"eval_steps_per_second": 9.616,
|
| 546 |
+
"step": 156
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 40.0,
|
| 550 |
+
"learning_rate": 6.750000000000001e-06,
|
| 551 |
+
"loss": 0.0094,
|
| 552 |
+
"step": 160
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 40.0,
|
| 556 |
+
"eval_accuracy": 0.9166666666666666,
|
| 557 |
+
"eval_loss": 0.07808921486139297,
|
| 558 |
+
"eval_runtime": 0.2086,
|
| 559 |
+
"eval_samples_per_second": 57.523,
|
| 560 |
+
"eval_steps_per_second": 9.587,
|
| 561 |
+
"step": 160
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 41.0,
|
| 565 |
+
"eval_accuracy": 0.9166666666666666,
|
| 566 |
+
"eval_loss": 0.22892427444458008,
|
| 567 |
+
"eval_runtime": 0.2158,
|
| 568 |
+
"eval_samples_per_second": 55.599,
|
| 569 |
+
"eval_steps_per_second": 9.267,
|
| 570 |
+
"step": 164
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 41.25,
|
| 574 |
+
"learning_rate": 6.6111111111111115e-06,
|
| 575 |
+
"loss": 0.0231,
|
| 576 |
+
"step": 165
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 42.0,
|
| 580 |
+
"eval_accuracy": 0.9166666666666666,
|
| 581 |
+
"eval_loss": 0.16741518676280975,
|
| 582 |
+
"eval_runtime": 0.2199,
|
| 583 |
+
"eval_samples_per_second": 54.565,
|
| 584 |
+
"eval_steps_per_second": 9.094,
|
| 585 |
+
"step": 168
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 42.5,
|
| 589 |
+
"learning_rate": 6.472222222222223e-06,
|
| 590 |
+
"loss": 0.0002,
|
| 591 |
+
"step": 170
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 43.0,
|
| 595 |
+
"eval_accuracy": 0.9166666666666666,
|
| 596 |
+
"eval_loss": 0.25974369049072266,
|
| 597 |
+
"eval_runtime": 0.297,
|
| 598 |
+
"eval_samples_per_second": 40.411,
|
| 599 |
+
"eval_steps_per_second": 6.735,
|
| 600 |
+
"step": 172
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 43.75,
|
| 604 |
+
"learning_rate": 6.333333333333333e-06,
|
| 605 |
+
"loss": 0.0785,
|
| 606 |
+
"step": 175
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 44.0,
|
| 610 |
+
"eval_accuracy": 0.9166666666666666,
|
| 611 |
+
"eval_loss": 0.2814592123031616,
|
| 612 |
+
"eval_runtime": 0.1943,
|
| 613 |
+
"eval_samples_per_second": 61.754,
|
| 614 |
+
"eval_steps_per_second": 10.292,
|
| 615 |
+
"step": 176
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 45.0,
|
| 619 |
+
"learning_rate": 6.194444444444445e-06,
|
| 620 |
+
"loss": 0.0009,
|
| 621 |
+
"step": 180
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 45.0,
|
| 625 |
+
"eval_accuracy": 0.9166666666666666,
|
| 626 |
+
"eval_loss": 0.2058378905057907,
|
| 627 |
+
"eval_runtime": 0.3052,
|
| 628 |
+
"eval_samples_per_second": 39.313,
|
| 629 |
+
"eval_steps_per_second": 6.552,
|
| 630 |
+
"step": 180
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 46.0,
|
| 634 |
+
"eval_accuracy": 1.0,
|
| 635 |
+
"eval_loss": 0.000926971435546875,
|
| 636 |
+
"eval_runtime": 0.2098,
|
| 637 |
+
"eval_samples_per_second": 57.192,
|
| 638 |
+
"eval_steps_per_second": 9.532,
|
| 639 |
+
"step": 184
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 46.25,
|
| 643 |
+
"learning_rate": 6.055555555555555e-06,
|
| 644 |
+
"loss": 0.0395,
|
| 645 |
+
"step": 185
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 47.0,
|
| 649 |
+
"eval_accuracy": 1.0,
|
| 650 |
+
"eval_loss": 0.0010882416972890496,
|
| 651 |
+
"eval_runtime": 0.3081,
|
| 652 |
+
"eval_samples_per_second": 38.948,
|
| 653 |
+
"eval_steps_per_second": 6.491,
|
| 654 |
+
"step": 188
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 47.5,
|
| 658 |
+
"learning_rate": 5.916666666666667e-06,
|
| 659 |
+
"loss": 0.0158,
|
| 660 |
+
"step": 190
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 48.0,
|
| 664 |
+
"eval_accuracy": 1.0,
|
| 665 |
+
"eval_loss": 0.000389377266401425,
|
| 666 |
+
"eval_runtime": 0.2087,
|
| 667 |
+
"eval_samples_per_second": 57.493,
|
| 668 |
+
"eval_steps_per_second": 9.582,
|
| 669 |
+
"step": 192
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 48.75,
|
| 673 |
+
"learning_rate": 5.777777777777778e-06,
|
| 674 |
+
"loss": 0.001,
|
| 675 |
+
"step": 195
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 49.0,
|
| 679 |
+
"eval_accuracy": 1.0,
|
| 680 |
+
"eval_loss": 0.0034777026157826185,
|
| 681 |
+
"eval_runtime": 0.2113,
|
| 682 |
+
"eval_samples_per_second": 56.801,
|
| 683 |
+
"eval_steps_per_second": 9.467,
|
| 684 |
+
"step": 196
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 50.0,
|
| 688 |
+
"learning_rate": 5.638888888888889e-06,
|
| 689 |
+
"loss": 0.0027,
|
| 690 |
+
"step": 200
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 50.0,
|
| 694 |
+
"eval_accuracy": 1.0,
|
| 695 |
+
"eval_loss": 0.0018774966010823846,
|
| 696 |
+
"eval_runtime": 0.1971,
|
| 697 |
+
"eval_samples_per_second": 60.883,
|
| 698 |
+
"eval_steps_per_second": 10.147,
|
| 699 |
+
"step": 200
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 51.0,
|
| 703 |
+
"eval_accuracy": 1.0,
|
| 704 |
+
"eval_loss": 0.0005501409177668393,
|
| 705 |
+
"eval_runtime": 0.2093,
|
| 706 |
+
"eval_samples_per_second": 57.332,
|
| 707 |
+
"eval_steps_per_second": 9.555,
|
| 708 |
+
"step": 204
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 51.25,
|
| 712 |
+
"learning_rate": 5.500000000000001e-06,
|
| 713 |
+
"loss": 0.0005,
|
| 714 |
+
"step": 205
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 52.0,
|
| 718 |
+
"eval_accuracy": 1.0,
|
| 719 |
+
"eval_loss": 0.0013666549930348992,
|
| 720 |
+
"eval_runtime": 0.2027,
|
| 721 |
+
"eval_samples_per_second": 59.215,
|
| 722 |
+
"eval_steps_per_second": 9.869,
|
| 723 |
+
"step": 208
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 52.5,
|
| 727 |
+
"learning_rate": 5.361111111111112e-06,
|
| 728 |
+
"loss": 0.003,
|
| 729 |
+
"step": 210
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 53.0,
|
| 733 |
+
"eval_accuracy": 1.0,
|
| 734 |
+
"eval_loss": 0.0018307765712961555,
|
| 735 |
+
"eval_runtime": 0.2153,
|
| 736 |
+
"eval_samples_per_second": 55.739,
|
| 737 |
+
"eval_steps_per_second": 9.29,
|
| 738 |
+
"step": 212
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 53.75,
|
| 742 |
+
"learning_rate": 5.2222222222222226e-06,
|
| 743 |
+
"loss": 0.0039,
|
| 744 |
+
"step": 215
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 54.0,
|
| 748 |
+
"eval_accuracy": 1.0,
|
| 749 |
+
"eval_loss": 0.01105162501335144,
|
| 750 |
+
"eval_runtime": 0.2005,
|
| 751 |
+
"eval_samples_per_second": 59.849,
|
| 752 |
+
"eval_steps_per_second": 9.975,
|
| 753 |
+
"step": 216
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 55.0,
|
| 757 |
+
"learning_rate": 5.0833333333333335e-06,
|
| 758 |
+
"loss": 0.0081,
|
| 759 |
+
"step": 220
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 55.0,
|
| 763 |
+
"eval_accuracy": 1.0,
|
| 764 |
+
"eval_loss": 0.021254947409033775,
|
| 765 |
+
"eval_runtime": 0.2174,
|
| 766 |
+
"eval_samples_per_second": 55.192,
|
| 767 |
+
"eval_steps_per_second": 9.199,
|
| 768 |
+
"step": 220
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 56.0,
|
| 772 |
+
"eval_accuracy": 1.0,
|
| 773 |
+
"eval_loss": 0.014328837394714355,
|
| 774 |
+
"eval_runtime": 0.2399,
|
| 775 |
+
"eval_samples_per_second": 50.017,
|
| 776 |
+
"eval_steps_per_second": 8.336,
|
| 777 |
+
"step": 224
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 56.25,
|
| 781 |
+
"learning_rate": 4.944444444444445e-06,
|
| 782 |
+
"loss": 0.0001,
|
| 783 |
+
"step": 225
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 57.0,
|
| 787 |
+
"eval_accuracy": 1.0,
|
| 788 |
+
"eval_loss": 0.00964482594281435,
|
| 789 |
+
"eval_runtime": 0.2188,
|
| 790 |
+
"eval_samples_per_second": 54.838,
|
| 791 |
+
"eval_steps_per_second": 9.14,
|
| 792 |
+
"step": 228
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 57.5,
|
| 796 |
+
"learning_rate": 4.805555555555556e-06,
|
| 797 |
+
"loss": 0.0002,
|
| 798 |
+
"step": 230
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 58.0,
|
| 802 |
+
"eval_accuracy": 1.0,
|
| 803 |
+
"eval_loss": 0.0017709335079416633,
|
| 804 |
+
"eval_runtime": 0.2916,
|
| 805 |
+
"eval_samples_per_second": 41.158,
|
| 806 |
+
"eval_steps_per_second": 6.86,
|
| 807 |
+
"step": 232
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 58.75,
|
| 811 |
+
"learning_rate": 4.666666666666667e-06,
|
| 812 |
+
"loss": 0.0035,
|
| 813 |
+
"step": 235
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 59.0,
|
| 817 |
+
"eval_accuracy": 1.0,
|
| 818 |
+
"eval_loss": 0.0005045731668360531,
|
| 819 |
+
"eval_runtime": 0.2036,
|
| 820 |
+
"eval_samples_per_second": 58.93,
|
| 821 |
+
"eval_steps_per_second": 9.822,
|
| 822 |
+
"step": 236
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 60.0,
|
| 826 |
+
"learning_rate": 4.527777777777778e-06,
|
| 827 |
+
"loss": 0.0049,
|
| 828 |
+
"step": 240
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 60.0,
|
| 832 |
+
"eval_accuracy": 1.0,
|
| 833 |
+
"eval_loss": 0.0014440218219533563,
|
| 834 |
+
"eval_runtime": 0.3276,
|
| 835 |
+
"eval_samples_per_second": 36.631,
|
| 836 |
+
"eval_steps_per_second": 6.105,
|
| 837 |
+
"step": 240
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 61.0,
|
| 841 |
+
"eval_accuracy": 1.0,
|
| 842 |
+
"eval_loss": 0.0020179252605885267,
|
| 843 |
+
"eval_runtime": 0.2179,
|
| 844 |
+
"eval_samples_per_second": 55.076,
|
| 845 |
+
"eval_steps_per_second": 9.179,
|
| 846 |
+
"step": 244
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 61.25,
|
| 850 |
+
"learning_rate": 4.388888888888889e-06,
|
| 851 |
+
"loss": 0.0006,
|
| 852 |
+
"step": 245
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 62.0,
|
| 856 |
+
"eval_accuracy": 1.0,
|
| 857 |
+
"eval_loss": 0.0003785987792070955,
|
| 858 |
+
"eval_runtime": 0.2267,
|
| 859 |
+
"eval_samples_per_second": 52.937,
|
| 860 |
+
"eval_steps_per_second": 8.823,
|
| 861 |
+
"step": 248
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 62.5,
|
| 865 |
+
"learning_rate": 4.25e-06,
|
| 866 |
+
"loss": 0.0001,
|
| 867 |
+
"step": 250
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 63.0,
|
| 871 |
+
"eval_accuracy": 1.0,
|
| 872 |
+
"eval_loss": 0.00021861989807803184,
|
| 873 |
+
"eval_runtime": 0.2108,
|
| 874 |
+
"eval_samples_per_second": 56.934,
|
| 875 |
+
"eval_steps_per_second": 9.489,
|
| 876 |
+
"step": 252
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 63.75,
|
| 880 |
+
"learning_rate": 4.111111111111111e-06,
|
| 881 |
+
"loss": 0.0272,
|
| 882 |
+
"step": 255
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 64.0,
|
| 886 |
+
"eval_accuracy": 1.0,
|
| 887 |
+
"eval_loss": 0.0002175172121496871,
|
| 888 |
+
"eval_runtime": 0.2182,
|
| 889 |
+
"eval_samples_per_second": 55.001,
|
| 890 |
+
"eval_steps_per_second": 9.167,
|
| 891 |
+
"step": 256
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 65.0,
|
| 895 |
+
"learning_rate": 3.972222222222223e-06,
|
| 896 |
+
"loss": 0.0016,
|
| 897 |
+
"step": 260
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 65.0,
|
| 901 |
+
"eval_accuracy": 1.0,
|
| 902 |
+
"eval_loss": 0.0003321866097394377,
|
| 903 |
+
"eval_runtime": 0.1986,
|
| 904 |
+
"eval_samples_per_second": 60.421,
|
| 905 |
+
"eval_steps_per_second": 10.07,
|
| 906 |
+
"step": 260
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 66.0,
|
| 910 |
+
"eval_accuracy": 1.0,
|
| 911 |
+
"eval_loss": 0.0012783011188730597,
|
| 912 |
+
"eval_runtime": 0.2135,
|
| 913 |
+
"eval_samples_per_second": 56.201,
|
| 914 |
+
"eval_steps_per_second": 9.367,
|
| 915 |
+
"step": 264
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 66.25,
|
| 919 |
+
"learning_rate": 3.833333333333334e-06,
|
| 920 |
+
"loss": 0.0002,
|
| 921 |
+
"step": 265
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 67.0,
|
| 925 |
+
"eval_accuracy": 1.0,
|
| 926 |
+
"eval_loss": 0.004446456674486399,
|
| 927 |
+
"eval_runtime": 0.2127,
|
| 928 |
+
"eval_samples_per_second": 56.422,
|
| 929 |
+
"eval_steps_per_second": 9.404,
|
| 930 |
+
"step": 268
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 67.5,
|
| 934 |
+
"learning_rate": 3.694444444444445e-06,
|
| 935 |
+
"loss": 0.0001,
|
| 936 |
+
"step": 270
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 68.0,
|
| 940 |
+
"eval_accuracy": 1.0,
|
| 941 |
+
"eval_loss": 0.013481984846293926,
|
| 942 |
+
"eval_runtime": 0.2043,
|
| 943 |
+
"eval_samples_per_second": 58.732,
|
| 944 |
+
"eval_steps_per_second": 9.789,
|
| 945 |
+
"step": 272
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 68.75,
|
| 949 |
+
"learning_rate": 3.555555555555556e-06,
|
| 950 |
+
"loss": 0.0002,
|
| 951 |
+
"step": 275
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 69.0,
|
| 955 |
+
"eval_accuracy": 1.0,
|
| 956 |
+
"eval_loss": 0.012796193361282349,
|
| 957 |
+
"eval_runtime": 0.2261,
|
| 958 |
+
"eval_samples_per_second": 53.079,
|
| 959 |
+
"eval_steps_per_second": 8.846,
|
| 960 |
+
"step": 276
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 70.0,
|
| 964 |
+
"learning_rate": 3.416666666666667e-06,
|
| 965 |
+
"loss": 0.0002,
|
| 966 |
+
"step": 280
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 70.0,
|
| 970 |
+
"eval_accuracy": 1.0,
|
| 971 |
+
"eval_loss": 0.01079349685460329,
|
| 972 |
+
"eval_runtime": 0.1991,
|
| 973 |
+
"eval_samples_per_second": 60.277,
|
| 974 |
+
"eval_steps_per_second": 10.046,
|
| 975 |
+
"step": 280
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 71.0,
|
| 979 |
+
"eval_accuracy": 1.0,
|
| 980 |
+
"eval_loss": 0.008993417024612427,
|
| 981 |
+
"eval_runtime": 0.2933,
|
| 982 |
+
"eval_samples_per_second": 40.92,
|
| 983 |
+
"eval_steps_per_second": 6.82,
|
| 984 |
+
"step": 284
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 71.25,
|
| 988 |
+
"learning_rate": 3.277777777777778e-06,
|
| 989 |
+
"loss": 0.0001,
|
| 990 |
+
"step": 285
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 72.0,
|
| 994 |
+
"eval_accuracy": 1.0,
|
| 995 |
+
"eval_loss": 0.006850639823824167,
|
| 996 |
+
"eval_runtime": 0.2074,
|
| 997 |
+
"eval_samples_per_second": 57.85,
|
| 998 |
+
"eval_steps_per_second": 9.642,
|
| 999 |
+
"step": 288
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 72.5,
|
| 1003 |
+
"learning_rate": 3.138888888888889e-06,
|
| 1004 |
+
"loss": 0.0001,
|
| 1005 |
+
"step": 290
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 73.0,
|
| 1009 |
+
"eval_accuracy": 1.0,
|
| 1010 |
+
"eval_loss": 0.0049311816692352295,
|
| 1011 |
+
"eval_runtime": 0.2813,
|
| 1012 |
+
"eval_samples_per_second": 42.663,
|
| 1013 |
+
"eval_steps_per_second": 7.111,
|
| 1014 |
+
"step": 292
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 73.75,
|
| 1018 |
+
"learning_rate": 3e-06,
|
| 1019 |
+
"loss": 0.0002,
|
| 1020 |
+
"step": 295
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 74.0,
|
| 1024 |
+
"eval_accuracy": 1.0,
|
| 1025 |
+
"eval_loss": 0.0031823813915252686,
|
| 1026 |
+
"eval_runtime": 0.2076,
|
| 1027 |
+
"eval_samples_per_second": 57.79,
|
| 1028 |
+
"eval_steps_per_second": 9.632,
|
| 1029 |
+
"step": 296
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 75.0,
|
| 1033 |
+
"learning_rate": 2.861111111111111e-06,
|
| 1034 |
+
"loss": 0.0,
|
| 1035 |
+
"step": 300
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 75.0,
|
| 1039 |
+
"eval_accuracy": 1.0,
|
| 1040 |
+
"eval_loss": 0.0018174747237935662,
|
| 1041 |
+
"eval_runtime": 0.1974,
|
| 1042 |
+
"eval_samples_per_second": 60.778,
|
| 1043 |
+
"eval_steps_per_second": 10.13,
|
| 1044 |
+
"step": 300
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 76.0,
|
| 1048 |
+
"eval_accuracy": 1.0,
|
| 1049 |
+
"eval_loss": 0.0033631324768066406,
|
| 1050 |
+
"eval_runtime": 0.2244,
|
| 1051 |
+
"eval_samples_per_second": 53.472,
|
| 1052 |
+
"eval_steps_per_second": 8.912,
|
| 1053 |
+
"step": 304
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 76.25,
|
| 1057 |
+
"learning_rate": 2.7222222222222224e-06,
|
| 1058 |
+
"loss": 0.0003,
|
| 1059 |
+
"step": 305
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 77.0,
|
| 1063 |
+
"eval_accuracy": 1.0,
|
| 1064 |
+
"eval_loss": 0.005034655332565308,
|
| 1065 |
+
"eval_runtime": 0.2054,
|
| 1066 |
+
"eval_samples_per_second": 58.436,
|
| 1067 |
+
"eval_steps_per_second": 9.739,
|
| 1068 |
+
"step": 308
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 77.5,
|
| 1072 |
+
"learning_rate": 2.5833333333333337e-06,
|
| 1073 |
+
"loss": 0.0001,
|
| 1074 |
+
"step": 310
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 78.0,
|
| 1078 |
+
"eval_accuracy": 1.0,
|
| 1079 |
+
"eval_loss": 0.0060021779499948025,
|
| 1080 |
+
"eval_runtime": 0.2165,
|
| 1081 |
+
"eval_samples_per_second": 55.438,
|
| 1082 |
+
"eval_steps_per_second": 9.24,
|
| 1083 |
+
"step": 312
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 78.75,
|
| 1087 |
+
"learning_rate": 2.4444444444444447e-06,
|
| 1088 |
+
"loss": 0.0,
|
| 1089 |
+
"step": 315
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 79.0,
|
| 1093 |
+
"eval_accuracy": 1.0,
|
| 1094 |
+
"eval_loss": 0.006388425827026367,
|
| 1095 |
+
"eval_runtime": 0.2076,
|
| 1096 |
+
"eval_samples_per_second": 57.793,
|
| 1097 |
+
"eval_steps_per_second": 9.632,
|
| 1098 |
+
"step": 316
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 80.0,
|
| 1102 |
+
"learning_rate": 2.305555555555556e-06,
|
| 1103 |
+
"loss": 0.0001,
|
| 1104 |
+
"step": 320
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 80.0,
|
| 1108 |
+
"eval_accuracy": 1.0,
|
| 1109 |
+
"eval_loss": 0.0061742267571389675,
|
| 1110 |
+
"eval_runtime": 0.2188,
|
| 1111 |
+
"eval_samples_per_second": 54.852,
|
| 1112 |
+
"eval_steps_per_second": 9.142,
|
| 1113 |
+
"step": 320
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 81.0,
|
| 1117 |
+
"eval_accuracy": 1.0,
|
| 1118 |
+
"eval_loss": 0.005842983722686768,
|
| 1119 |
+
"eval_runtime": 0.2181,
|
| 1120 |
+
"eval_samples_per_second": 55.023,
|
| 1121 |
+
"eval_steps_per_second": 9.171,
|
| 1122 |
+
"step": 324
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 81.25,
|
| 1126 |
+
"learning_rate": 2.166666666666667e-06,
|
| 1127 |
+
"loss": 0.0001,
|
| 1128 |
+
"step": 325
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 82.0,
|
| 1132 |
+
"eval_accuracy": 1.0,
|
| 1133 |
+
"eval_loss": 0.0059954822063446045,
|
| 1134 |
+
"eval_runtime": 0.2026,
|
| 1135 |
+
"eval_samples_per_second": 59.232,
|
| 1136 |
+
"eval_steps_per_second": 9.872,
|
| 1137 |
+
"step": 328
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 82.5,
|
| 1141 |
+
"learning_rate": 2.027777777777778e-06,
|
| 1142 |
+
"loss": 0.0001,
|
| 1143 |
+
"step": 330
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 83.0,
|
| 1147 |
+
"eval_accuracy": 1.0,
|
| 1148 |
+
"eval_loss": 0.006137927528470755,
|
| 1149 |
+
"eval_runtime": 0.2087,
|
| 1150 |
+
"eval_samples_per_second": 57.509,
|
| 1151 |
+
"eval_steps_per_second": 9.585,
|
| 1152 |
+
"step": 332
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 83.75,
|
| 1156 |
+
"learning_rate": 1.888888888888889e-06,
|
| 1157 |
+
"loss": 0.0,
|
| 1158 |
+
"step": 335
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 84.0,
|
| 1162 |
+
"eval_accuracy": 1.0,
|
| 1163 |
+
"eval_loss": 0.007910847663879395,
|
| 1164 |
+
"eval_runtime": 0.2136,
|
| 1165 |
+
"eval_samples_per_second": 56.177,
|
| 1166 |
+
"eval_steps_per_second": 9.363,
|
| 1167 |
+
"step": 336
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 85.0,
|
| 1171 |
+
"learning_rate": 1.75e-06,
|
| 1172 |
+
"loss": 0.0005,
|
| 1173 |
+
"step": 340
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 85.0,
|
| 1177 |
+
"eval_accuracy": 1.0,
|
| 1178 |
+
"eval_loss": 0.012745578773319721,
|
| 1179 |
+
"eval_runtime": 0.2199,
|
| 1180 |
+
"eval_samples_per_second": 54.578,
|
| 1181 |
+
"eval_steps_per_second": 9.096,
|
| 1182 |
+
"step": 340
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 86.0,
|
| 1186 |
+
"eval_accuracy": 1.0,
|
| 1187 |
+
"eval_loss": 0.021893838420510292,
|
| 1188 |
+
"eval_runtime": 0.2149,
|
| 1189 |
+
"eval_samples_per_second": 55.845,
|
| 1190 |
+
"eval_steps_per_second": 9.307,
|
| 1191 |
+
"step": 344
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 86.25,
|
| 1195 |
+
"learning_rate": 1.6111111111111113e-06,
|
| 1196 |
+
"loss": 0.0002,
|
| 1197 |
+
"step": 345
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 87.0,
|
| 1201 |
+
"eval_accuracy": 1.0,
|
| 1202 |
+
"eval_loss": 0.031531721353530884,
|
| 1203 |
+
"eval_runtime": 0.2233,
|
| 1204 |
+
"eval_samples_per_second": 53.743,
|
| 1205 |
+
"eval_steps_per_second": 8.957,
|
| 1206 |
+
"step": 348
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 87.5,
|
| 1210 |
+
"learning_rate": 1.4722222222222225e-06,
|
| 1211 |
+
"loss": 0.0001,
|
| 1212 |
+
"step": 350
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 88.0,
|
| 1216 |
+
"eval_accuracy": 1.0,
|
| 1217 |
+
"eval_loss": 0.03811268135905266,
|
| 1218 |
+
"eval_runtime": 0.2096,
|
| 1219 |
+
"eval_samples_per_second": 57.255,
|
| 1220 |
+
"eval_steps_per_second": 9.542,
|
| 1221 |
+
"step": 352
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 88.75,
|
| 1225 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 1226 |
+
"loss": 0.0001,
|
| 1227 |
+
"step": 355
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 89.0,
|
| 1231 |
+
"eval_accuracy": 1.0,
|
| 1232 |
+
"eval_loss": 0.04021235182881355,
|
| 1233 |
+
"eval_runtime": 0.2105,
|
| 1234 |
+
"eval_samples_per_second": 57.008,
|
| 1235 |
+
"eval_steps_per_second": 9.501,
|
| 1236 |
+
"step": 356
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 90.0,
|
| 1240 |
+
"learning_rate": 1.1944444444444446e-06,
|
| 1241 |
+
"loss": 0.0,
|
| 1242 |
+
"step": 360
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 90.0,
|
| 1246 |
+
"eval_accuracy": 1.0,
|
| 1247 |
+
"eval_loss": 0.040413498878479004,
|
| 1248 |
+
"eval_runtime": 0.2044,
|
| 1249 |
+
"eval_samples_per_second": 58.696,
|
| 1250 |
+
"eval_steps_per_second": 9.783,
|
| 1251 |
+
"step": 360
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 91.0,
|
| 1255 |
+
"eval_accuracy": 1.0,
|
| 1256 |
+
"eval_loss": 0.0374186746776104,
|
| 1257 |
+
"eval_runtime": 0.316,
|
| 1258 |
+
"eval_samples_per_second": 37.973,
|
| 1259 |
+
"eval_steps_per_second": 6.329,
|
| 1260 |
+
"step": 364
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 91.25,
|
| 1264 |
+
"learning_rate": 1.0555555555555557e-06,
|
| 1265 |
+
"loss": 0.0,
|
| 1266 |
+
"step": 365
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 92.0,
|
| 1270 |
+
"eval_accuracy": 1.0,
|
| 1271 |
+
"eval_loss": 0.03140836954116821,
|
| 1272 |
+
"eval_runtime": 0.2142,
|
| 1273 |
+
"eval_samples_per_second": 56.034,
|
| 1274 |
+
"eval_steps_per_second": 9.339,
|
| 1275 |
+
"step": 368
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 92.5,
|
| 1279 |
+
"learning_rate": 9.166666666666666e-07,
|
| 1280 |
+
"loss": 0.0003,
|
| 1281 |
+
"step": 370
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 93.0,
|
| 1285 |
+
"eval_accuracy": 1.0,
|
| 1286 |
+
"eval_loss": 0.01769857667386532,
|
| 1287 |
+
"eval_runtime": 0.2018,
|
| 1288 |
+
"eval_samples_per_second": 59.478,
|
| 1289 |
+
"eval_steps_per_second": 9.913,
|
| 1290 |
+
"step": 372
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 93.75,
|
| 1294 |
+
"learning_rate": 7.777777777777779e-07,
|
| 1295 |
+
"loss": 0.0001,
|
| 1296 |
+
"step": 375
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 94.0,
|
| 1300 |
+
"eval_accuracy": 1.0,
|
| 1301 |
+
"eval_loss": 0.011687318794429302,
|
| 1302 |
+
"eval_runtime": 0.2027,
|
| 1303 |
+
"eval_samples_per_second": 59.211,
|
| 1304 |
+
"eval_steps_per_second": 9.869,
|
| 1305 |
+
"step": 376
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 95.0,
|
| 1309 |
+
"learning_rate": 6.388888888888889e-07,
|
| 1310 |
+
"loss": 0.0001,
|
| 1311 |
+
"step": 380
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 95.0,
|
| 1315 |
+
"eval_accuracy": 1.0,
|
| 1316 |
+
"eval_loss": 0.0088284807279706,
|
| 1317 |
+
"eval_runtime": 0.2065,
|
| 1318 |
+
"eval_samples_per_second": 58.106,
|
| 1319 |
+
"eval_steps_per_second": 9.684,
|
| 1320 |
+
"step": 380
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 96.0,
|
| 1324 |
+
"eval_accuracy": 1.0,
|
| 1325 |
+
"eval_loss": 0.007670491933822632,
|
| 1326 |
+
"eval_runtime": 0.2024,
|
| 1327 |
+
"eval_samples_per_second": 59.303,
|
| 1328 |
+
"eval_steps_per_second": 9.884,
|
| 1329 |
+
"step": 384
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 96.25,
|
| 1333 |
+
"learning_rate": 5.000000000000001e-07,
|
| 1334 |
+
"loss": 0.0,
|
| 1335 |
+
"step": 385
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 97.0,
|
| 1339 |
+
"eval_accuracy": 1.0,
|
| 1340 |
+
"eval_loss": 0.00695762038230896,
|
| 1341 |
+
"eval_runtime": 0.2135,
|
| 1342 |
+
"eval_samples_per_second": 56.218,
|
| 1343 |
+
"eval_steps_per_second": 9.37,
|
| 1344 |
+
"step": 388
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 97.5,
|
| 1348 |
+
"learning_rate": 3.611111111111111e-07,
|
| 1349 |
+
"loss": 0.0001,
|
| 1350 |
+
"step": 390
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 98.0,
|
| 1354 |
+
"eval_accuracy": 1.0,
|
| 1355 |
+
"eval_loss": 0.0064276158809661865,
|
| 1356 |
+
"eval_runtime": 0.2059,
|
| 1357 |
+
"eval_samples_per_second": 58.292,
|
| 1358 |
+
"eval_steps_per_second": 9.715,
|
| 1359 |
+
"step": 392
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 98.75,
|
| 1363 |
+
"learning_rate": 2.2222222222222224e-07,
|
| 1364 |
+
"loss": 0.0001,
|
| 1365 |
+
"step": 395
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 99.0,
|
| 1369 |
+
"eval_accuracy": 1.0,
|
| 1370 |
+
"eval_loss": 0.006080567836761475,
|
| 1371 |
+
"eval_runtime": 0.2091,
|
| 1372 |
+
"eval_samples_per_second": 57.386,
|
| 1373 |
+
"eval_steps_per_second": 9.564,
|
| 1374 |
+
"step": 396
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 100.0,
|
| 1378 |
+
"learning_rate": 8.333333333333334e-08,
|
| 1379 |
+
"loss": 0.0001,
|
| 1380 |
+
"step": 400
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 100.0,
|
| 1384 |
+
"eval_accuracy": 1.0,
|
| 1385 |
+
"eval_loss": 0.005917658563703299,
|
| 1386 |
+
"eval_runtime": 0.2205,
|
| 1387 |
+
"eval_samples_per_second": 54.429,
|
| 1388 |
+
"eval_steps_per_second": 9.072,
|
| 1389 |
+
"step": 400
|
| 1390 |
+
}
|
| 1391 |
+
],
|
| 1392 |
+
"logging_steps": 5,
|
| 1393 |
+
"max_steps": 400,
|
| 1394 |
+
"num_train_epochs": 100,
|
| 1395 |
+
"save_steps": 500,
|
| 1396 |
+
"total_flos": 4.70073333768192e+17,
|
| 1397 |
+
"trial_name": null,
|
| 1398 |
+
"trial_params": null
|
| 1399 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5af0d25da3647f02cb549dbce3f3e15125788273231f5cddb0a0a9e60095d680
|
| 3 |
+
size 4600
|