Upload 8 files
Browse files- config.json +54 -0
- optimizer.pt +3 -0
- preprocessor_config.json +27 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2638 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/bit-50",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BitForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"conv_layer": "std_conv",
|
| 7 |
+
"depths": [
|
| 8 |
+
3,
|
| 9 |
+
4,
|
| 10 |
+
6,
|
| 11 |
+
3
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.0,
|
| 14 |
+
"embedding_dynamic_padding": false,
|
| 15 |
+
"embedding_size": 64,
|
| 16 |
+
"global_padding": null,
|
| 17 |
+
"hidden_act": "relu",
|
| 18 |
+
"hidden_sizes": [
|
| 19 |
+
256,
|
| 20 |
+
512,
|
| 21 |
+
1024,
|
| 22 |
+
2048
|
| 23 |
+
],
|
| 24 |
+
"id2label": {
|
| 25 |
+
"0": "glaucoma",
|
| 26 |
+
"1": "normal"
|
| 27 |
+
},
|
| 28 |
+
"label2id": {
|
| 29 |
+
"glaucoma": 0,
|
| 30 |
+
"normal": 1
|
| 31 |
+
},
|
| 32 |
+
"layer_type": "preactivation",
|
| 33 |
+
"model_type": "bit",
|
| 34 |
+
"num_channels": 3,
|
| 35 |
+
"num_groups": 32,
|
| 36 |
+
"out_features": [
|
| 37 |
+
"stage4"
|
| 38 |
+
],
|
| 39 |
+
"out_indices": [
|
| 40 |
+
4
|
| 41 |
+
],
|
| 42 |
+
"output_stride": 32,
|
| 43 |
+
"problem_type": "single_label_classification",
|
| 44 |
+
"stage_names": [
|
| 45 |
+
"stem",
|
| 46 |
+
"stage1",
|
| 47 |
+
"stage2",
|
| 48 |
+
"stage3",
|
| 49 |
+
"stage4"
|
| 50 |
+
],
|
| 51 |
+
"torch_dtype": "float32",
|
| 52 |
+
"transformers_version": "4.31.0",
|
| 53 |
+
"width_factor": 1
|
| 54 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:caa6a00ab903d697709af2442e051d8636ccdf44c907569aaad3e640d96f322a
|
| 3 |
+
size 188126085
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 448,
|
| 4 |
+
"width": 448
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": true,
|
| 7 |
+
"do_convert_rgb": true,
|
| 8 |
+
"do_normalize": true,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"image_mean": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"image_processor_type": "BitImageProcessor",
|
| 17 |
+
"image_std": [
|
| 18 |
+
0.5,
|
| 19 |
+
0.5,
|
| 20 |
+
0.5
|
| 21 |
+
],
|
| 22 |
+
"resample": 2,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"shortest_edge": 448
|
| 26 |
+
}
|
| 27 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:540bd946cbf8df39e5d8716dbb3953e50c0f14956c8db0ac7c0223dab7585df1
|
| 3 |
+
size 94069425
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62f00a8afa2437c04569329e0b1ef3cc347ac6d475d155806975dad22a1ac07a
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0e6e669f2e3e9ee0ef63d22b9d88fba05e247e5b55055948219b85580f27d4d
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2638 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9184013322231473,
|
| 3 |
+
"best_model_checkpoint": "bit-50-Glaucoma\\checkpoint-3718",
|
| 4 |
+
"epoch": 44.0,
|
| 5 |
+
"global_step": 3718,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.12,
|
| 12 |
+
"learning_rate": 1.1904761904761904e-06,
|
| 13 |
+
"loss": 0.8071,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.24,
|
| 18 |
+
"learning_rate": 2.3809523809523808e-06,
|
| 19 |
+
"loss": 0.7134,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.36,
|
| 24 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 25 |
+
"loss": 0.647,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.47,
|
| 30 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 31 |
+
"loss": 0.547,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.59,
|
| 36 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 37 |
+
"loss": 0.519,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.71,
|
| 42 |
+
"learning_rate": 7.142857142857143e-06,
|
| 43 |
+
"loss": 0.4948,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.83,
|
| 48 |
+
"learning_rate": 8.333333333333334e-06,
|
| 49 |
+
"loss": 0.4692,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.95,
|
| 54 |
+
"learning_rate": 9.523809523809523e-06,
|
| 55 |
+
"loss": 0.4444,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.99,
|
| 60 |
+
"eval_accuracy": 0.813488759367194,
|
| 61 |
+
"eval_loss": 0.4316416382789612,
|
| 62 |
+
"eval_runtime": 16.626,
|
| 63 |
+
"eval_samples_per_second": 72.236,
|
| 64 |
+
"eval_steps_per_second": 2.286,
|
| 65 |
+
"step": 84
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.07,
|
| 69 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 70 |
+
"loss": 0.4251,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.18,
|
| 75 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 76 |
+
"loss": 0.4638,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.3,
|
| 81 |
+
"learning_rate": 1.3095238095238096e-05,
|
| 82 |
+
"loss": 0.4005,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.42,
|
| 87 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 88 |
+
"loss": 0.4005,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.54,
|
| 93 |
+
"learning_rate": 1.5476190476190476e-05,
|
| 94 |
+
"loss": 0.4023,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.66,
|
| 99 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 100 |
+
"loss": 0.3837,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.78,
|
| 105 |
+
"learning_rate": 1.785714285714286e-05,
|
| 106 |
+
"loss": 0.3927,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.89,
|
| 111 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 112 |
+
"loss": 0.3947,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.0,
|
| 117 |
+
"eval_accuracy": 0.8592839300582847,
|
| 118 |
+
"eval_loss": 0.31854528188705444,
|
| 119 |
+
"eval_runtime": 15.8188,
|
| 120 |
+
"eval_samples_per_second": 75.922,
|
| 121 |
+
"eval_steps_per_second": 2.402,
|
| 122 |
+
"step": 169
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.01,
|
| 126 |
+
"learning_rate": 2.023809523809524e-05,
|
| 127 |
+
"loss": 0.3676,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.13,
|
| 132 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 133 |
+
"loss": 0.3767,
|
| 134 |
+
"step": 180
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.25,
|
| 138 |
+
"learning_rate": 2.261904761904762e-05,
|
| 139 |
+
"loss": 0.385,
|
| 140 |
+
"step": 190
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.37,
|
| 144 |
+
"learning_rate": 2.380952380952381e-05,
|
| 145 |
+
"loss": 0.3561,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.49,
|
| 150 |
+
"learning_rate": 2.5e-05,
|
| 151 |
+
"loss": 0.356,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.6,
|
| 156 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 157 |
+
"loss": 0.3683,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.72,
|
| 162 |
+
"learning_rate": 2.7380952380952383e-05,
|
| 163 |
+
"loss": 0.3506,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.84,
|
| 168 |
+
"learning_rate": 2.857142857142857e-05,
|
| 169 |
+
"loss": 0.3569,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.96,
|
| 174 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 175 |
+
"loss": 0.3475,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.99,
|
| 180 |
+
"eval_accuracy": 0.8759367194004996,
|
| 181 |
+
"eval_loss": 0.31600221991539,
|
| 182 |
+
"eval_runtime": 16.0329,
|
| 183 |
+
"eval_samples_per_second": 74.908,
|
| 184 |
+
"eval_steps_per_second": 2.37,
|
| 185 |
+
"step": 253
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.08,
|
| 189 |
+
"learning_rate": 3.095238095238095e-05,
|
| 190 |
+
"loss": 0.3372,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 3.2,
|
| 195 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 196 |
+
"loss": 0.3394,
|
| 197 |
+
"step": 270
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.31,
|
| 201 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 202 |
+
"loss": 0.3077,
|
| 203 |
+
"step": 280
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.43,
|
| 207 |
+
"learning_rate": 3.4523809523809526e-05,
|
| 208 |
+
"loss": 0.3221,
|
| 209 |
+
"step": 290
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.55,
|
| 213 |
+
"learning_rate": 3.571428571428572e-05,
|
| 214 |
+
"loss": 0.3581,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.67,
|
| 219 |
+
"learning_rate": 3.690476190476191e-05,
|
| 220 |
+
"loss": 0.3742,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.79,
|
| 225 |
+
"learning_rate": 3.809523809523809e-05,
|
| 226 |
+
"loss": 0.377,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.91,
|
| 231 |
+
"learning_rate": 3.928571428571429e-05,
|
| 232 |
+
"loss": 0.3796,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 4.0,
|
| 237 |
+
"eval_accuracy": 0.8834304746044963,
|
| 238 |
+
"eval_loss": 0.29224836826324463,
|
| 239 |
+
"eval_runtime": 15.8421,
|
| 240 |
+
"eval_samples_per_second": 75.811,
|
| 241 |
+
"eval_steps_per_second": 2.399,
|
| 242 |
+
"step": 338
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.02,
|
| 246 |
+
"learning_rate": 4.047619047619048e-05,
|
| 247 |
+
"loss": 0.3687,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.14,
|
| 252 |
+
"learning_rate": 4.166666666666667e-05,
|
| 253 |
+
"loss": 0.3326,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.26,
|
| 258 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 259 |
+
"loss": 0.3275,
|
| 260 |
+
"step": 360
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.38,
|
| 264 |
+
"learning_rate": 4.404761904761905e-05,
|
| 265 |
+
"loss": 0.3296,
|
| 266 |
+
"step": 370
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.5,
|
| 270 |
+
"learning_rate": 4.523809523809524e-05,
|
| 271 |
+
"loss": 0.3257,
|
| 272 |
+
"step": 380
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.62,
|
| 276 |
+
"learning_rate": 4.642857142857143e-05,
|
| 277 |
+
"loss": 0.3362,
|
| 278 |
+
"step": 390
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.73,
|
| 282 |
+
"learning_rate": 4.761904761904762e-05,
|
| 283 |
+
"loss": 0.3471,
|
| 284 |
+
"step": 400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.85,
|
| 288 |
+
"learning_rate": 4.880952380952381e-05,
|
| 289 |
+
"loss": 0.3511,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.97,
|
| 294 |
+
"learning_rate": 5e-05,
|
| 295 |
+
"loss": 0.3618,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.99,
|
| 300 |
+
"eval_accuracy": 0.8859283930058285,
|
| 301 |
+
"eval_loss": 0.27965524792671204,
|
| 302 |
+
"eval_runtime": 15.8956,
|
| 303 |
+
"eval_samples_per_second": 75.555,
|
| 304 |
+
"eval_steps_per_second": 2.391,
|
| 305 |
+
"step": 422
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 5.09,
|
| 309 |
+
"learning_rate": 4.986772486772487e-05,
|
| 310 |
+
"loss": 0.3145,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 5.21,
|
| 315 |
+
"learning_rate": 4.973544973544973e-05,
|
| 316 |
+
"loss": 0.3305,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 5.33,
|
| 321 |
+
"learning_rate": 4.960317460317461e-05,
|
| 322 |
+
"loss": 0.3104,
|
| 323 |
+
"step": 450
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 5.44,
|
| 327 |
+
"learning_rate": 4.9470899470899475e-05,
|
| 328 |
+
"loss": 0.3015,
|
| 329 |
+
"step": 460
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.56,
|
| 333 |
+
"learning_rate": 4.933862433862434e-05,
|
| 334 |
+
"loss": 0.3249,
|
| 335 |
+
"step": 470
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 5.68,
|
| 339 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 340 |
+
"loss": 0.2987,
|
| 341 |
+
"step": 480
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.8,
|
| 345 |
+
"learning_rate": 4.9074074074074075e-05,
|
| 346 |
+
"loss": 0.3434,
|
| 347 |
+
"step": 490
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 5.92,
|
| 351 |
+
"learning_rate": 4.894179894179895e-05,
|
| 352 |
+
"loss": 0.3205,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 6.0,
|
| 357 |
+
"eval_accuracy": 0.8409658617818485,
|
| 358 |
+
"eval_loss": 0.3533245027065277,
|
| 359 |
+
"eval_runtime": 15.8631,
|
| 360 |
+
"eval_samples_per_second": 75.71,
|
| 361 |
+
"eval_steps_per_second": 2.395,
|
| 362 |
+
"step": 507
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 6.04,
|
| 366 |
+
"learning_rate": 4.880952380952381e-05,
|
| 367 |
+
"loss": 0.3518,
|
| 368 |
+
"step": 510
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 6.15,
|
| 372 |
+
"learning_rate": 4.8677248677248676e-05,
|
| 373 |
+
"loss": 0.2728,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 6.27,
|
| 378 |
+
"learning_rate": 4.854497354497355e-05,
|
| 379 |
+
"loss": 0.3404,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 6.39,
|
| 384 |
+
"learning_rate": 4.841269841269841e-05,
|
| 385 |
+
"loss": 0.3298,
|
| 386 |
+
"step": 540
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.51,
|
| 390 |
+
"learning_rate": 4.8280423280423284e-05,
|
| 391 |
+
"loss": 0.3178,
|
| 392 |
+
"step": 550
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 6.63,
|
| 396 |
+
"learning_rate": 4.814814814814815e-05,
|
| 397 |
+
"loss": 0.3299,
|
| 398 |
+
"step": 560
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 6.75,
|
| 402 |
+
"learning_rate": 4.801587301587302e-05,
|
| 403 |
+
"loss": 0.3369,
|
| 404 |
+
"step": 570
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 6.86,
|
| 408 |
+
"learning_rate": 4.7883597883597884e-05,
|
| 409 |
+
"loss": 0.2974,
|
| 410 |
+
"step": 580
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 6.98,
|
| 414 |
+
"learning_rate": 4.7751322751322756e-05,
|
| 415 |
+
"loss": 0.2743,
|
| 416 |
+
"step": 590
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 6.99,
|
| 420 |
+
"eval_accuracy": 0.8601165695253955,
|
| 421 |
+
"eval_loss": 0.2907889485359192,
|
| 422 |
+
"eval_runtime": 15.8108,
|
| 423 |
+
"eval_samples_per_second": 75.961,
|
| 424 |
+
"eval_steps_per_second": 2.403,
|
| 425 |
+
"step": 591
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 7.1,
|
| 429 |
+
"learning_rate": 4.761904761904762e-05,
|
| 430 |
+
"loss": 0.2966,
|
| 431 |
+
"step": 600
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 7.22,
|
| 435 |
+
"learning_rate": 4.748677248677249e-05,
|
| 436 |
+
"loss": 0.3083,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 7.34,
|
| 441 |
+
"learning_rate": 4.7354497354497356e-05,
|
| 442 |
+
"loss": 0.299,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.46,
|
| 447 |
+
"learning_rate": 4.722222222222222e-05,
|
| 448 |
+
"loss": 0.2669,
|
| 449 |
+
"step": 630
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 7.57,
|
| 453 |
+
"learning_rate": 4.708994708994709e-05,
|
| 454 |
+
"loss": 0.2996,
|
| 455 |
+
"step": 640
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 7.69,
|
| 459 |
+
"learning_rate": 4.6957671957671964e-05,
|
| 460 |
+
"loss": 0.2686,
|
| 461 |
+
"step": 650
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 7.81,
|
| 465 |
+
"learning_rate": 4.682539682539683e-05,
|
| 466 |
+
"loss": 0.3172,
|
| 467 |
+
"step": 660
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 7.93,
|
| 471 |
+
"learning_rate": 4.669312169312169e-05,
|
| 472 |
+
"loss": 0.2765,
|
| 473 |
+
"step": 670
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.0,
|
| 477 |
+
"eval_accuracy": 0.8601165695253955,
|
| 478 |
+
"eval_loss": 0.2945786714553833,
|
| 479 |
+
"eval_runtime": 16.0316,
|
| 480 |
+
"eval_samples_per_second": 74.914,
|
| 481 |
+
"eval_steps_per_second": 2.37,
|
| 482 |
+
"step": 676
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 8.05,
|
| 486 |
+
"learning_rate": 4.656084656084656e-05,
|
| 487 |
+
"loss": 0.2692,
|
| 488 |
+
"step": 680
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 8.17,
|
| 492 |
+
"learning_rate": 4.642857142857143e-05,
|
| 493 |
+
"loss": 0.273,
|
| 494 |
+
"step": 690
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.28,
|
| 498 |
+
"learning_rate": 4.62962962962963e-05,
|
| 499 |
+
"loss": 0.252,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.4,
|
| 504 |
+
"learning_rate": 4.6164021164021165e-05,
|
| 505 |
+
"loss": 0.2719,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 8.52,
|
| 510 |
+
"learning_rate": 4.603174603174603e-05,
|
| 511 |
+
"loss": 0.273,
|
| 512 |
+
"step": 720
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 8.64,
|
| 516 |
+
"learning_rate": 4.58994708994709e-05,
|
| 517 |
+
"loss": 0.2853,
|
| 518 |
+
"step": 730
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 8.76,
|
| 522 |
+
"learning_rate": 4.576719576719577e-05,
|
| 523 |
+
"loss": 0.2965,
|
| 524 |
+
"step": 740
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 8.88,
|
| 528 |
+
"learning_rate": 4.563492063492064e-05,
|
| 529 |
+
"loss": 0.2842,
|
| 530 |
+
"step": 750
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 8.99,
|
| 534 |
+
"learning_rate": 4.55026455026455e-05,
|
| 535 |
+
"loss": 0.2528,
|
| 536 |
+
"step": 760
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 8.99,
|
| 540 |
+
"eval_accuracy": 0.8992506244796004,
|
| 541 |
+
"eval_loss": 0.24736784398555756,
|
| 542 |
+
"eval_runtime": 15.802,
|
| 543 |
+
"eval_samples_per_second": 76.003,
|
| 544 |
+
"eval_steps_per_second": 2.405,
|
| 545 |
+
"step": 760
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 9.11,
|
| 549 |
+
"learning_rate": 4.5370370370370374e-05,
|
| 550 |
+
"loss": 0.2585,
|
| 551 |
+
"step": 770
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 9.23,
|
| 555 |
+
"learning_rate": 4.523809523809524e-05,
|
| 556 |
+
"loss": 0.2567,
|
| 557 |
+
"step": 780
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 9.35,
|
| 561 |
+
"learning_rate": 4.510582010582011e-05,
|
| 562 |
+
"loss": 0.2794,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 9.47,
|
| 567 |
+
"learning_rate": 4.4973544973544974e-05,
|
| 568 |
+
"loss": 0.2698,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 9.59,
|
| 573 |
+
"learning_rate": 4.4841269841269846e-05,
|
| 574 |
+
"loss": 0.2472,
|
| 575 |
+
"step": 810
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 9.7,
|
| 579 |
+
"learning_rate": 4.470899470899471e-05,
|
| 580 |
+
"loss": 0.2537,
|
| 581 |
+
"step": 820
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 9.82,
|
| 585 |
+
"learning_rate": 4.4576719576719575e-05,
|
| 586 |
+
"loss": 0.2635,
|
| 587 |
+
"step": 830
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 9.94,
|
| 591 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 592 |
+
"loss": 0.2678,
|
| 593 |
+
"step": 840
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 10.0,
|
| 597 |
+
"eval_accuracy": 0.9084096586178185,
|
| 598 |
+
"eval_loss": 0.2332490235567093,
|
| 599 |
+
"eval_runtime": 15.7703,
|
| 600 |
+
"eval_samples_per_second": 76.156,
|
| 601 |
+
"eval_steps_per_second": 2.41,
|
| 602 |
+
"step": 845
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 10.06,
|
| 606 |
+
"learning_rate": 4.431216931216932e-05,
|
| 607 |
+
"loss": 0.2777,
|
| 608 |
+
"step": 850
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 10.18,
|
| 612 |
+
"learning_rate": 4.417989417989418e-05,
|
| 613 |
+
"loss": 0.2543,
|
| 614 |
+
"step": 860
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 10.3,
|
| 618 |
+
"learning_rate": 4.404761904761905e-05,
|
| 619 |
+
"loss": 0.2434,
|
| 620 |
+
"step": 870
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 10.41,
|
| 624 |
+
"learning_rate": 4.391534391534391e-05,
|
| 625 |
+
"loss": 0.2629,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 10.53,
|
| 630 |
+
"learning_rate": 4.378306878306879e-05,
|
| 631 |
+
"loss": 0.2453,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 10.65,
|
| 636 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 637 |
+
"loss": 0.266,
|
| 638 |
+
"step": 900
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 10.77,
|
| 642 |
+
"learning_rate": 4.351851851851852e-05,
|
| 643 |
+
"loss": 0.2349,
|
| 644 |
+
"step": 910
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 10.89,
|
| 648 |
+
"learning_rate": 4.3386243386243384e-05,
|
| 649 |
+
"loss": 0.2314,
|
| 650 |
+
"step": 920
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 10.99,
|
| 654 |
+
"eval_accuracy": 0.8992506244796004,
|
| 655 |
+
"eval_loss": 0.2373119592666626,
|
| 656 |
+
"eval_runtime": 15.6414,
|
| 657 |
+
"eval_samples_per_second": 76.783,
|
| 658 |
+
"eval_steps_per_second": 2.429,
|
| 659 |
+
"step": 929
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 11.01,
|
| 663 |
+
"learning_rate": 4.3253968253968256e-05,
|
| 664 |
+
"loss": 0.2436,
|
| 665 |
+
"step": 930
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 11.12,
|
| 669 |
+
"learning_rate": 4.312169312169313e-05,
|
| 670 |
+
"loss": 0.2576,
|
| 671 |
+
"step": 940
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 11.24,
|
| 675 |
+
"learning_rate": 4.298941798941799e-05,
|
| 676 |
+
"loss": 0.2982,
|
| 677 |
+
"step": 950
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 11.36,
|
| 681 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 682 |
+
"loss": 0.2399,
|
| 683 |
+
"step": 960
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 11.48,
|
| 687 |
+
"learning_rate": 4.272486772486773e-05,
|
| 688 |
+
"loss": 0.2442,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 11.6,
|
| 693 |
+
"learning_rate": 4.259259259259259e-05,
|
| 694 |
+
"loss": 0.2629,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 11.72,
|
| 699 |
+
"learning_rate": 4.2460317460317464e-05,
|
| 700 |
+
"loss": 0.2484,
|
| 701 |
+
"step": 990
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 11.83,
|
| 705 |
+
"learning_rate": 4.232804232804233e-05,
|
| 706 |
+
"loss": 0.2462,
|
| 707 |
+
"step": 1000
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 11.95,
|
| 711 |
+
"learning_rate": 4.21957671957672e-05,
|
| 712 |
+
"loss": 0.244,
|
| 713 |
+
"step": 1010
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 12.0,
|
| 717 |
+
"eval_accuracy": 0.9034138218151541,
|
| 718 |
+
"eval_loss": 0.2284502238035202,
|
| 719 |
+
"eval_runtime": 15.8599,
|
| 720 |
+
"eval_samples_per_second": 75.726,
|
| 721 |
+
"eval_steps_per_second": 2.396,
|
| 722 |
+
"step": 1014
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 12.07,
|
| 726 |
+
"learning_rate": 4.2063492063492065e-05,
|
| 727 |
+
"loss": 0.2496,
|
| 728 |
+
"step": 1020
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 12.19,
|
| 732 |
+
"learning_rate": 4.193121693121693e-05,
|
| 733 |
+
"loss": 0.2235,
|
| 734 |
+
"step": 1030
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 12.31,
|
| 738 |
+
"learning_rate": 4.17989417989418e-05,
|
| 739 |
+
"loss": 0.2194,
|
| 740 |
+
"step": 1040
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 12.43,
|
| 744 |
+
"learning_rate": 4.166666666666667e-05,
|
| 745 |
+
"loss": 0.2358,
|
| 746 |
+
"step": 1050
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 12.54,
|
| 750 |
+
"learning_rate": 4.153439153439154e-05,
|
| 751 |
+
"loss": 0.2447,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 12.66,
|
| 756 |
+
"learning_rate": 4.14021164021164e-05,
|
| 757 |
+
"loss": 0.2537,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 12.78,
|
| 762 |
+
"learning_rate": 4.126984126984127e-05,
|
| 763 |
+
"loss": 0.2246,
|
| 764 |
+
"step": 1080
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 12.9,
|
| 768 |
+
"learning_rate": 4.1137566137566144e-05,
|
| 769 |
+
"loss": 0.2313,
|
| 770 |
+
"step": 1090
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 12.99,
|
| 774 |
+
"eval_accuracy": 0.9017485428809325,
|
| 775 |
+
"eval_loss": 0.2355087250471115,
|
| 776 |
+
"eval_runtime": 15.8512,
|
| 777 |
+
"eval_samples_per_second": 75.767,
|
| 778 |
+
"eval_steps_per_second": 2.397,
|
| 779 |
+
"step": 1098
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 13.02,
|
| 783 |
+
"learning_rate": 4.100529100529101e-05,
|
| 784 |
+
"loss": 0.2217,
|
| 785 |
+
"step": 1100
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 13.14,
|
| 789 |
+
"learning_rate": 4.0873015873015874e-05,
|
| 790 |
+
"loss": 0.196,
|
| 791 |
+
"step": 1110
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 13.25,
|
| 795 |
+
"learning_rate": 4.074074074074074e-05,
|
| 796 |
+
"loss": 0.2298,
|
| 797 |
+
"step": 1120
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 13.37,
|
| 801 |
+
"learning_rate": 4.060846560846561e-05,
|
| 802 |
+
"loss": 0.2352,
|
| 803 |
+
"step": 1130
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 13.49,
|
| 807 |
+
"learning_rate": 4.047619047619048e-05,
|
| 808 |
+
"loss": 0.2582,
|
| 809 |
+
"step": 1140
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 13.61,
|
| 813 |
+
"learning_rate": 4.0343915343915346e-05,
|
| 814 |
+
"loss": 0.2168,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 13.73,
|
| 819 |
+
"learning_rate": 4.021164021164021e-05,
|
| 820 |
+
"loss": 0.2278,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 13.85,
|
| 825 |
+
"learning_rate": 4.007936507936508e-05,
|
| 826 |
+
"loss": 0.2511,
|
| 827 |
+
"step": 1170
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.96,
|
| 831 |
+
"learning_rate": 3.9947089947089946e-05,
|
| 832 |
+
"loss": 0.2055,
|
| 833 |
+
"step": 1180
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 14.0,
|
| 837 |
+
"eval_accuracy": 0.8975853455453788,
|
| 838 |
+
"eval_loss": 0.2440473437309265,
|
| 839 |
+
"eval_runtime": 15.5821,
|
| 840 |
+
"eval_samples_per_second": 77.075,
|
| 841 |
+
"eval_steps_per_second": 2.439,
|
| 842 |
+
"step": 1183
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 14.08,
|
| 846 |
+
"learning_rate": 3.981481481481482e-05,
|
| 847 |
+
"loss": 0.2228,
|
| 848 |
+
"step": 1190
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 14.2,
|
| 852 |
+
"learning_rate": 3.968253968253968e-05,
|
| 853 |
+
"loss": 0.2118,
|
| 854 |
+
"step": 1200
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 14.32,
|
| 858 |
+
"learning_rate": 3.9550264550264554e-05,
|
| 859 |
+
"loss": 0.2281,
|
| 860 |
+
"step": 1210
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 14.44,
|
| 864 |
+
"learning_rate": 3.941798941798942e-05,
|
| 865 |
+
"loss": 0.2107,
|
| 866 |
+
"step": 1220
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 14.56,
|
| 870 |
+
"learning_rate": 3.928571428571429e-05,
|
| 871 |
+
"loss": 0.2133,
|
| 872 |
+
"step": 1230
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 14.67,
|
| 876 |
+
"learning_rate": 3.9153439153439155e-05,
|
| 877 |
+
"loss": 0.2271,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 14.79,
|
| 882 |
+
"learning_rate": 3.9021164021164026e-05,
|
| 883 |
+
"loss": 0.2252,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 14.91,
|
| 888 |
+
"learning_rate": 3.888888888888889e-05,
|
| 889 |
+
"loss": 0.2271,
|
| 890 |
+
"step": 1260
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 14.99,
|
| 894 |
+
"eval_accuracy": 0.8884263114071607,
|
| 895 |
+
"eval_loss": 0.2461342066526413,
|
| 896 |
+
"eval_runtime": 15.57,
|
| 897 |
+
"eval_samples_per_second": 77.136,
|
| 898 |
+
"eval_steps_per_second": 2.441,
|
| 899 |
+
"step": 1267
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 15.03,
|
| 903 |
+
"learning_rate": 3.8756613756613755e-05,
|
| 904 |
+
"loss": 0.223,
|
| 905 |
+
"step": 1270
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 15.15,
|
| 909 |
+
"learning_rate": 3.862433862433863e-05,
|
| 910 |
+
"loss": 0.1974,
|
| 911 |
+
"step": 1280
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 15.27,
|
| 915 |
+
"learning_rate": 3.84920634920635e-05,
|
| 916 |
+
"loss": 0.2514,
|
| 917 |
+
"step": 1290
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 15.38,
|
| 921 |
+
"learning_rate": 3.835978835978836e-05,
|
| 922 |
+
"loss": 0.2005,
|
| 923 |
+
"step": 1300
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 15.5,
|
| 927 |
+
"learning_rate": 3.822751322751323e-05,
|
| 928 |
+
"loss": 0.2014,
|
| 929 |
+
"step": 1310
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 15.62,
|
| 933 |
+
"learning_rate": 3.809523809523809e-05,
|
| 934 |
+
"loss": 0.1952,
|
| 935 |
+
"step": 1320
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 15.74,
|
| 939 |
+
"learning_rate": 3.7962962962962964e-05,
|
| 940 |
+
"loss": 0.2002,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 15.86,
|
| 945 |
+
"learning_rate": 3.7830687830687835e-05,
|
| 946 |
+
"loss": 0.2351,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 15.98,
|
| 951 |
+
"learning_rate": 3.76984126984127e-05,
|
| 952 |
+
"loss": 0.2106,
|
| 953 |
+
"step": 1350
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 16.0,
|
| 957 |
+
"eval_accuracy": 0.8784346378018318,
|
| 958 |
+
"eval_loss": 0.26777663826942444,
|
| 959 |
+
"eval_runtime": 15.6876,
|
| 960 |
+
"eval_samples_per_second": 76.557,
|
| 961 |
+
"eval_steps_per_second": 2.422,
|
| 962 |
+
"step": 1352
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 16.09,
|
| 966 |
+
"learning_rate": 3.7566137566137564e-05,
|
| 967 |
+
"loss": 0.2066,
|
| 968 |
+
"step": 1360
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 16.21,
|
| 972 |
+
"learning_rate": 3.7433862433862436e-05,
|
| 973 |
+
"loss": 0.2009,
|
| 974 |
+
"step": 1370
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 16.33,
|
| 978 |
+
"learning_rate": 3.730158730158731e-05,
|
| 979 |
+
"loss": 0.2162,
|
| 980 |
+
"step": 1380
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 16.45,
|
| 984 |
+
"learning_rate": 3.716931216931217e-05,
|
| 985 |
+
"loss": 0.2155,
|
| 986 |
+
"step": 1390
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 16.57,
|
| 990 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 991 |
+
"loss": 0.1995,
|
| 992 |
+
"step": 1400
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 16.69,
|
| 996 |
+
"learning_rate": 3.690476190476191e-05,
|
| 997 |
+
"loss": 0.1992,
|
| 998 |
+
"step": 1410
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 16.8,
|
| 1002 |
+
"learning_rate": 3.677248677248677e-05,
|
| 1003 |
+
"loss": 0.2366,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 16.92,
|
| 1008 |
+
"learning_rate": 3.6640211640211644e-05,
|
| 1009 |
+
"loss": 0.2076,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 16.99,
|
| 1014 |
+
"eval_accuracy": 0.9150707743547044,
|
| 1015 |
+
"eval_loss": 0.22340074181556702,
|
| 1016 |
+
"eval_runtime": 16.0204,
|
| 1017 |
+
"eval_samples_per_second": 74.967,
|
| 1018 |
+
"eval_steps_per_second": 2.372,
|
| 1019 |
+
"step": 1436
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 17.04,
|
| 1023 |
+
"learning_rate": 3.650793650793651e-05,
|
| 1024 |
+
"loss": 0.2054,
|
| 1025 |
+
"step": 1440
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 17.16,
|
| 1029 |
+
"learning_rate": 3.637566137566138e-05,
|
| 1030 |
+
"loss": 0.1922,
|
| 1031 |
+
"step": 1450
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 17.28,
|
| 1035 |
+
"learning_rate": 3.6243386243386245e-05,
|
| 1036 |
+
"loss": 0.1768,
|
| 1037 |
+
"step": 1460
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 17.4,
|
| 1041 |
+
"learning_rate": 3.611111111111111e-05,
|
| 1042 |
+
"loss": 0.2221,
|
| 1043 |
+
"step": 1470
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 17.51,
|
| 1047 |
+
"learning_rate": 3.597883597883598e-05,
|
| 1048 |
+
"loss": 0.1813,
|
| 1049 |
+
"step": 1480
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 17.63,
|
| 1053 |
+
"learning_rate": 3.584656084656085e-05,
|
| 1054 |
+
"loss": 0.2035,
|
| 1055 |
+
"step": 1490
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 17.75,
|
| 1059 |
+
"learning_rate": 3.571428571428572e-05,
|
| 1060 |
+
"loss": 0.2015,
|
| 1061 |
+
"step": 1500
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 17.87,
|
| 1065 |
+
"learning_rate": 3.558201058201058e-05,
|
| 1066 |
+
"loss": 0.2246,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 17.99,
|
| 1071 |
+
"learning_rate": 3.5449735449735446e-05,
|
| 1072 |
+
"loss": 0.2074,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 18.0,
|
| 1077 |
+
"eval_accuracy": 0.8934221482098251,
|
| 1078 |
+
"eval_loss": 0.24407817423343658,
|
| 1079 |
+
"eval_runtime": 15.7214,
|
| 1080 |
+
"eval_samples_per_second": 76.393,
|
| 1081 |
+
"eval_steps_per_second": 2.417,
|
| 1082 |
+
"step": 1521
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 18.11,
|
| 1086 |
+
"learning_rate": 3.5317460317460324e-05,
|
| 1087 |
+
"loss": 0.208,
|
| 1088 |
+
"step": 1530
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 18.22,
|
| 1092 |
+
"learning_rate": 3.518518518518519e-05,
|
| 1093 |
+
"loss": 0.1872,
|
| 1094 |
+
"step": 1540
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 18.34,
|
| 1098 |
+
"learning_rate": 3.5052910052910054e-05,
|
| 1099 |
+
"loss": 0.2111,
|
| 1100 |
+
"step": 1550
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 18.46,
|
| 1104 |
+
"learning_rate": 3.492063492063492e-05,
|
| 1105 |
+
"loss": 0.1953,
|
| 1106 |
+
"step": 1560
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 18.58,
|
| 1110 |
+
"learning_rate": 3.478835978835979e-05,
|
| 1111 |
+
"loss": 0.1802,
|
| 1112 |
+
"step": 1570
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 18.7,
|
| 1116 |
+
"learning_rate": 3.465608465608466e-05,
|
| 1117 |
+
"loss": 0.2181,
|
| 1118 |
+
"step": 1580
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 18.82,
|
| 1122 |
+
"learning_rate": 3.4523809523809526e-05,
|
| 1123 |
+
"loss": 0.2199,
|
| 1124 |
+
"step": 1590
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 18.93,
|
| 1128 |
+
"learning_rate": 3.439153439153439e-05,
|
| 1129 |
+
"loss": 0.2043,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 18.99,
|
| 1134 |
+
"eval_accuracy": 0.9000832639467111,
|
| 1135 |
+
"eval_loss": 0.2535579800605774,
|
| 1136 |
+
"eval_runtime": 15.8014,
|
| 1137 |
+
"eval_samples_per_second": 76.006,
|
| 1138 |
+
"eval_steps_per_second": 2.405,
|
| 1139 |
+
"step": 1605
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 19.05,
|
| 1143 |
+
"learning_rate": 3.425925925925926e-05,
|
| 1144 |
+
"loss": 0.1837,
|
| 1145 |
+
"step": 1610
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 19.17,
|
| 1149 |
+
"learning_rate": 3.412698412698413e-05,
|
| 1150 |
+
"loss": 0.1705,
|
| 1151 |
+
"step": 1620
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 19.29,
|
| 1155 |
+
"learning_rate": 3.3994708994709e-05,
|
| 1156 |
+
"loss": 0.1925,
|
| 1157 |
+
"step": 1630
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 19.41,
|
| 1161 |
+
"learning_rate": 3.386243386243386e-05,
|
| 1162 |
+
"loss": 0.1857,
|
| 1163 |
+
"step": 1640
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 19.53,
|
| 1167 |
+
"learning_rate": 3.3730158730158734e-05,
|
| 1168 |
+
"loss": 0.181,
|
| 1169 |
+
"step": 1650
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 19.64,
|
| 1173 |
+
"learning_rate": 3.35978835978836e-05,
|
| 1174 |
+
"loss": 0.2066,
|
| 1175 |
+
"step": 1660
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 19.76,
|
| 1179 |
+
"learning_rate": 3.3465608465608464e-05,
|
| 1180 |
+
"loss": 0.2078,
|
| 1181 |
+
"step": 1670
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 19.88,
|
| 1185 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 1186 |
+
"loss": 0.2028,
|
| 1187 |
+
"step": 1680
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 20.0,
|
| 1191 |
+
"learning_rate": 3.3201058201058206e-05,
|
| 1192 |
+
"loss": 0.2058,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 20.0,
|
| 1197 |
+
"eval_accuracy": 0.8834304746044963,
|
| 1198 |
+
"eval_loss": 0.2481454610824585,
|
| 1199 |
+
"eval_runtime": 16.3173,
|
| 1200 |
+
"eval_samples_per_second": 73.603,
|
| 1201 |
+
"eval_steps_per_second": 2.329,
|
| 1202 |
+
"step": 1690
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 20.12,
|
| 1206 |
+
"learning_rate": 3.306878306878307e-05,
|
| 1207 |
+
"loss": 0.1715,
|
| 1208 |
+
"step": 1700
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 20.24,
|
| 1212 |
+
"learning_rate": 3.2936507936507936e-05,
|
| 1213 |
+
"loss": 0.1735,
|
| 1214 |
+
"step": 1710
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 20.36,
|
| 1218 |
+
"learning_rate": 3.280423280423281e-05,
|
| 1219 |
+
"loss": 0.1703,
|
| 1220 |
+
"step": 1720
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 20.47,
|
| 1224 |
+
"learning_rate": 3.267195767195768e-05,
|
| 1225 |
+
"loss": 0.1718,
|
| 1226 |
+
"step": 1730
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 20.59,
|
| 1230 |
+
"learning_rate": 3.253968253968254e-05,
|
| 1231 |
+
"loss": 0.1817,
|
| 1232 |
+
"step": 1740
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 20.71,
|
| 1236 |
+
"learning_rate": 3.240740740740741e-05,
|
| 1237 |
+
"loss": 0.1959,
|
| 1238 |
+
"step": 1750
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 20.83,
|
| 1242 |
+
"learning_rate": 3.227513227513227e-05,
|
| 1243 |
+
"loss": 0.1753,
|
| 1244 |
+
"step": 1760
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 20.95,
|
| 1248 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 1249 |
+
"loss": 0.1575,
|
| 1250 |
+
"step": 1770
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 20.99,
|
| 1254 |
+
"eval_accuracy": 0.9025811823480433,
|
| 1255 |
+
"eval_loss": 0.2368643879890442,
|
| 1256 |
+
"eval_runtime": 15.885,
|
| 1257 |
+
"eval_samples_per_second": 75.606,
|
| 1258 |
+
"eval_steps_per_second": 2.392,
|
| 1259 |
+
"step": 1774
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 21.07,
|
| 1263 |
+
"learning_rate": 3.2010582010582015e-05,
|
| 1264 |
+
"loss": 0.1754,
|
| 1265 |
+
"step": 1780
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 21.18,
|
| 1269 |
+
"learning_rate": 3.187830687830688e-05,
|
| 1270 |
+
"loss": 0.168,
|
| 1271 |
+
"step": 1790
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 21.3,
|
| 1275 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 1276 |
+
"loss": 0.1746,
|
| 1277 |
+
"step": 1800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 21.42,
|
| 1281 |
+
"learning_rate": 3.1613756613756616e-05,
|
| 1282 |
+
"loss": 0.1883,
|
| 1283 |
+
"step": 1810
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 21.54,
|
| 1287 |
+
"learning_rate": 3.148148148148148e-05,
|
| 1288 |
+
"loss": 0.1732,
|
| 1289 |
+
"step": 1820
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 21.66,
|
| 1293 |
+
"learning_rate": 3.134920634920635e-05,
|
| 1294 |
+
"loss": 0.147,
|
| 1295 |
+
"step": 1830
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 21.78,
|
| 1299 |
+
"learning_rate": 3.121693121693122e-05,
|
| 1300 |
+
"loss": 0.1961,
|
| 1301 |
+
"step": 1840
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 21.89,
|
| 1305 |
+
"learning_rate": 3.108465608465609e-05,
|
| 1306 |
+
"loss": 0.1889,
|
| 1307 |
+
"step": 1850
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 22.0,
|
| 1311 |
+
"eval_accuracy": 0.9117402164862615,
|
| 1312 |
+
"eval_loss": 0.2264406383037567,
|
| 1313 |
+
"eval_runtime": 15.8707,
|
| 1314 |
+
"eval_samples_per_second": 75.674,
|
| 1315 |
+
"eval_steps_per_second": 2.394,
|
| 1316 |
+
"step": 1859
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 22.01,
|
| 1320 |
+
"learning_rate": 3.095238095238095e-05,
|
| 1321 |
+
"loss": 0.1844,
|
| 1322 |
+
"step": 1860
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 22.13,
|
| 1326 |
+
"learning_rate": 3.0820105820105824e-05,
|
| 1327 |
+
"loss": 0.175,
|
| 1328 |
+
"step": 1870
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 22.25,
|
| 1332 |
+
"learning_rate": 3.068783068783069e-05,
|
| 1333 |
+
"loss": 0.1638,
|
| 1334 |
+
"step": 1880
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 22.37,
|
| 1338 |
+
"learning_rate": 3.055555555555556e-05,
|
| 1339 |
+
"loss": 0.1828,
|
| 1340 |
+
"step": 1890
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 22.49,
|
| 1344 |
+
"learning_rate": 3.0423280423280425e-05,
|
| 1345 |
+
"loss": 0.1869,
|
| 1346 |
+
"step": 1900
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 22.6,
|
| 1350 |
+
"learning_rate": 3.0291005291005293e-05,
|
| 1351 |
+
"loss": 0.1354,
|
| 1352 |
+
"step": 1910
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 22.72,
|
| 1356 |
+
"learning_rate": 3.0158730158730158e-05,
|
| 1357 |
+
"loss": 0.1544,
|
| 1358 |
+
"step": 1920
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 22.84,
|
| 1362 |
+
"learning_rate": 3.002645502645503e-05,
|
| 1363 |
+
"loss": 0.1607,
|
| 1364 |
+
"step": 1930
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 22.96,
|
| 1368 |
+
"learning_rate": 2.9894179894179897e-05,
|
| 1369 |
+
"loss": 0.1797,
|
| 1370 |
+
"step": 1940
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 22.99,
|
| 1374 |
+
"eval_accuracy": 0.9084096586178185,
|
| 1375 |
+
"eval_loss": 0.23159563541412354,
|
| 1376 |
+
"eval_runtime": 15.8807,
|
| 1377 |
+
"eval_samples_per_second": 75.626,
|
| 1378 |
+
"eval_steps_per_second": 2.393,
|
| 1379 |
+
"step": 1943
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 23.08,
|
| 1383 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 1384 |
+
"loss": 0.164,
|
| 1385 |
+
"step": 1950
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 23.2,
|
| 1389 |
+
"learning_rate": 2.962962962962963e-05,
|
| 1390 |
+
"loss": 0.1444,
|
| 1391 |
+
"step": 1960
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 23.31,
|
| 1395 |
+
"learning_rate": 2.94973544973545e-05,
|
| 1396 |
+
"loss": 0.1518,
|
| 1397 |
+
"step": 1970
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 23.43,
|
| 1401 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 1402 |
+
"loss": 0.1498,
|
| 1403 |
+
"step": 1980
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 23.55,
|
| 1407 |
+
"learning_rate": 2.9232804232804234e-05,
|
| 1408 |
+
"loss": 0.1624,
|
| 1409 |
+
"step": 1990
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 23.67,
|
| 1413 |
+
"learning_rate": 2.91005291005291e-05,
|
| 1414 |
+
"loss": 0.1367,
|
| 1415 |
+
"step": 2000
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 23.79,
|
| 1419 |
+
"learning_rate": 2.8968253968253974e-05,
|
| 1420 |
+
"loss": 0.1645,
|
| 1421 |
+
"step": 2010
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 23.91,
|
| 1425 |
+
"learning_rate": 2.8835978835978838e-05,
|
| 1426 |
+
"loss": 0.1729,
|
| 1427 |
+
"step": 2020
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 24.0,
|
| 1431 |
+
"eval_accuracy": 0.9042464612822648,
|
| 1432 |
+
"eval_loss": 0.24072496592998505,
|
| 1433 |
+
"eval_runtime": 15.9775,
|
| 1434 |
+
"eval_samples_per_second": 75.168,
|
| 1435 |
+
"eval_steps_per_second": 2.378,
|
| 1436 |
+
"step": 2028
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 24.02,
|
| 1440 |
+
"learning_rate": 2.8703703703703706e-05,
|
| 1441 |
+
"loss": 0.1603,
|
| 1442 |
+
"step": 2030
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 24.14,
|
| 1446 |
+
"learning_rate": 2.857142857142857e-05,
|
| 1447 |
+
"loss": 0.1448,
|
| 1448 |
+
"step": 2040
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 24.26,
|
| 1452 |
+
"learning_rate": 2.8439153439153442e-05,
|
| 1453 |
+
"loss": 0.1389,
|
| 1454 |
+
"step": 2050
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 24.38,
|
| 1458 |
+
"learning_rate": 2.830687830687831e-05,
|
| 1459 |
+
"loss": 0.1612,
|
| 1460 |
+
"step": 2060
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 24.5,
|
| 1464 |
+
"learning_rate": 2.8174603174603175e-05,
|
| 1465 |
+
"loss": 0.1429,
|
| 1466 |
+
"step": 2070
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 24.62,
|
| 1470 |
+
"learning_rate": 2.8042328042328043e-05,
|
| 1471 |
+
"loss": 0.1498,
|
| 1472 |
+
"step": 2080
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 24.73,
|
| 1476 |
+
"learning_rate": 2.7910052910052914e-05,
|
| 1477 |
+
"loss": 0.1699,
|
| 1478 |
+
"step": 2090
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 24.85,
|
| 1482 |
+
"learning_rate": 2.777777777777778e-05,
|
| 1483 |
+
"loss": 0.1409,
|
| 1484 |
+
"step": 2100
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 24.97,
|
| 1488 |
+
"learning_rate": 2.7645502645502647e-05,
|
| 1489 |
+
"loss": 0.163,
|
| 1490 |
+
"step": 2110
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 24.99,
|
| 1494 |
+
"eval_accuracy": 0.9009159034138218,
|
| 1495 |
+
"eval_loss": 0.27579963207244873,
|
| 1496 |
+
"eval_runtime": 15.7181,
|
| 1497 |
+
"eval_samples_per_second": 76.409,
|
| 1498 |
+
"eval_steps_per_second": 2.418,
|
| 1499 |
+
"step": 2112
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 25.09,
|
| 1503 |
+
"learning_rate": 2.7513227513227512e-05,
|
| 1504 |
+
"loss": 0.1668,
|
| 1505 |
+
"step": 2120
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 25.21,
|
| 1509 |
+
"learning_rate": 2.7380952380952383e-05,
|
| 1510 |
+
"loss": 0.1529,
|
| 1511 |
+
"step": 2130
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 25.33,
|
| 1515 |
+
"learning_rate": 2.724867724867725e-05,
|
| 1516 |
+
"loss": 0.1581,
|
| 1517 |
+
"step": 2140
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 25.44,
|
| 1521 |
+
"learning_rate": 2.7116402116402116e-05,
|
| 1522 |
+
"loss": 0.1408,
|
| 1523 |
+
"step": 2150
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 25.56,
|
| 1527 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 1528 |
+
"loss": 0.1463,
|
| 1529 |
+
"step": 2160
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 25.68,
|
| 1533 |
+
"learning_rate": 2.6851851851851855e-05,
|
| 1534 |
+
"loss": 0.1431,
|
| 1535 |
+
"step": 2170
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 25.8,
|
| 1539 |
+
"learning_rate": 2.6719576719576723e-05,
|
| 1540 |
+
"loss": 0.1397,
|
| 1541 |
+
"step": 2180
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 25.92,
|
| 1545 |
+
"learning_rate": 2.6587301587301588e-05,
|
| 1546 |
+
"loss": 0.1609,
|
| 1547 |
+
"step": 2190
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 26.0,
|
| 1551 |
+
"eval_accuracy": 0.8934221482098251,
|
| 1552 |
+
"eval_loss": 0.25153836607933044,
|
| 1553 |
+
"eval_runtime": 15.7973,
|
| 1554 |
+
"eval_samples_per_second": 76.026,
|
| 1555 |
+
"eval_steps_per_second": 2.405,
|
| 1556 |
+
"step": 2197
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 26.04,
|
| 1560 |
+
"learning_rate": 2.6455026455026456e-05,
|
| 1561 |
+
"loss": 0.1503,
|
| 1562 |
+
"step": 2200
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 26.15,
|
| 1566 |
+
"learning_rate": 2.6322751322751328e-05,
|
| 1567 |
+
"loss": 0.141,
|
| 1568 |
+
"step": 2210
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 26.27,
|
| 1572 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 1573 |
+
"loss": 0.1384,
|
| 1574 |
+
"step": 2220
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 26.39,
|
| 1578 |
+
"learning_rate": 2.605820105820106e-05,
|
| 1579 |
+
"loss": 0.1505,
|
| 1580 |
+
"step": 2230
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 26.51,
|
| 1584 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 1585 |
+
"loss": 0.153,
|
| 1586 |
+
"step": 2240
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 26.63,
|
| 1590 |
+
"learning_rate": 2.5793650793650796e-05,
|
| 1591 |
+
"loss": 0.1367,
|
| 1592 |
+
"step": 2250
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 26.75,
|
| 1596 |
+
"learning_rate": 2.5661375661375664e-05,
|
| 1597 |
+
"loss": 0.1587,
|
| 1598 |
+
"step": 2260
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 26.86,
|
| 1602 |
+
"learning_rate": 2.552910052910053e-05,
|
| 1603 |
+
"loss": 0.1493,
|
| 1604 |
+
"step": 2270
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 26.98,
|
| 1608 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 1609 |
+
"loss": 0.1345,
|
| 1610 |
+
"step": 2280
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 26.99,
|
| 1614 |
+
"eval_accuracy": 0.8950874271440467,
|
| 1615 |
+
"eval_loss": 0.2521910071372986,
|
| 1616 |
+
"eval_runtime": 15.9043,
|
| 1617 |
+
"eval_samples_per_second": 75.514,
|
| 1618 |
+
"eval_steps_per_second": 2.389,
|
| 1619 |
+
"step": 2281
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 27.1,
|
| 1623 |
+
"learning_rate": 2.526455026455027e-05,
|
| 1624 |
+
"loss": 0.1349,
|
| 1625 |
+
"step": 2290
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 27.22,
|
| 1629 |
+
"learning_rate": 2.5132275132275137e-05,
|
| 1630 |
+
"loss": 0.1301,
|
| 1631 |
+
"step": 2300
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 27.34,
|
| 1635 |
+
"learning_rate": 2.5e-05,
|
| 1636 |
+
"loss": 0.1382,
|
| 1637 |
+
"step": 2310
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 27.46,
|
| 1641 |
+
"learning_rate": 2.4867724867724866e-05,
|
| 1642 |
+
"loss": 0.1336,
|
| 1643 |
+
"step": 2320
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 27.57,
|
| 1647 |
+
"learning_rate": 2.4735449735449737e-05,
|
| 1648 |
+
"loss": 0.1308,
|
| 1649 |
+
"step": 2330
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 27.69,
|
| 1653 |
+
"learning_rate": 2.4603174603174602e-05,
|
| 1654 |
+
"loss": 0.1285,
|
| 1655 |
+
"step": 2340
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 27.81,
|
| 1659 |
+
"learning_rate": 2.4470899470899473e-05,
|
| 1660 |
+
"loss": 0.1424,
|
| 1661 |
+
"step": 2350
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 27.93,
|
| 1665 |
+
"learning_rate": 2.4338624338624338e-05,
|
| 1666 |
+
"loss": 0.1567,
|
| 1667 |
+
"step": 2360
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 28.0,
|
| 1671 |
+
"eval_accuracy": 0.9084096586178185,
|
| 1672 |
+
"eval_loss": 0.24210087954998016,
|
| 1673 |
+
"eval_runtime": 15.8265,
|
| 1674 |
+
"eval_samples_per_second": 75.885,
|
| 1675 |
+
"eval_steps_per_second": 2.401,
|
| 1676 |
+
"step": 2366
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 28.05,
|
| 1680 |
+
"learning_rate": 2.4206349206349206e-05,
|
| 1681 |
+
"loss": 0.135,
|
| 1682 |
+
"step": 2370
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 28.17,
|
| 1686 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 1687 |
+
"loss": 0.1305,
|
| 1688 |
+
"step": 2380
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 28.28,
|
| 1692 |
+
"learning_rate": 2.3941798941798942e-05,
|
| 1693 |
+
"loss": 0.1351,
|
| 1694 |
+
"step": 2390
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 28.4,
|
| 1698 |
+
"learning_rate": 2.380952380952381e-05,
|
| 1699 |
+
"loss": 0.1351,
|
| 1700 |
+
"step": 2400
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 28.52,
|
| 1704 |
+
"learning_rate": 2.3677248677248678e-05,
|
| 1705 |
+
"loss": 0.1272,
|
| 1706 |
+
"step": 2410
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 28.64,
|
| 1710 |
+
"learning_rate": 2.3544973544973546e-05,
|
| 1711 |
+
"loss": 0.1522,
|
| 1712 |
+
"step": 2420
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 28.76,
|
| 1716 |
+
"learning_rate": 2.3412698412698414e-05,
|
| 1717 |
+
"loss": 0.1295,
|
| 1718 |
+
"step": 2430
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 28.88,
|
| 1722 |
+
"learning_rate": 2.328042328042328e-05,
|
| 1723 |
+
"loss": 0.141,
|
| 1724 |
+
"step": 2440
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 28.99,
|
| 1728 |
+
"learning_rate": 2.314814814814815e-05,
|
| 1729 |
+
"loss": 0.1441,
|
| 1730 |
+
"step": 2450
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 28.99,
|
| 1734 |
+
"eval_accuracy": 0.9167360532889259,
|
| 1735 |
+
"eval_loss": 0.2525836229324341,
|
| 1736 |
+
"eval_runtime": 16.1677,
|
| 1737 |
+
"eval_samples_per_second": 74.284,
|
| 1738 |
+
"eval_steps_per_second": 2.35,
|
| 1739 |
+
"step": 2450
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 29.11,
|
| 1743 |
+
"learning_rate": 2.3015873015873015e-05,
|
| 1744 |
+
"loss": 0.1166,
|
| 1745 |
+
"step": 2460
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 29.23,
|
| 1749 |
+
"learning_rate": 2.2883597883597886e-05,
|
| 1750 |
+
"loss": 0.1318,
|
| 1751 |
+
"step": 2470
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 29.35,
|
| 1755 |
+
"learning_rate": 2.275132275132275e-05,
|
| 1756 |
+
"loss": 0.1171,
|
| 1757 |
+
"step": 2480
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 29.47,
|
| 1761 |
+
"learning_rate": 2.261904761904762e-05,
|
| 1762 |
+
"loss": 0.1276,
|
| 1763 |
+
"step": 2490
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 29.59,
|
| 1767 |
+
"learning_rate": 2.2486772486772487e-05,
|
| 1768 |
+
"loss": 0.1442,
|
| 1769 |
+
"step": 2500
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 29.7,
|
| 1773 |
+
"learning_rate": 2.2354497354497355e-05,
|
| 1774 |
+
"loss": 0.1522,
|
| 1775 |
+
"step": 2510
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 29.82,
|
| 1779 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 1780 |
+
"loss": 0.1357,
|
| 1781 |
+
"step": 2520
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 29.94,
|
| 1785 |
+
"learning_rate": 2.208994708994709e-05,
|
| 1786 |
+
"loss": 0.1219,
|
| 1787 |
+
"step": 2530
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 30.0,
|
| 1791 |
+
"eval_accuracy": 0.9109075770191507,
|
| 1792 |
+
"eval_loss": 0.2636389136314392,
|
| 1793 |
+
"eval_runtime": 15.8974,
|
| 1794 |
+
"eval_samples_per_second": 75.547,
|
| 1795 |
+
"eval_steps_per_second": 2.39,
|
| 1796 |
+
"step": 2535
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 30.06,
|
| 1800 |
+
"learning_rate": 2.1957671957671956e-05,
|
| 1801 |
+
"loss": 0.1387,
|
| 1802 |
+
"step": 2540
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 30.18,
|
| 1806 |
+
"learning_rate": 2.1825396825396827e-05,
|
| 1807 |
+
"loss": 0.1182,
|
| 1808 |
+
"step": 2550
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 30.3,
|
| 1812 |
+
"learning_rate": 2.1693121693121692e-05,
|
| 1813 |
+
"loss": 0.1103,
|
| 1814 |
+
"step": 2560
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 30.41,
|
| 1818 |
+
"learning_rate": 2.1560846560846563e-05,
|
| 1819 |
+
"loss": 0.1385,
|
| 1820 |
+
"step": 2570
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 30.53,
|
| 1824 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 1825 |
+
"loss": 0.1279,
|
| 1826 |
+
"step": 2580
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 30.65,
|
| 1830 |
+
"learning_rate": 2.1296296296296296e-05,
|
| 1831 |
+
"loss": 0.1133,
|
| 1832 |
+
"step": 2590
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 30.77,
|
| 1836 |
+
"learning_rate": 2.1164021164021164e-05,
|
| 1837 |
+
"loss": 0.1475,
|
| 1838 |
+
"step": 2600
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 30.89,
|
| 1842 |
+
"learning_rate": 2.1031746031746032e-05,
|
| 1843 |
+
"loss": 0.1137,
|
| 1844 |
+
"step": 2610
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 30.99,
|
| 1848 |
+
"eval_accuracy": 0.9034138218151541,
|
| 1849 |
+
"eval_loss": 0.2889064848423004,
|
| 1850 |
+
"eval_runtime": 15.4125,
|
| 1851 |
+
"eval_samples_per_second": 77.924,
|
| 1852 |
+
"eval_steps_per_second": 2.466,
|
| 1853 |
+
"step": 2619
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 31.01,
|
| 1857 |
+
"learning_rate": 2.08994708994709e-05,
|
| 1858 |
+
"loss": 0.1316,
|
| 1859 |
+
"step": 2620
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 31.12,
|
| 1863 |
+
"learning_rate": 2.076719576719577e-05,
|
| 1864 |
+
"loss": 0.1245,
|
| 1865 |
+
"step": 2630
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 31.24,
|
| 1869 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 1870 |
+
"loss": 0.1104,
|
| 1871 |
+
"step": 2640
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 31.36,
|
| 1875 |
+
"learning_rate": 2.0502645502645504e-05,
|
| 1876 |
+
"loss": 0.1201,
|
| 1877 |
+
"step": 2650
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 31.48,
|
| 1881 |
+
"learning_rate": 2.037037037037037e-05,
|
| 1882 |
+
"loss": 0.1269,
|
| 1883 |
+
"step": 2660
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 31.6,
|
| 1887 |
+
"learning_rate": 2.023809523809524e-05,
|
| 1888 |
+
"loss": 0.1034,
|
| 1889 |
+
"step": 2670
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 31.72,
|
| 1893 |
+
"learning_rate": 2.0105820105820105e-05,
|
| 1894 |
+
"loss": 0.1039,
|
| 1895 |
+
"step": 2680
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 31.83,
|
| 1899 |
+
"learning_rate": 1.9973544973544973e-05,
|
| 1900 |
+
"loss": 0.1297,
|
| 1901 |
+
"step": 2690
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 31.95,
|
| 1905 |
+
"learning_rate": 1.984126984126984e-05,
|
| 1906 |
+
"loss": 0.1109,
|
| 1907 |
+
"step": 2700
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 32.0,
|
| 1911 |
+
"eval_accuracy": 0.8917568692756037,
|
| 1912 |
+
"eval_loss": 0.3240314722061157,
|
| 1913 |
+
"eval_runtime": 15.6837,
|
| 1914 |
+
"eval_samples_per_second": 76.576,
|
| 1915 |
+
"eval_steps_per_second": 2.423,
|
| 1916 |
+
"step": 2704
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 32.07,
|
| 1920 |
+
"learning_rate": 1.970899470899471e-05,
|
| 1921 |
+
"loss": 0.1219,
|
| 1922 |
+
"step": 2710
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 32.19,
|
| 1926 |
+
"learning_rate": 1.9576719576719577e-05,
|
| 1927 |
+
"loss": 0.125,
|
| 1928 |
+
"step": 2720
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 32.31,
|
| 1932 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 1933 |
+
"loss": 0.1037,
|
| 1934 |
+
"step": 2730
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 32.43,
|
| 1938 |
+
"learning_rate": 1.9312169312169313e-05,
|
| 1939 |
+
"loss": 0.0952,
|
| 1940 |
+
"step": 2740
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 32.54,
|
| 1944 |
+
"learning_rate": 1.917989417989418e-05,
|
| 1945 |
+
"loss": 0.1125,
|
| 1946 |
+
"step": 2750
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 32.66,
|
| 1950 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 1951 |
+
"loss": 0.1036,
|
| 1952 |
+
"step": 2760
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 32.78,
|
| 1956 |
+
"learning_rate": 1.8915343915343918e-05,
|
| 1957 |
+
"loss": 0.1192,
|
| 1958 |
+
"step": 2770
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 32.9,
|
| 1962 |
+
"learning_rate": 1.8783068783068782e-05,
|
| 1963 |
+
"loss": 0.1254,
|
| 1964 |
+
"step": 2780
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 32.99,
|
| 1968 |
+
"eval_accuracy": 0.9150707743547044,
|
| 1969 |
+
"eval_loss": 0.2847664952278137,
|
| 1970 |
+
"eval_runtime": 15.3587,
|
| 1971 |
+
"eval_samples_per_second": 78.196,
|
| 1972 |
+
"eval_steps_per_second": 2.474,
|
| 1973 |
+
"step": 2788
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 33.02,
|
| 1977 |
+
"learning_rate": 1.8650793650793654e-05,
|
| 1978 |
+
"loss": 0.0989,
|
| 1979 |
+
"step": 2790
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 33.14,
|
| 1983 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 1984 |
+
"loss": 0.1006,
|
| 1985 |
+
"step": 2800
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 33.25,
|
| 1989 |
+
"learning_rate": 1.8386243386243386e-05,
|
| 1990 |
+
"loss": 0.1213,
|
| 1991 |
+
"step": 2810
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 33.37,
|
| 1995 |
+
"learning_rate": 1.8253968253968254e-05,
|
| 1996 |
+
"loss": 0.1153,
|
| 1997 |
+
"step": 2820
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 33.49,
|
| 2001 |
+
"learning_rate": 1.8121693121693122e-05,
|
| 2002 |
+
"loss": 0.1199,
|
| 2003 |
+
"step": 2830
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 33.61,
|
| 2007 |
+
"learning_rate": 1.798941798941799e-05,
|
| 2008 |
+
"loss": 0.1034,
|
| 2009 |
+
"step": 2840
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 33.73,
|
| 2013 |
+
"learning_rate": 1.785714285714286e-05,
|
| 2014 |
+
"loss": 0.1013,
|
| 2015 |
+
"step": 2850
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 33.85,
|
| 2019 |
+
"learning_rate": 1.7724867724867723e-05,
|
| 2020 |
+
"loss": 0.1197,
|
| 2021 |
+
"step": 2860
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 33.96,
|
| 2025 |
+
"learning_rate": 1.7592592592592595e-05,
|
| 2026 |
+
"loss": 0.1425,
|
| 2027 |
+
"step": 2870
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 34.0,
|
| 2031 |
+
"eval_accuracy": 0.9092422980849292,
|
| 2032 |
+
"eval_loss": 0.27417123317718506,
|
| 2033 |
+
"eval_runtime": 15.4475,
|
| 2034 |
+
"eval_samples_per_second": 77.747,
|
| 2035 |
+
"eval_steps_per_second": 2.46,
|
| 2036 |
+
"step": 2873
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 34.08,
|
| 2040 |
+
"learning_rate": 1.746031746031746e-05,
|
| 2041 |
+
"loss": 0.0986,
|
| 2042 |
+
"step": 2880
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 34.2,
|
| 2046 |
+
"learning_rate": 1.732804232804233e-05,
|
| 2047 |
+
"loss": 0.0942,
|
| 2048 |
+
"step": 2890
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 34.32,
|
| 2052 |
+
"learning_rate": 1.7195767195767195e-05,
|
| 2053 |
+
"loss": 0.1186,
|
| 2054 |
+
"step": 2900
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 34.44,
|
| 2058 |
+
"learning_rate": 1.7063492063492063e-05,
|
| 2059 |
+
"loss": 0.0965,
|
| 2060 |
+
"step": 2910
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 34.56,
|
| 2064 |
+
"learning_rate": 1.693121693121693e-05,
|
| 2065 |
+
"loss": 0.1119,
|
| 2066 |
+
"step": 2920
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 34.67,
|
| 2070 |
+
"learning_rate": 1.67989417989418e-05,
|
| 2071 |
+
"loss": 0.1238,
|
| 2072 |
+
"step": 2930
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 34.79,
|
| 2076 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 2077 |
+
"loss": 0.125,
|
| 2078 |
+
"step": 2940
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 34.91,
|
| 2082 |
+
"learning_rate": 1.6534391534391536e-05,
|
| 2083 |
+
"loss": 0.1164,
|
| 2084 |
+
"step": 2950
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 34.99,
|
| 2088 |
+
"eval_accuracy": 0.9150707743547044,
|
| 2089 |
+
"eval_loss": 0.3028131425380707,
|
| 2090 |
+
"eval_runtime": 15.5308,
|
| 2091 |
+
"eval_samples_per_second": 77.33,
|
| 2092 |
+
"eval_steps_per_second": 2.447,
|
| 2093 |
+
"step": 2957
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 35.03,
|
| 2097 |
+
"learning_rate": 1.6402116402116404e-05,
|
| 2098 |
+
"loss": 0.0972,
|
| 2099 |
+
"step": 2960
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 35.15,
|
| 2103 |
+
"learning_rate": 1.626984126984127e-05,
|
| 2104 |
+
"loss": 0.1096,
|
| 2105 |
+
"step": 2970
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 35.27,
|
| 2109 |
+
"learning_rate": 1.6137566137566136e-05,
|
| 2110 |
+
"loss": 0.1163,
|
| 2111 |
+
"step": 2980
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 35.38,
|
| 2115 |
+
"learning_rate": 1.6005291005291008e-05,
|
| 2116 |
+
"loss": 0.1034,
|
| 2117 |
+
"step": 2990
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 35.5,
|
| 2121 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 2122 |
+
"loss": 0.0962,
|
| 2123 |
+
"step": 3000
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 35.62,
|
| 2127 |
+
"learning_rate": 1.574074074074074e-05,
|
| 2128 |
+
"loss": 0.0955,
|
| 2129 |
+
"step": 3010
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 35.74,
|
| 2133 |
+
"learning_rate": 1.560846560846561e-05,
|
| 2134 |
+
"loss": 0.1082,
|
| 2135 |
+
"step": 3020
|
| 2136 |
+
},
|
| 2137 |
+
{
|
| 2138 |
+
"epoch": 35.86,
|
| 2139 |
+
"learning_rate": 1.5476190476190476e-05,
|
| 2140 |
+
"loss": 0.0875,
|
| 2141 |
+
"step": 3030
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 35.98,
|
| 2145 |
+
"learning_rate": 1.5343915343915344e-05,
|
| 2146 |
+
"loss": 0.0781,
|
| 2147 |
+
"step": 3040
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 36.0,
|
| 2151 |
+
"eval_accuracy": 0.9059117402164862,
|
| 2152 |
+
"eval_loss": 0.33245834708213806,
|
| 2153 |
+
"eval_runtime": 15.6522,
|
| 2154 |
+
"eval_samples_per_second": 76.73,
|
| 2155 |
+
"eval_steps_per_second": 2.428,
|
| 2156 |
+
"step": 3042
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 36.09,
|
| 2160 |
+
"learning_rate": 1.5211640211640213e-05,
|
| 2161 |
+
"loss": 0.1148,
|
| 2162 |
+
"step": 3050
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 36.21,
|
| 2166 |
+
"learning_rate": 1.5079365079365079e-05,
|
| 2167 |
+
"loss": 0.0983,
|
| 2168 |
+
"step": 3060
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 36.33,
|
| 2172 |
+
"learning_rate": 1.4947089947089949e-05,
|
| 2173 |
+
"loss": 0.1018,
|
| 2174 |
+
"step": 3070
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 36.45,
|
| 2178 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 2179 |
+
"loss": 0.1093,
|
| 2180 |
+
"step": 3080
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 36.57,
|
| 2184 |
+
"learning_rate": 1.4682539682539683e-05,
|
| 2185 |
+
"loss": 0.1056,
|
| 2186 |
+
"step": 3090
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 36.69,
|
| 2190 |
+
"learning_rate": 1.455026455026455e-05,
|
| 2191 |
+
"loss": 0.0843,
|
| 2192 |
+
"step": 3100
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 36.8,
|
| 2196 |
+
"learning_rate": 1.4417989417989419e-05,
|
| 2197 |
+
"loss": 0.0945,
|
| 2198 |
+
"step": 3110
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 36.92,
|
| 2202 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 2203 |
+
"loss": 0.1004,
|
| 2204 |
+
"step": 3120
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 36.99,
|
| 2208 |
+
"eval_accuracy": 0.9017485428809325,
|
| 2209 |
+
"eval_loss": 0.314680814743042,
|
| 2210 |
+
"eval_runtime": 15.403,
|
| 2211 |
+
"eval_samples_per_second": 77.972,
|
| 2212 |
+
"eval_steps_per_second": 2.467,
|
| 2213 |
+
"step": 3126
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 37.04,
|
| 2217 |
+
"learning_rate": 1.4153439153439155e-05,
|
| 2218 |
+
"loss": 0.0903,
|
| 2219 |
+
"step": 3130
|
| 2220 |
+
},
|
| 2221 |
+
{
|
| 2222 |
+
"epoch": 37.16,
|
| 2223 |
+
"learning_rate": 1.4021164021164022e-05,
|
| 2224 |
+
"loss": 0.0932,
|
| 2225 |
+
"step": 3140
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"epoch": 37.28,
|
| 2229 |
+
"learning_rate": 1.388888888888889e-05,
|
| 2230 |
+
"loss": 0.0949,
|
| 2231 |
+
"step": 3150
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"epoch": 37.4,
|
| 2235 |
+
"learning_rate": 1.3756613756613756e-05,
|
| 2236 |
+
"loss": 0.0923,
|
| 2237 |
+
"step": 3160
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 37.51,
|
| 2241 |
+
"learning_rate": 1.3624338624338626e-05,
|
| 2242 |
+
"loss": 0.0837,
|
| 2243 |
+
"step": 3170
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 37.63,
|
| 2247 |
+
"learning_rate": 1.3492063492063492e-05,
|
| 2248 |
+
"loss": 0.1104,
|
| 2249 |
+
"step": 3180
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 37.75,
|
| 2253 |
+
"learning_rate": 1.3359788359788362e-05,
|
| 2254 |
+
"loss": 0.1168,
|
| 2255 |
+
"step": 3190
|
| 2256 |
+
},
|
| 2257 |
+
{
|
| 2258 |
+
"epoch": 37.87,
|
| 2259 |
+
"learning_rate": 1.3227513227513228e-05,
|
| 2260 |
+
"loss": 0.098,
|
| 2261 |
+
"step": 3200
|
| 2262 |
+
},
|
| 2263 |
+
{
|
| 2264 |
+
"epoch": 37.99,
|
| 2265 |
+
"learning_rate": 1.3095238095238096e-05,
|
| 2266 |
+
"loss": 0.0955,
|
| 2267 |
+
"step": 3210
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"epoch": 38.0,
|
| 2271 |
+
"eval_accuracy": 0.9117402164862615,
|
| 2272 |
+
"eval_loss": 0.31323495507240295,
|
| 2273 |
+
"eval_runtime": 15.436,
|
| 2274 |
+
"eval_samples_per_second": 77.805,
|
| 2275 |
+
"eval_steps_per_second": 2.462,
|
| 2276 |
+
"step": 3211
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 38.11,
|
| 2280 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 2281 |
+
"loss": 0.1021,
|
| 2282 |
+
"step": 3220
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 38.22,
|
| 2286 |
+
"learning_rate": 1.2830687830687832e-05,
|
| 2287 |
+
"loss": 0.082,
|
| 2288 |
+
"step": 3230
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 38.34,
|
| 2292 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 2293 |
+
"loss": 0.1104,
|
| 2294 |
+
"step": 3240
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 38.46,
|
| 2298 |
+
"learning_rate": 1.2566137566137568e-05,
|
| 2299 |
+
"loss": 0.0954,
|
| 2300 |
+
"step": 3250
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 38.58,
|
| 2304 |
+
"learning_rate": 1.2433862433862433e-05,
|
| 2305 |
+
"loss": 0.084,
|
| 2306 |
+
"step": 3260
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 38.7,
|
| 2310 |
+
"learning_rate": 1.2301587301587301e-05,
|
| 2311 |
+
"loss": 0.0762,
|
| 2312 |
+
"step": 3270
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 38.82,
|
| 2316 |
+
"learning_rate": 1.2169312169312169e-05,
|
| 2317 |
+
"loss": 0.0789,
|
| 2318 |
+
"step": 3280
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"epoch": 38.93,
|
| 2322 |
+
"learning_rate": 1.2037037037037037e-05,
|
| 2323 |
+
"loss": 0.0942,
|
| 2324 |
+
"step": 3290
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"epoch": 38.99,
|
| 2328 |
+
"eval_accuracy": 0.9142381348875936,
|
| 2329 |
+
"eval_loss": 0.3243829607963562,
|
| 2330 |
+
"eval_runtime": 16.3904,
|
| 2331 |
+
"eval_samples_per_second": 73.275,
|
| 2332 |
+
"eval_steps_per_second": 2.318,
|
| 2333 |
+
"step": 3295
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 39.05,
|
| 2337 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 2338 |
+
"loss": 0.0888,
|
| 2339 |
+
"step": 3300
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 39.17,
|
| 2343 |
+
"learning_rate": 1.1772486772486773e-05,
|
| 2344 |
+
"loss": 0.0895,
|
| 2345 |
+
"step": 3310
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 39.29,
|
| 2349 |
+
"learning_rate": 1.164021164021164e-05,
|
| 2350 |
+
"loss": 0.0973,
|
| 2351 |
+
"step": 3320
|
| 2352 |
+
},
|
| 2353 |
+
{
|
| 2354 |
+
"epoch": 39.41,
|
| 2355 |
+
"learning_rate": 1.1507936507936508e-05,
|
| 2356 |
+
"loss": 0.0648,
|
| 2357 |
+
"step": 3330
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 39.53,
|
| 2361 |
+
"learning_rate": 1.1375661375661376e-05,
|
| 2362 |
+
"loss": 0.11,
|
| 2363 |
+
"step": 3340
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 39.64,
|
| 2367 |
+
"learning_rate": 1.1243386243386244e-05,
|
| 2368 |
+
"loss": 0.0698,
|
| 2369 |
+
"step": 3350
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 39.76,
|
| 2373 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 2374 |
+
"loss": 0.0988,
|
| 2375 |
+
"step": 3360
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 39.88,
|
| 2379 |
+
"learning_rate": 1.0978835978835978e-05,
|
| 2380 |
+
"loss": 0.0803,
|
| 2381 |
+
"step": 3370
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 40.0,
|
| 2385 |
+
"learning_rate": 1.0846560846560846e-05,
|
| 2386 |
+
"loss": 0.1077,
|
| 2387 |
+
"step": 3380
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 40.0,
|
| 2391 |
+
"eval_accuracy": 0.9134054954204829,
|
| 2392 |
+
"eval_loss": 0.32237449288368225,
|
| 2393 |
+
"eval_runtime": 15.4633,
|
| 2394 |
+
"eval_samples_per_second": 77.668,
|
| 2395 |
+
"eval_steps_per_second": 2.457,
|
| 2396 |
+
"step": 3380
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 40.12,
|
| 2400 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 2401 |
+
"loss": 0.063,
|
| 2402 |
+
"step": 3390
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 40.24,
|
| 2406 |
+
"learning_rate": 1.0582010582010582e-05,
|
| 2407 |
+
"loss": 0.077,
|
| 2408 |
+
"step": 3400
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 40.36,
|
| 2412 |
+
"learning_rate": 1.044973544973545e-05,
|
| 2413 |
+
"loss": 0.0976,
|
| 2414 |
+
"step": 3410
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 40.47,
|
| 2418 |
+
"learning_rate": 1.0317460317460318e-05,
|
| 2419 |
+
"loss": 0.1123,
|
| 2420 |
+
"step": 3420
|
| 2421 |
+
},
|
| 2422 |
+
{
|
| 2423 |
+
"epoch": 40.59,
|
| 2424 |
+
"learning_rate": 1.0185185185185185e-05,
|
| 2425 |
+
"loss": 0.0968,
|
| 2426 |
+
"step": 3430
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 40.71,
|
| 2430 |
+
"learning_rate": 1.0052910052910053e-05,
|
| 2431 |
+
"loss": 0.0808,
|
| 2432 |
+
"step": 3440
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 40.83,
|
| 2436 |
+
"learning_rate": 9.92063492063492e-06,
|
| 2437 |
+
"loss": 0.0861,
|
| 2438 |
+
"step": 3450
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 40.95,
|
| 2442 |
+
"learning_rate": 9.788359788359789e-06,
|
| 2443 |
+
"loss": 0.079,
|
| 2444 |
+
"step": 3460
|
| 2445 |
+
},
|
| 2446 |
+
{
|
| 2447 |
+
"epoch": 40.99,
|
| 2448 |
+
"eval_accuracy": 0.9134054954204829,
|
| 2449 |
+
"eval_loss": 0.33289557695388794,
|
| 2450 |
+
"eval_runtime": 15.7987,
|
| 2451 |
+
"eval_samples_per_second": 76.019,
|
| 2452 |
+
"eval_steps_per_second": 2.405,
|
| 2453 |
+
"step": 3464
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 41.07,
|
| 2457 |
+
"learning_rate": 9.656084656084657e-06,
|
| 2458 |
+
"loss": 0.0968,
|
| 2459 |
+
"step": 3470
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 41.18,
|
| 2463 |
+
"learning_rate": 9.523809523809523e-06,
|
| 2464 |
+
"loss": 0.0785,
|
| 2465 |
+
"step": 3480
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 41.3,
|
| 2469 |
+
"learning_rate": 9.391534391534391e-06,
|
| 2470 |
+
"loss": 0.0711,
|
| 2471 |
+
"step": 3490
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 41.42,
|
| 2475 |
+
"learning_rate": 9.259259259259259e-06,
|
| 2476 |
+
"loss": 0.0785,
|
| 2477 |
+
"step": 3500
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 41.54,
|
| 2481 |
+
"learning_rate": 9.126984126984127e-06,
|
| 2482 |
+
"loss": 0.0888,
|
| 2483 |
+
"step": 3510
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 41.66,
|
| 2487 |
+
"learning_rate": 8.994708994708995e-06,
|
| 2488 |
+
"loss": 0.0727,
|
| 2489 |
+
"step": 3520
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 41.78,
|
| 2493 |
+
"learning_rate": 8.862433862433862e-06,
|
| 2494 |
+
"loss": 0.0806,
|
| 2495 |
+
"step": 3530
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 41.89,
|
| 2499 |
+
"learning_rate": 8.73015873015873e-06,
|
| 2500 |
+
"loss": 0.0826,
|
| 2501 |
+
"step": 3540
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 42.0,
|
| 2505 |
+
"eval_accuracy": 0.9109075770191507,
|
| 2506 |
+
"eval_loss": 0.33648884296417236,
|
| 2507 |
+
"eval_runtime": 15.7707,
|
| 2508 |
+
"eval_samples_per_second": 76.154,
|
| 2509 |
+
"eval_steps_per_second": 2.41,
|
| 2510 |
+
"step": 3549
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 42.01,
|
| 2514 |
+
"learning_rate": 8.597883597883598e-06,
|
| 2515 |
+
"loss": 0.0678,
|
| 2516 |
+
"step": 3550
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 42.13,
|
| 2520 |
+
"learning_rate": 8.465608465608466e-06,
|
| 2521 |
+
"loss": 0.0653,
|
| 2522 |
+
"step": 3560
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 42.25,
|
| 2526 |
+
"learning_rate": 8.333333333333334e-06,
|
| 2527 |
+
"loss": 0.0788,
|
| 2528 |
+
"step": 3570
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 42.37,
|
| 2532 |
+
"learning_rate": 8.201058201058202e-06,
|
| 2533 |
+
"loss": 0.0798,
|
| 2534 |
+
"step": 3580
|
| 2535 |
+
},
|
| 2536 |
+
{
|
| 2537 |
+
"epoch": 42.49,
|
| 2538 |
+
"learning_rate": 8.068783068783068e-06,
|
| 2539 |
+
"loss": 0.0832,
|
| 2540 |
+
"step": 3590
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 42.6,
|
| 2544 |
+
"learning_rate": 7.936507936507936e-06,
|
| 2545 |
+
"loss": 0.0819,
|
| 2546 |
+
"step": 3600
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 42.72,
|
| 2550 |
+
"learning_rate": 7.804232804232804e-06,
|
| 2551 |
+
"loss": 0.0813,
|
| 2552 |
+
"step": 3610
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 42.84,
|
| 2556 |
+
"learning_rate": 7.671957671957672e-06,
|
| 2557 |
+
"loss": 0.0738,
|
| 2558 |
+
"step": 3620
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 42.96,
|
| 2562 |
+
"learning_rate": 7.5396825396825394e-06,
|
| 2563 |
+
"loss": 0.07,
|
| 2564 |
+
"step": 3630
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 42.99,
|
| 2568 |
+
"eval_accuracy": 0.9100749375520399,
|
| 2569 |
+
"eval_loss": 0.3693588972091675,
|
| 2570 |
+
"eval_runtime": 15.7316,
|
| 2571 |
+
"eval_samples_per_second": 76.343,
|
| 2572 |
+
"eval_steps_per_second": 2.416,
|
| 2573 |
+
"step": 3633
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 43.08,
|
| 2577 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 2578 |
+
"loss": 0.0717,
|
| 2579 |
+
"step": 3640
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 43.2,
|
| 2583 |
+
"learning_rate": 7.275132275132275e-06,
|
| 2584 |
+
"loss": 0.0835,
|
| 2585 |
+
"step": 3650
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 43.31,
|
| 2589 |
+
"learning_rate": 7.142857142857143e-06,
|
| 2590 |
+
"loss": 0.0765,
|
| 2591 |
+
"step": 3660
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 43.43,
|
| 2595 |
+
"learning_rate": 7.010582010582011e-06,
|
| 2596 |
+
"loss": 0.0721,
|
| 2597 |
+
"step": 3670
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 43.55,
|
| 2601 |
+
"learning_rate": 6.878306878306878e-06,
|
| 2602 |
+
"loss": 0.0733,
|
| 2603 |
+
"step": 3680
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 43.67,
|
| 2607 |
+
"learning_rate": 6.746031746031746e-06,
|
| 2608 |
+
"loss": 0.0709,
|
| 2609 |
+
"step": 3690
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 43.79,
|
| 2613 |
+
"learning_rate": 6.613756613756614e-06,
|
| 2614 |
+
"loss": 0.09,
|
| 2615 |
+
"step": 3700
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 43.91,
|
| 2619 |
+
"learning_rate": 6.481481481481481e-06,
|
| 2620 |
+
"loss": 0.0698,
|
| 2621 |
+
"step": 3710
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 44.0,
|
| 2625 |
+
"eval_accuracy": 0.9184013322231473,
|
| 2626 |
+
"eval_loss": 0.3492058217525482,
|
| 2627 |
+
"eval_runtime": 15.97,
|
| 2628 |
+
"eval_samples_per_second": 75.204,
|
| 2629 |
+
"eval_steps_per_second": 2.379,
|
| 2630 |
+
"step": 3718
|
| 2631 |
+
}
|
| 2632 |
+
],
|
| 2633 |
+
"max_steps": 4200,
|
| 2634 |
+
"num_train_epochs": 50,
|
| 2635 |
+
"total_flos": 4.035107025911808e+19,
|
| 2636 |
+
"trial_name": null,
|
| 2637 |
+
"trial_params": null
|
| 2638 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd714c320d57b3046e0a922a49d3fe30947df6baa9b6bfa29227dc7db6d4fc76
|
| 3 |
+
size 4027
|