Upload 8 files
Browse files- config.json +56 -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 +958 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "melanoma",
|
| 26 |
+
"1": "nevus",
|
| 27 |
+
"2": "seborrheic_keratosis"
|
| 28 |
+
},
|
| 29 |
+
"label2id": {
|
| 30 |
+
"melanoma": 0,
|
| 31 |
+
"nevus": 1,
|
| 32 |
+
"seborrheic_keratosis": 2
|
| 33 |
+
},
|
| 34 |
+
"layer_type": "preactivation",
|
| 35 |
+
"model_type": "bit",
|
| 36 |
+
"num_channels": 3,
|
| 37 |
+
"num_groups": 32,
|
| 38 |
+
"out_features": [
|
| 39 |
+
"stage4"
|
| 40 |
+
],
|
| 41 |
+
"out_indices": [
|
| 42 |
+
4
|
| 43 |
+
],
|
| 44 |
+
"output_stride": 32,
|
| 45 |
+
"problem_type": "single_label_classification",
|
| 46 |
+
"stage_names": [
|
| 47 |
+
"stem",
|
| 48 |
+
"stage1",
|
| 49 |
+
"stage2",
|
| 50 |
+
"stage3",
|
| 51 |
+
"stage4"
|
| 52 |
+
],
|
| 53 |
+
"torch_dtype": "float32",
|
| 54 |
+
"transformers_version": "4.31.0",
|
| 55 |
+
"width_factor": 1
|
| 56 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e5972aef562932591283eca95a4ffbe11204bd97d515cdb2cdf547073d77e63
|
| 3 |
+
size 188142469
|
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 sha256:6ba1fe789de85b5656a87c2a49163ab836851947b0b51669182448afc44b8076
|
| 3 |
+
size 94077617
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d492226c0859c69b7d7954e558f40046f471ce1a604961375b17757a76021572
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:861929b6e2c3d66c3399c5946123c089831a1b0206860b461cd934bb6336b5cb
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,958 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8733333333333333,
|
| 3 |
+
"best_model_checkpoint": "bit-50-skin-lesions\\checkpoint-1062",
|
| 4 |
+
"epoch": 33.984,
|
| 5 |
+
"global_step": 1062,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.32,
|
| 12 |
+
"learning_rate": 3.225806451612903e-06,
|
| 13 |
+
"loss": 1.2002,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.64,
|
| 18 |
+
"learning_rate": 6.451612903225806e-06,
|
| 19 |
+
"loss": 0.9773,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.96,
|
| 24 |
+
"learning_rate": 9.67741935483871e-06,
|
| 25 |
+
"loss": 0.8102,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.99,
|
| 30 |
+
"eval_accuracy": 0.5333333333333333,
|
| 31 |
+
"eval_loss": 0.968501091003418,
|
| 32 |
+
"eval_runtime": 21.6336,
|
| 33 |
+
"eval_samples_per_second": 6.934,
|
| 34 |
+
"eval_steps_per_second": 0.462,
|
| 35 |
+
"step": 31
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.28,
|
| 39 |
+
"learning_rate": 1.2903225806451613e-05,
|
| 40 |
+
"loss": 0.699,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.6,
|
| 45 |
+
"learning_rate": 1.6129032258064517e-05,
|
| 46 |
+
"loss": 0.7167,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.92,
|
| 51 |
+
"learning_rate": 1.935483870967742e-05,
|
| 52 |
+
"loss": 0.5819,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.98,
|
| 57 |
+
"eval_accuracy": 0.74,
|
| 58 |
+
"eval_loss": 0.6370717287063599,
|
| 59 |
+
"eval_runtime": 20.2966,
|
| 60 |
+
"eval_samples_per_second": 7.39,
|
| 61 |
+
"eval_steps_per_second": 0.493,
|
| 62 |
+
"step": 62
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.24,
|
| 66 |
+
"learning_rate": 2.258064516129032e-05,
|
| 67 |
+
"loss": 0.5839,
|
| 68 |
+
"step": 70
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.56,
|
| 72 |
+
"learning_rate": 2.5806451612903226e-05,
|
| 73 |
+
"loss": 0.547,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.88,
|
| 78 |
+
"learning_rate": 2.9032258064516133e-05,
|
| 79 |
+
"loss": 0.5664,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.98,
|
| 84 |
+
"eval_accuracy": 0.7333333333333333,
|
| 85 |
+
"eval_loss": 0.6208174824714661,
|
| 86 |
+
"eval_runtime": 20.1891,
|
| 87 |
+
"eval_samples_per_second": 7.43,
|
| 88 |
+
"eval_steps_per_second": 0.495,
|
| 89 |
+
"step": 93
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.2,
|
| 93 |
+
"learning_rate": 3.2258064516129034e-05,
|
| 94 |
+
"loss": 0.5493,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 3.52,
|
| 99 |
+
"learning_rate": 3.548387096774194e-05,
|
| 100 |
+
"loss": 0.5026,
|
| 101 |
+
"step": 110
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.84,
|
| 105 |
+
"learning_rate": 3.870967741935484e-05,
|
| 106 |
+
"loss": 0.552,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.0,
|
| 111 |
+
"eval_accuracy": 0.7466666666666667,
|
| 112 |
+
"eval_loss": 0.6024346947669983,
|
| 113 |
+
"eval_runtime": 20.2094,
|
| 114 |
+
"eval_samples_per_second": 7.422,
|
| 115 |
+
"eval_steps_per_second": 0.495,
|
| 116 |
+
"step": 125
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.16,
|
| 120 |
+
"learning_rate": 4.1935483870967746e-05,
|
| 121 |
+
"loss": 0.4447,
|
| 122 |
+
"step": 130
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.48,
|
| 126 |
+
"learning_rate": 4.516129032258064e-05,
|
| 127 |
+
"loss": 0.5273,
|
| 128 |
+
"step": 140
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.8,
|
| 132 |
+
"learning_rate": 4.8387096774193554e-05,
|
| 133 |
+
"loss": 0.4998,
|
| 134 |
+
"step": 150
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.99,
|
| 138 |
+
"eval_accuracy": 0.76,
|
| 139 |
+
"eval_loss": 0.5782171487808228,
|
| 140 |
+
"eval_runtime": 19.9613,
|
| 141 |
+
"eval_samples_per_second": 7.515,
|
| 142 |
+
"eval_steps_per_second": 0.501,
|
| 143 |
+
"step": 156
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 5.12,
|
| 147 |
+
"learning_rate": 4.982078853046595e-05,
|
| 148 |
+
"loss": 0.5239,
|
| 149 |
+
"step": 160
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 5.44,
|
| 153 |
+
"learning_rate": 4.9462365591397855e-05,
|
| 154 |
+
"loss": 0.4935,
|
| 155 |
+
"step": 170
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 5.76,
|
| 159 |
+
"learning_rate": 4.910394265232976e-05,
|
| 160 |
+
"loss": 0.4417,
|
| 161 |
+
"step": 180
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.98,
|
| 165 |
+
"eval_accuracy": 0.78,
|
| 166 |
+
"eval_loss": 0.504960298538208,
|
| 167 |
+
"eval_runtime": 20.0953,
|
| 168 |
+
"eval_samples_per_second": 7.464,
|
| 169 |
+
"eval_steps_per_second": 0.498,
|
| 170 |
+
"step": 187
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.08,
|
| 174 |
+
"learning_rate": 4.874551971326165e-05,
|
| 175 |
+
"loss": 0.428,
|
| 176 |
+
"step": 190
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.4,
|
| 180 |
+
"learning_rate": 4.8387096774193554e-05,
|
| 181 |
+
"loss": 0.4105,
|
| 182 |
+
"step": 200
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.72,
|
| 186 |
+
"learning_rate": 4.802867383512545e-05,
|
| 187 |
+
"loss": 0.4161,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 6.98,
|
| 192 |
+
"eval_accuracy": 0.7666666666666667,
|
| 193 |
+
"eval_loss": 0.6188395619392395,
|
| 194 |
+
"eval_runtime": 20.1038,
|
| 195 |
+
"eval_samples_per_second": 7.461,
|
| 196 |
+
"eval_steps_per_second": 0.497,
|
| 197 |
+
"step": 218
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.04,
|
| 201 |
+
"learning_rate": 4.767025089605735e-05,
|
| 202 |
+
"loss": 0.4169,
|
| 203 |
+
"step": 220
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 7.36,
|
| 207 |
+
"learning_rate": 4.731182795698925e-05,
|
| 208 |
+
"loss": 0.306,
|
| 209 |
+
"step": 230
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 7.68,
|
| 213 |
+
"learning_rate": 4.695340501792115e-05,
|
| 214 |
+
"loss": 0.3994,
|
| 215 |
+
"step": 240
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 8.0,
|
| 219 |
+
"learning_rate": 4.659498207885305e-05,
|
| 220 |
+
"loss": 0.4077,
|
| 221 |
+
"step": 250
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 8.0,
|
| 225 |
+
"eval_accuracy": 0.8,
|
| 226 |
+
"eval_loss": 0.630321741104126,
|
| 227 |
+
"eval_runtime": 20.0628,
|
| 228 |
+
"eval_samples_per_second": 7.477,
|
| 229 |
+
"eval_steps_per_second": 0.498,
|
| 230 |
+
"step": 250
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 8.32,
|
| 234 |
+
"learning_rate": 4.6236559139784944e-05,
|
| 235 |
+
"loss": 0.3623,
|
| 236 |
+
"step": 260
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 8.64,
|
| 240 |
+
"learning_rate": 4.5878136200716846e-05,
|
| 241 |
+
"loss": 0.3911,
|
| 242 |
+
"step": 270
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 8.96,
|
| 246 |
+
"learning_rate": 4.551971326164875e-05,
|
| 247 |
+
"loss": 0.4163,
|
| 248 |
+
"step": 280
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.99,
|
| 252 |
+
"eval_accuracy": 0.7933333333333333,
|
| 253 |
+
"eval_loss": 0.548830509185791,
|
| 254 |
+
"eval_runtime": 20.0908,
|
| 255 |
+
"eval_samples_per_second": 7.466,
|
| 256 |
+
"eval_steps_per_second": 0.498,
|
| 257 |
+
"step": 281
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 9.28,
|
| 261 |
+
"learning_rate": 4.516129032258064e-05,
|
| 262 |
+
"loss": 0.319,
|
| 263 |
+
"step": 290
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.6,
|
| 267 |
+
"learning_rate": 4.4802867383512545e-05,
|
| 268 |
+
"loss": 0.332,
|
| 269 |
+
"step": 300
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 9.92,
|
| 273 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 274 |
+
"loss": 0.367,
|
| 275 |
+
"step": 310
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 9.98,
|
| 279 |
+
"eval_accuracy": 0.8066666666666666,
|
| 280 |
+
"eval_loss": 0.5251256227493286,
|
| 281 |
+
"eval_runtime": 20.2039,
|
| 282 |
+
"eval_samples_per_second": 7.424,
|
| 283 |
+
"eval_steps_per_second": 0.495,
|
| 284 |
+
"step": 312
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.24,
|
| 288 |
+
"learning_rate": 4.408602150537635e-05,
|
| 289 |
+
"loss": 0.4215,
|
| 290 |
+
"step": 320
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 10.56,
|
| 294 |
+
"learning_rate": 4.372759856630825e-05,
|
| 295 |
+
"loss": 0.3571,
|
| 296 |
+
"step": 330
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 10.88,
|
| 300 |
+
"learning_rate": 4.3369175627240145e-05,
|
| 301 |
+
"loss": 0.3391,
|
| 302 |
+
"step": 340
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 10.98,
|
| 306 |
+
"eval_accuracy": 0.78,
|
| 307 |
+
"eval_loss": 0.5417120456695557,
|
| 308 |
+
"eval_runtime": 20.1843,
|
| 309 |
+
"eval_samples_per_second": 7.432,
|
| 310 |
+
"eval_steps_per_second": 0.495,
|
| 311 |
+
"step": 343
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 11.2,
|
| 315 |
+
"learning_rate": 4.301075268817205e-05,
|
| 316 |
+
"loss": 0.3091,
|
| 317 |
+
"step": 350
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 11.52,
|
| 321 |
+
"learning_rate": 4.265232974910394e-05,
|
| 322 |
+
"loss": 0.2611,
|
| 323 |
+
"step": 360
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 11.84,
|
| 327 |
+
"learning_rate": 4.2293906810035844e-05,
|
| 328 |
+
"loss": 0.3298,
|
| 329 |
+
"step": 370
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 12.0,
|
| 333 |
+
"eval_accuracy": 0.7933333333333333,
|
| 334 |
+
"eval_loss": 0.4787960946559906,
|
| 335 |
+
"eval_runtime": 20.0289,
|
| 336 |
+
"eval_samples_per_second": 7.489,
|
| 337 |
+
"eval_steps_per_second": 0.499,
|
| 338 |
+
"step": 375
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 12.16,
|
| 342 |
+
"learning_rate": 4.1935483870967746e-05,
|
| 343 |
+
"loss": 0.2416,
|
| 344 |
+
"step": 380
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 12.48,
|
| 348 |
+
"learning_rate": 4.157706093189964e-05,
|
| 349 |
+
"loss": 0.2494,
|
| 350 |
+
"step": 390
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.8,
|
| 354 |
+
"learning_rate": 4.121863799283154e-05,
|
| 355 |
+
"loss": 0.3133,
|
| 356 |
+
"step": 400
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 12.99,
|
| 360 |
+
"eval_accuracy": 0.7733333333333333,
|
| 361 |
+
"eval_loss": 0.6023210883140564,
|
| 362 |
+
"eval_runtime": 19.8848,
|
| 363 |
+
"eval_samples_per_second": 7.543,
|
| 364 |
+
"eval_steps_per_second": 0.503,
|
| 365 |
+
"step": 406
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 13.12,
|
| 369 |
+
"learning_rate": 4.0860215053763444e-05,
|
| 370 |
+
"loss": 0.2826,
|
| 371 |
+
"step": 410
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 13.44,
|
| 375 |
+
"learning_rate": 4.050179211469534e-05,
|
| 376 |
+
"loss": 0.2571,
|
| 377 |
+
"step": 420
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 13.76,
|
| 381 |
+
"learning_rate": 4.014336917562724e-05,
|
| 382 |
+
"loss": 0.2916,
|
| 383 |
+
"step": 430
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 13.98,
|
| 387 |
+
"eval_accuracy": 0.76,
|
| 388 |
+
"eval_loss": 0.5615163445472717,
|
| 389 |
+
"eval_runtime": 19.8565,
|
| 390 |
+
"eval_samples_per_second": 7.554,
|
| 391 |
+
"eval_steps_per_second": 0.504,
|
| 392 |
+
"step": 437
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 14.08,
|
| 396 |
+
"learning_rate": 3.978494623655914e-05,
|
| 397 |
+
"loss": 0.3033,
|
| 398 |
+
"step": 440
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 14.4,
|
| 402 |
+
"learning_rate": 3.9426523297491045e-05,
|
| 403 |
+
"loss": 0.2611,
|
| 404 |
+
"step": 450
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 14.72,
|
| 408 |
+
"learning_rate": 3.906810035842295e-05,
|
| 409 |
+
"loss": 0.2491,
|
| 410 |
+
"step": 460
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 14.98,
|
| 414 |
+
"eval_accuracy": 0.84,
|
| 415 |
+
"eval_loss": 0.4784340560436249,
|
| 416 |
+
"eval_runtime": 20.1708,
|
| 417 |
+
"eval_samples_per_second": 7.437,
|
| 418 |
+
"eval_steps_per_second": 0.496,
|
| 419 |
+
"step": 468
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 15.04,
|
| 423 |
+
"learning_rate": 3.870967741935484e-05,
|
| 424 |
+
"loss": 0.1961,
|
| 425 |
+
"step": 470
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 15.36,
|
| 429 |
+
"learning_rate": 3.8351254480286743e-05,
|
| 430 |
+
"loss": 0.2239,
|
| 431 |
+
"step": 480
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 15.68,
|
| 435 |
+
"learning_rate": 3.799283154121864e-05,
|
| 436 |
+
"loss": 0.2206,
|
| 437 |
+
"step": 490
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.0,
|
| 441 |
+
"learning_rate": 3.763440860215054e-05,
|
| 442 |
+
"loss": 0.2213,
|
| 443 |
+
"step": 500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 16.0,
|
| 447 |
+
"eval_accuracy": 0.8,
|
| 448 |
+
"eval_loss": 0.5498819351196289,
|
| 449 |
+
"eval_runtime": 20.1376,
|
| 450 |
+
"eval_samples_per_second": 7.449,
|
| 451 |
+
"eval_steps_per_second": 0.497,
|
| 452 |
+
"step": 500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 16.32,
|
| 456 |
+
"learning_rate": 3.727598566308244e-05,
|
| 457 |
+
"loss": 0.2294,
|
| 458 |
+
"step": 510
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 16.64,
|
| 462 |
+
"learning_rate": 3.691756272401434e-05,
|
| 463 |
+
"loss": 0.228,
|
| 464 |
+
"step": 520
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 16.96,
|
| 468 |
+
"learning_rate": 3.655913978494624e-05,
|
| 469 |
+
"loss": 0.2334,
|
| 470 |
+
"step": 530
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 16.99,
|
| 474 |
+
"eval_accuracy": 0.8133333333333334,
|
| 475 |
+
"eval_loss": 0.5679303407669067,
|
| 476 |
+
"eval_runtime": 19.8409,
|
| 477 |
+
"eval_samples_per_second": 7.56,
|
| 478 |
+
"eval_steps_per_second": 0.504,
|
| 479 |
+
"step": 531
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 17.28,
|
| 483 |
+
"learning_rate": 3.6200716845878134e-05,
|
| 484 |
+
"loss": 0.1944,
|
| 485 |
+
"step": 540
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 17.6,
|
| 489 |
+
"learning_rate": 3.5842293906810036e-05,
|
| 490 |
+
"loss": 0.1782,
|
| 491 |
+
"step": 550
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 17.92,
|
| 495 |
+
"learning_rate": 3.548387096774194e-05,
|
| 496 |
+
"loss": 0.1882,
|
| 497 |
+
"step": 560
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 17.98,
|
| 501 |
+
"eval_accuracy": 0.82,
|
| 502 |
+
"eval_loss": 0.5990158319473267,
|
| 503 |
+
"eval_runtime": 19.6405,
|
| 504 |
+
"eval_samples_per_second": 7.637,
|
| 505 |
+
"eval_steps_per_second": 0.509,
|
| 506 |
+
"step": 562
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 18.24,
|
| 510 |
+
"learning_rate": 3.512544802867383e-05,
|
| 511 |
+
"loss": 0.1918,
|
| 512 |
+
"step": 570
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 18.56,
|
| 516 |
+
"learning_rate": 3.4767025089605734e-05,
|
| 517 |
+
"loss": 0.208,
|
| 518 |
+
"step": 580
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 18.88,
|
| 522 |
+
"learning_rate": 3.4408602150537636e-05,
|
| 523 |
+
"loss": 0.2121,
|
| 524 |
+
"step": 590
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 18.98,
|
| 528 |
+
"eval_accuracy": 0.8333333333333334,
|
| 529 |
+
"eval_loss": 0.4990116357803345,
|
| 530 |
+
"eval_runtime": 20.1236,
|
| 531 |
+
"eval_samples_per_second": 7.454,
|
| 532 |
+
"eval_steps_per_second": 0.497,
|
| 533 |
+
"step": 593
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 19.2,
|
| 537 |
+
"learning_rate": 3.405017921146954e-05,
|
| 538 |
+
"loss": 0.1591,
|
| 539 |
+
"step": 600
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 19.52,
|
| 543 |
+
"learning_rate": 3.369175627240144e-05,
|
| 544 |
+
"loss": 0.1472,
|
| 545 |
+
"step": 610
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 19.84,
|
| 549 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 550 |
+
"loss": 0.1456,
|
| 551 |
+
"step": 620
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 20.0,
|
| 555 |
+
"eval_accuracy": 0.8066666666666666,
|
| 556 |
+
"eval_loss": 0.7644553184509277,
|
| 557 |
+
"eval_runtime": 19.8391,
|
| 558 |
+
"eval_samples_per_second": 7.561,
|
| 559 |
+
"eval_steps_per_second": 0.504,
|
| 560 |
+
"step": 625
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 20.16,
|
| 564 |
+
"learning_rate": 3.297491039426524e-05,
|
| 565 |
+
"loss": 0.1569,
|
| 566 |
+
"step": 630
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 20.48,
|
| 570 |
+
"learning_rate": 3.261648745519714e-05,
|
| 571 |
+
"loss": 0.1653,
|
| 572 |
+
"step": 640
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 20.8,
|
| 576 |
+
"learning_rate": 3.2258064516129034e-05,
|
| 577 |
+
"loss": 0.1938,
|
| 578 |
+
"step": 650
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 20.99,
|
| 582 |
+
"eval_accuracy": 0.8266666666666667,
|
| 583 |
+
"eval_loss": 0.579462468624115,
|
| 584 |
+
"eval_runtime": 19.9203,
|
| 585 |
+
"eval_samples_per_second": 7.53,
|
| 586 |
+
"eval_steps_per_second": 0.502,
|
| 587 |
+
"step": 656
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 21.12,
|
| 591 |
+
"learning_rate": 3.1899641577060935e-05,
|
| 592 |
+
"loss": 0.1467,
|
| 593 |
+
"step": 660
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 21.44,
|
| 597 |
+
"learning_rate": 3.154121863799283e-05,
|
| 598 |
+
"loss": 0.1463,
|
| 599 |
+
"step": 670
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 21.76,
|
| 603 |
+
"learning_rate": 3.118279569892473e-05,
|
| 604 |
+
"loss": 0.1489,
|
| 605 |
+
"step": 680
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 21.98,
|
| 609 |
+
"eval_accuracy": 0.8333333333333334,
|
| 610 |
+
"eval_loss": 0.4865401089191437,
|
| 611 |
+
"eval_runtime": 20.1519,
|
| 612 |
+
"eval_samples_per_second": 7.443,
|
| 613 |
+
"eval_steps_per_second": 0.496,
|
| 614 |
+
"step": 687
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 22.08,
|
| 618 |
+
"learning_rate": 3.0824372759856634e-05,
|
| 619 |
+
"loss": 0.1771,
|
| 620 |
+
"step": 690
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 22.4,
|
| 624 |
+
"learning_rate": 3.046594982078853e-05,
|
| 625 |
+
"loss": 0.1178,
|
| 626 |
+
"step": 700
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 22.72,
|
| 630 |
+
"learning_rate": 3.010752688172043e-05,
|
| 631 |
+
"loss": 0.1887,
|
| 632 |
+
"step": 710
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 22.98,
|
| 636 |
+
"eval_accuracy": 0.8,
|
| 637 |
+
"eval_loss": 0.641372561454773,
|
| 638 |
+
"eval_runtime": 20.0159,
|
| 639 |
+
"eval_samples_per_second": 7.494,
|
| 640 |
+
"eval_steps_per_second": 0.5,
|
| 641 |
+
"step": 718
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 23.04,
|
| 645 |
+
"learning_rate": 2.974910394265233e-05,
|
| 646 |
+
"loss": 0.1736,
|
| 647 |
+
"step": 720
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 23.36,
|
| 651 |
+
"learning_rate": 2.939068100358423e-05,
|
| 652 |
+
"loss": 0.1307,
|
| 653 |
+
"step": 730
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 23.68,
|
| 657 |
+
"learning_rate": 2.9032258064516133e-05,
|
| 658 |
+
"loss": 0.1716,
|
| 659 |
+
"step": 740
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 24.0,
|
| 663 |
+
"learning_rate": 2.8673835125448028e-05,
|
| 664 |
+
"loss": 0.1713,
|
| 665 |
+
"step": 750
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 24.0,
|
| 669 |
+
"eval_accuracy": 0.84,
|
| 670 |
+
"eval_loss": 0.5748192071914673,
|
| 671 |
+
"eval_runtime": 20.2017,
|
| 672 |
+
"eval_samples_per_second": 7.425,
|
| 673 |
+
"eval_steps_per_second": 0.495,
|
| 674 |
+
"step": 750
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 24.32,
|
| 678 |
+
"learning_rate": 2.831541218637993e-05,
|
| 679 |
+
"loss": 0.1503,
|
| 680 |
+
"step": 760
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 24.64,
|
| 684 |
+
"learning_rate": 2.7956989247311828e-05,
|
| 685 |
+
"loss": 0.1278,
|
| 686 |
+
"step": 770
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 24.96,
|
| 690 |
+
"learning_rate": 2.759856630824373e-05,
|
| 691 |
+
"loss": 0.154,
|
| 692 |
+
"step": 780
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 24.99,
|
| 696 |
+
"eval_accuracy": 0.8266666666666667,
|
| 697 |
+
"eval_loss": 0.5349760055541992,
|
| 698 |
+
"eval_runtime": 20.1924,
|
| 699 |
+
"eval_samples_per_second": 7.429,
|
| 700 |
+
"eval_steps_per_second": 0.495,
|
| 701 |
+
"step": 781
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 25.28,
|
| 705 |
+
"learning_rate": 2.7240143369175632e-05,
|
| 706 |
+
"loss": 0.0884,
|
| 707 |
+
"step": 790
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 25.6,
|
| 711 |
+
"learning_rate": 2.6881720430107527e-05,
|
| 712 |
+
"loss": 0.134,
|
| 713 |
+
"step": 800
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 25.92,
|
| 717 |
+
"learning_rate": 2.652329749103943e-05,
|
| 718 |
+
"loss": 0.1232,
|
| 719 |
+
"step": 810
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 25.98,
|
| 723 |
+
"eval_accuracy": 0.8466666666666667,
|
| 724 |
+
"eval_loss": 0.5574176907539368,
|
| 725 |
+
"eval_runtime": 20.0038,
|
| 726 |
+
"eval_samples_per_second": 7.499,
|
| 727 |
+
"eval_steps_per_second": 0.5,
|
| 728 |
+
"step": 812
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 26.24,
|
| 732 |
+
"learning_rate": 2.616487455197133e-05,
|
| 733 |
+
"loss": 0.1021,
|
| 734 |
+
"step": 820
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 26.56,
|
| 738 |
+
"learning_rate": 2.5806451612903226e-05,
|
| 739 |
+
"loss": 0.1958,
|
| 740 |
+
"step": 830
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 26.88,
|
| 744 |
+
"learning_rate": 2.5448028673835127e-05,
|
| 745 |
+
"loss": 0.1514,
|
| 746 |
+
"step": 840
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 26.98,
|
| 750 |
+
"eval_accuracy": 0.8666666666666667,
|
| 751 |
+
"eval_loss": 0.4726710021495819,
|
| 752 |
+
"eval_runtime": 20.1425,
|
| 753 |
+
"eval_samples_per_second": 7.447,
|
| 754 |
+
"eval_steps_per_second": 0.496,
|
| 755 |
+
"step": 843
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 27.2,
|
| 759 |
+
"learning_rate": 2.5089605734767026e-05,
|
| 760 |
+
"loss": 0.1184,
|
| 761 |
+
"step": 850
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 27.52,
|
| 765 |
+
"learning_rate": 2.4731182795698928e-05,
|
| 766 |
+
"loss": 0.1252,
|
| 767 |
+
"step": 860
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 27.84,
|
| 771 |
+
"learning_rate": 2.4372759856630826e-05,
|
| 772 |
+
"loss": 0.0835,
|
| 773 |
+
"step": 870
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 28.0,
|
| 777 |
+
"eval_accuracy": 0.8133333333333334,
|
| 778 |
+
"eval_loss": 0.8025414943695068,
|
| 779 |
+
"eval_runtime": 19.9723,
|
| 780 |
+
"eval_samples_per_second": 7.51,
|
| 781 |
+
"eval_steps_per_second": 0.501,
|
| 782 |
+
"step": 875
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 28.16,
|
| 786 |
+
"learning_rate": 2.4014336917562724e-05,
|
| 787 |
+
"loss": 0.1916,
|
| 788 |
+
"step": 880
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 28.48,
|
| 792 |
+
"learning_rate": 2.3655913978494626e-05,
|
| 793 |
+
"loss": 0.1547,
|
| 794 |
+
"step": 890
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 28.8,
|
| 798 |
+
"learning_rate": 2.3297491039426525e-05,
|
| 799 |
+
"loss": 0.1012,
|
| 800 |
+
"step": 900
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 28.99,
|
| 804 |
+
"eval_accuracy": 0.8666666666666667,
|
| 805 |
+
"eval_loss": 0.5923764705657959,
|
| 806 |
+
"eval_runtime": 19.9464,
|
| 807 |
+
"eval_samples_per_second": 7.52,
|
| 808 |
+
"eval_steps_per_second": 0.501,
|
| 809 |
+
"step": 906
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 29.12,
|
| 813 |
+
"learning_rate": 2.2939068100358423e-05,
|
| 814 |
+
"loss": 0.0884,
|
| 815 |
+
"step": 910
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 29.44,
|
| 819 |
+
"learning_rate": 2.258064516129032e-05,
|
| 820 |
+
"loss": 0.0869,
|
| 821 |
+
"step": 920
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 29.76,
|
| 825 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 826 |
+
"loss": 0.1328,
|
| 827 |
+
"step": 930
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 29.98,
|
| 831 |
+
"eval_accuracy": 0.86,
|
| 832 |
+
"eval_loss": 0.5381821990013123,
|
| 833 |
+
"eval_runtime": 20.537,
|
| 834 |
+
"eval_samples_per_second": 7.304,
|
| 835 |
+
"eval_steps_per_second": 0.487,
|
| 836 |
+
"step": 937
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 30.08,
|
| 840 |
+
"learning_rate": 2.1863799283154125e-05,
|
| 841 |
+
"loss": 0.1074,
|
| 842 |
+
"step": 940
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 30.4,
|
| 846 |
+
"learning_rate": 2.1505376344086024e-05,
|
| 847 |
+
"loss": 0.1311,
|
| 848 |
+
"step": 950
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 30.72,
|
| 852 |
+
"learning_rate": 2.1146953405017922e-05,
|
| 853 |
+
"loss": 0.1243,
|
| 854 |
+
"step": 960
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 30.98,
|
| 858 |
+
"eval_accuracy": 0.8466666666666667,
|
| 859 |
+
"eval_loss": 0.5992811322212219,
|
| 860 |
+
"eval_runtime": 19.7967,
|
| 861 |
+
"eval_samples_per_second": 7.577,
|
| 862 |
+
"eval_steps_per_second": 0.505,
|
| 863 |
+
"step": 968
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 31.04,
|
| 867 |
+
"learning_rate": 2.078853046594982e-05,
|
| 868 |
+
"loss": 0.122,
|
| 869 |
+
"step": 970
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 31.36,
|
| 873 |
+
"learning_rate": 2.0430107526881722e-05,
|
| 874 |
+
"loss": 0.0976,
|
| 875 |
+
"step": 980
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 31.68,
|
| 879 |
+
"learning_rate": 2.007168458781362e-05,
|
| 880 |
+
"loss": 0.114,
|
| 881 |
+
"step": 990
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 32.0,
|
| 885 |
+
"learning_rate": 1.9713261648745522e-05,
|
| 886 |
+
"loss": 0.0924,
|
| 887 |
+
"step": 1000
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 32.0,
|
| 891 |
+
"eval_accuracy": 0.8133333333333334,
|
| 892 |
+
"eval_loss": 0.7414191961288452,
|
| 893 |
+
"eval_runtime": 20.4518,
|
| 894 |
+
"eval_samples_per_second": 7.334,
|
| 895 |
+
"eval_steps_per_second": 0.489,
|
| 896 |
+
"step": 1000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 32.32,
|
| 900 |
+
"learning_rate": 1.935483870967742e-05,
|
| 901 |
+
"loss": 0.117,
|
| 902 |
+
"step": 1010
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 32.64,
|
| 906 |
+
"learning_rate": 1.899641577060932e-05,
|
| 907 |
+
"loss": 0.1152,
|
| 908 |
+
"step": 1020
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 32.96,
|
| 912 |
+
"learning_rate": 1.863799283154122e-05,
|
| 913 |
+
"loss": 0.1101,
|
| 914 |
+
"step": 1030
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 32.99,
|
| 918 |
+
"eval_accuracy": 0.8333333333333334,
|
| 919 |
+
"eval_loss": 0.7131091952323914,
|
| 920 |
+
"eval_runtime": 20.5438,
|
| 921 |
+
"eval_samples_per_second": 7.301,
|
| 922 |
+
"eval_steps_per_second": 0.487,
|
| 923 |
+
"step": 1031
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 33.28,
|
| 927 |
+
"learning_rate": 1.827956989247312e-05,
|
| 928 |
+
"loss": 0.0776,
|
| 929 |
+
"step": 1040
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 33.6,
|
| 933 |
+
"learning_rate": 1.7921146953405018e-05,
|
| 934 |
+
"loss": 0.0753,
|
| 935 |
+
"step": 1050
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 33.92,
|
| 939 |
+
"learning_rate": 1.7562724014336916e-05,
|
| 940 |
+
"loss": 0.0785,
|
| 941 |
+
"step": 1060
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 33.98,
|
| 945 |
+
"eval_accuracy": 0.8733333333333333,
|
| 946 |
+
"eval_loss": 0.5796133279800415,
|
| 947 |
+
"eval_runtime": 19.6695,
|
| 948 |
+
"eval_samples_per_second": 7.626,
|
| 949 |
+
"eval_steps_per_second": 0.508,
|
| 950 |
+
"step": 1062
|
| 951 |
+
}
|
| 952 |
+
],
|
| 953 |
+
"max_steps": 1550,
|
| 954 |
+
"num_train_epochs": 50,
|
| 955 |
+
"total_flos": 5.774646411362304e+18,
|
| 956 |
+
"trial_name": null,
|
| 957 |
+
"trial_params": null
|
| 958 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f25ef23bf58a2c6bcb4b87345ed869b5952faad8168d73b0c5bbaeafee2f66d
|
| 3 |
+
size 4027
|