Upload 8 files
Browse files- config.json +99 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +22 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +471 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,99 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"EfficientNetForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"batch_norm_eps": 0.001,
|
| 6 |
+
"batch_norm_momentum": 0.99,
|
| 7 |
+
"depth_coefficient": 3.1,
|
| 8 |
+
"depth_divisor": 8,
|
| 9 |
+
"depthwise_padding": [],
|
| 10 |
+
"drop_connect_rate": 0.2,
|
| 11 |
+
"dropout_rate": 0.5,
|
| 12 |
+
"expand_ratios": [
|
| 13 |
+
1,
|
| 14 |
+
6,
|
| 15 |
+
6,
|
| 16 |
+
6,
|
| 17 |
+
6,
|
| 18 |
+
6,
|
| 19 |
+
6
|
| 20 |
+
],
|
| 21 |
+
"hidden_act": "swish",
|
| 22 |
+
"hidden_dim": 2560,
|
| 23 |
+
"id2label": {
|
| 24 |
+
"0": "Melanoma",
|
| 25 |
+
"1": "Melanocytic nevus",
|
| 26 |
+
"2": "Basal cell carcinoma",
|
| 27 |
+
"3": "Actinic keratosis",
|
| 28 |
+
"4": "Benign keratosis",
|
| 29 |
+
"5": "Dermatofibroma",
|
| 30 |
+
"6": "Vascular lesion",
|
| 31 |
+
"7": "Squamous cell carcinoma"
|
| 32 |
+
},
|
| 33 |
+
"image_size": 600,
|
| 34 |
+
"in_channels": [
|
| 35 |
+
32,
|
| 36 |
+
16,
|
| 37 |
+
24,
|
| 38 |
+
40,
|
| 39 |
+
80,
|
| 40 |
+
112,
|
| 41 |
+
192
|
| 42 |
+
],
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"kernel_sizes": [
|
| 45 |
+
3,
|
| 46 |
+
3,
|
| 47 |
+
5,
|
| 48 |
+
3,
|
| 49 |
+
5,
|
| 50 |
+
5,
|
| 51 |
+
3
|
| 52 |
+
],
|
| 53 |
+
"label2id": {
|
| 54 |
+
"Actinic keratosis": "3",
|
| 55 |
+
"Basal cell carcinoma": "2",
|
| 56 |
+
"Benign keratosis": "4",
|
| 57 |
+
"Dermatofibroma": "5",
|
| 58 |
+
"Melanocytic nevus": "1",
|
| 59 |
+
"Melanoma": "0",
|
| 60 |
+
"Squamous cell carcinoma": "7",
|
| 61 |
+
"Vascular lesion": "6"
|
| 62 |
+
},
|
| 63 |
+
"model_type": "efficientnet",
|
| 64 |
+
"num_block_repeats": [
|
| 65 |
+
1,
|
| 66 |
+
2,
|
| 67 |
+
2,
|
| 68 |
+
3,
|
| 69 |
+
3,
|
| 70 |
+
4,
|
| 71 |
+
1
|
| 72 |
+
],
|
| 73 |
+
"num_channels": 3,
|
| 74 |
+
"num_hidden_layers": 64,
|
| 75 |
+
"out_channels": [
|
| 76 |
+
16,
|
| 77 |
+
24,
|
| 78 |
+
40,
|
| 79 |
+
80,
|
| 80 |
+
112,
|
| 81 |
+
192,
|
| 82 |
+
320
|
| 83 |
+
],
|
| 84 |
+
"pooling_type": "mean",
|
| 85 |
+
"problem_type": "single_label_classification",
|
| 86 |
+
"squeeze_expansion_ratio": 0.25,
|
| 87 |
+
"strides": [
|
| 88 |
+
1,
|
| 89 |
+
2,
|
| 90 |
+
2,
|
| 91 |
+
2,
|
| 92 |
+
1,
|
| 93 |
+
2,
|
| 94 |
+
1
|
| 95 |
+
],
|
| 96 |
+
"torch_dtype": "float32",
|
| 97 |
+
"transformers_version": "4.36.2",
|
| 98 |
+
"width_coefficient": 2.0
|
| 99 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04822af36520552107c63f30eea7fa616b6f1d709ed81b7ac1eb275d66d24d7e
|
| 3 |
+
size 256631328
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d963fc8c26e686b48a8bef54d7768b6ae6a19ba5759f6b7017488101c17d8e5
|
| 3 |
+
size 511057319
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:889de10fb992b121100f458ef9f2f4499ecc7b00accce13dcd308479c6f90e30
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:795aae61e054f7c273a548f291a22c10185e466cd7af31fed7886b963079b475
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,471 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6808762581409118,
|
| 3 |
+
"best_model_checkpoint": "Distilled-Melanoma-Classification-EfficientNet/checkpoint-63325",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 75990,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"learning_rate": 4.8333333333333334e-05,
|
| 14 |
+
"loss": 1.8409,
|
| 15 |
+
"step": 2533
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"eval_accuracy": 0.5792382080126307,
|
| 20 |
+
"eval_loss": 1.383086085319519,
|
| 21 |
+
"eval_runtime": 438.7758,
|
| 22 |
+
"eval_samples_per_second": 11.548,
|
| 23 |
+
"eval_steps_per_second": 1.445,
|
| 24 |
+
"step": 2533
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.0,
|
| 28 |
+
"learning_rate": 4.666666666666667e-05,
|
| 29 |
+
"loss": 1.549,
|
| 30 |
+
"step": 5066
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 2.0,
|
| 34 |
+
"eval_accuracy": 0.5859482928754687,
|
| 35 |
+
"eval_loss": 1.3424272537231445,
|
| 36 |
+
"eval_runtime": 452.7141,
|
| 37 |
+
"eval_samples_per_second": 11.192,
|
| 38 |
+
"eval_steps_per_second": 1.4,
|
| 39 |
+
"step": 5066
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 3.0,
|
| 43 |
+
"learning_rate": 4.5e-05,
|
| 44 |
+
"loss": 1.45,
|
| 45 |
+
"step": 7599
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 3.0,
|
| 49 |
+
"eval_accuracy": 0.6295638444839156,
|
| 50 |
+
"eval_loss": 1.33952796459198,
|
| 51 |
+
"eval_runtime": 461.2274,
|
| 52 |
+
"eval_samples_per_second": 10.986,
|
| 53 |
+
"eval_steps_per_second": 1.375,
|
| 54 |
+
"step": 7599
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 4.0,
|
| 58 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 59 |
+
"loss": 1.3614,
|
| 60 |
+
"step": 10132
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"eval_accuracy": 0.6277876455496348,
|
| 65 |
+
"eval_loss": 1.282675862312317,
|
| 66 |
+
"eval_runtime": 466.1134,
|
| 67 |
+
"eval_samples_per_second": 10.871,
|
| 68 |
+
"eval_steps_per_second": 1.36,
|
| 69 |
+
"step": 10132
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 5.0,
|
| 73 |
+
"learning_rate": 4.166666666666667e-05,
|
| 74 |
+
"loss": 1.2856,
|
| 75 |
+
"step": 12665
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.0,
|
| 79 |
+
"eval_accuracy": 0.6481152555752911,
|
| 80 |
+
"eval_loss": 1.0961527824401855,
|
| 81 |
+
"eval_runtime": 474.1304,
|
| 82 |
+
"eval_samples_per_second": 10.687,
|
| 83 |
+
"eval_steps_per_second": 1.337,
|
| 84 |
+
"step": 12665
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.0,
|
| 88 |
+
"learning_rate": 4e-05,
|
| 89 |
+
"loss": 1.2093,
|
| 90 |
+
"step": 15198
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 6.0,
|
| 94 |
+
"eval_accuracy": 0.6502861653838563,
|
| 95 |
+
"eval_loss": 1.1280168294906616,
|
| 96 |
+
"eval_runtime": 453.7056,
|
| 97 |
+
"eval_samples_per_second": 11.168,
|
| 98 |
+
"eval_steps_per_second": 1.397,
|
| 99 |
+
"step": 15198
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 7.0,
|
| 103 |
+
"learning_rate": 3.8333333333333334e-05,
|
| 104 |
+
"loss": 1.1307,
|
| 105 |
+
"step": 17731
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"eval_accuracy": 0.6599565818038287,
|
| 110 |
+
"eval_loss": 1.045525312423706,
|
| 111 |
+
"eval_runtime": 440.9718,
|
| 112 |
+
"eval_samples_per_second": 11.491,
|
| 113 |
+
"eval_steps_per_second": 1.438,
|
| 114 |
+
"step": 17731
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 8.0,
|
| 118 |
+
"learning_rate": 3.6666666666666666e-05,
|
| 119 |
+
"loss": 1.043,
|
| 120 |
+
"step": 20264
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"eval_accuracy": 0.6623248470495362,
|
| 125 |
+
"eval_loss": 1.115827202796936,
|
| 126 |
+
"eval_runtime": 441.2314,
|
| 127 |
+
"eval_samples_per_second": 11.484,
|
| 128 |
+
"eval_steps_per_second": 1.437,
|
| 129 |
+
"step": 20264
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 9.0,
|
| 133 |
+
"learning_rate": 3.5e-05,
|
| 134 |
+
"loss": 0.9456,
|
| 135 |
+
"step": 22797
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 9.0,
|
| 139 |
+
"eval_accuracy": 0.6562068284981252,
|
| 140 |
+
"eval_loss": 1.2251527309417725,
|
| 141 |
+
"eval_runtime": 449.8822,
|
| 142 |
+
"eval_samples_per_second": 11.263,
|
| 143 |
+
"eval_steps_per_second": 1.409,
|
| 144 |
+
"step": 22797
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 10.0,
|
| 148 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 149 |
+
"loss": 0.8582,
|
| 150 |
+
"step": 25330
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 10.0,
|
| 154 |
+
"eval_accuracy": 0.6573909611209788,
|
| 155 |
+
"eval_loss": 1.1615424156188965,
|
| 156 |
+
"eval_runtime": 447.1634,
|
| 157 |
+
"eval_samples_per_second": 11.331,
|
| 158 |
+
"eval_steps_per_second": 1.418,
|
| 159 |
+
"step": 25330
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 11.0,
|
| 163 |
+
"learning_rate": 3.1666666666666666e-05,
|
| 164 |
+
"loss": 0.7617,
|
| 165 |
+
"step": 27863
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 11.0,
|
| 169 |
+
"eval_accuracy": 0.6637063351095323,
|
| 170 |
+
"eval_loss": 1.1989778280258179,
|
| 171 |
+
"eval_runtime": 453.7466,
|
| 172 |
+
"eval_samples_per_second": 11.167,
|
| 173 |
+
"eval_steps_per_second": 1.397,
|
| 174 |
+
"step": 27863
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 12.0,
|
| 178 |
+
"learning_rate": 3e-05,
|
| 179 |
+
"loss": 0.6806,
|
| 180 |
+
"step": 30396
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 12.0,
|
| 184 |
+
"eval_accuracy": 0.658180382869548,
|
| 185 |
+
"eval_loss": 1.3517282009124756,
|
| 186 |
+
"eval_runtime": 458.6814,
|
| 187 |
+
"eval_samples_per_second": 11.047,
|
| 188 |
+
"eval_steps_per_second": 1.382,
|
| 189 |
+
"step": 30396
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 13.0,
|
| 193 |
+
"learning_rate": 2.8333333333333335e-05,
|
| 194 |
+
"loss": 0.6083,
|
| 195 |
+
"step": 32929
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 13.0,
|
| 199 |
+
"eval_accuracy": 0.6250246694296427,
|
| 200 |
+
"eval_loss": 1.7902203798294067,
|
| 201 |
+
"eval_runtime": 459.1268,
|
| 202 |
+
"eval_samples_per_second": 11.036,
|
| 203 |
+
"eval_steps_per_second": 1.381,
|
| 204 |
+
"step": 32929
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 14.0,
|
| 208 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 209 |
+
"loss": 0.5496,
|
| 210 |
+
"step": 35462
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 14.0,
|
| 214 |
+
"eval_accuracy": 0.6694296427866587,
|
| 215 |
+
"eval_loss": 1.3157224655151367,
|
| 216 |
+
"eval_runtime": 459.084,
|
| 217 |
+
"eval_samples_per_second": 11.037,
|
| 218 |
+
"eval_steps_per_second": 1.381,
|
| 219 |
+
"step": 35462
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 15.0,
|
| 223 |
+
"learning_rate": 2.5e-05,
|
| 224 |
+
"loss": 0.4897,
|
| 225 |
+
"step": 37995
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 15.0,
|
| 229 |
+
"eval_accuracy": 0.6510755871324255,
|
| 230 |
+
"eval_loss": 1.738616943359375,
|
| 231 |
+
"eval_runtime": 460.8902,
|
| 232 |
+
"eval_samples_per_second": 10.994,
|
| 233 |
+
"eval_steps_per_second": 1.376,
|
| 234 |
+
"step": 37995
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 16.0,
|
| 238 |
+
"learning_rate": 2.3333333333333336e-05,
|
| 239 |
+
"loss": 0.4478,
|
| 240 |
+
"step": 40528
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 16.0,
|
| 244 |
+
"eval_accuracy": 0.6560094730609828,
|
| 245 |
+
"eval_loss": 1.4945268630981445,
|
| 246 |
+
"eval_runtime": 459.5191,
|
| 247 |
+
"eval_samples_per_second": 11.027,
|
| 248 |
+
"eval_steps_per_second": 1.38,
|
| 249 |
+
"step": 40528
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 17.0,
|
| 253 |
+
"learning_rate": 2.1666666666666667e-05,
|
| 254 |
+
"loss": 0.4032,
|
| 255 |
+
"step": 43061
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 17.0,
|
| 259 |
+
"eval_accuracy": 0.6692322873495165,
|
| 260 |
+
"eval_loss": 1.2260057926177979,
|
| 261 |
+
"eval_runtime": 465.3259,
|
| 262 |
+
"eval_samples_per_second": 10.889,
|
| 263 |
+
"eval_steps_per_second": 1.362,
|
| 264 |
+
"step": 43061
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 18.0,
|
| 268 |
+
"learning_rate": 2e-05,
|
| 269 |
+
"loss": 0.3728,
|
| 270 |
+
"step": 45594
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 18.0,
|
| 274 |
+
"eval_accuracy": 0.6315373988553384,
|
| 275 |
+
"eval_loss": 1.3818838596343994,
|
| 276 |
+
"eval_runtime": 466.0275,
|
| 277 |
+
"eval_samples_per_second": 10.873,
|
| 278 |
+
"eval_steps_per_second": 1.36,
|
| 279 |
+
"step": 45594
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 19.0,
|
| 283 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 284 |
+
"loss": 0.3323,
|
| 285 |
+
"step": 48127
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 19.0,
|
| 289 |
+
"eval_accuracy": 0.6605486481152556,
|
| 290 |
+
"eval_loss": 1.3816238641738892,
|
| 291 |
+
"eval_runtime": 469.6622,
|
| 292 |
+
"eval_samples_per_second": 10.789,
|
| 293 |
+
"eval_steps_per_second": 1.35,
|
| 294 |
+
"step": 48127
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 20.0,
|
| 298 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 299 |
+
"loss": 0.3101,
|
| 300 |
+
"step": 50660
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 20.0,
|
| 304 |
+
"eval_accuracy": 0.6668640221038089,
|
| 305 |
+
"eval_loss": 1.2838419675827026,
|
| 306 |
+
"eval_runtime": 468.0234,
|
| 307 |
+
"eval_samples_per_second": 10.826,
|
| 308 |
+
"eval_steps_per_second": 1.355,
|
| 309 |
+
"step": 50660
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 21.0,
|
| 313 |
+
"learning_rate": 1.5e-05,
|
| 314 |
+
"loss": 0.2834,
|
| 315 |
+
"step": 53193
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 21.0,
|
| 319 |
+
"eval_accuracy": 0.6779159265837774,
|
| 320 |
+
"eval_loss": 1.144907832145691,
|
| 321 |
+
"eval_runtime": 467.7943,
|
| 322 |
+
"eval_samples_per_second": 10.832,
|
| 323 |
+
"eval_steps_per_second": 1.355,
|
| 324 |
+
"step": 53193
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 22.0,
|
| 328 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 329 |
+
"loss": 0.2604,
|
| 330 |
+
"step": 55726
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 22.0,
|
| 334 |
+
"eval_accuracy": 0.6639036905466745,
|
| 335 |
+
"eval_loss": 1.1774488687515259,
|
| 336 |
+
"eval_runtime": 466.3084,
|
| 337 |
+
"eval_samples_per_second": 10.866,
|
| 338 |
+
"eval_steps_per_second": 1.36,
|
| 339 |
+
"step": 55726
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 23.0,
|
| 343 |
+
"learning_rate": 1.1666666666666668e-05,
|
| 344 |
+
"loss": 0.2408,
|
| 345 |
+
"step": 58259
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 23.0,
|
| 349 |
+
"eval_accuracy": 0.670219064535228,
|
| 350 |
+
"eval_loss": 1.0753474235534668,
|
| 351 |
+
"eval_runtime": 472.9792,
|
| 352 |
+
"eval_samples_per_second": 10.713,
|
| 353 |
+
"eval_steps_per_second": 1.34,
|
| 354 |
+
"step": 58259
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 24.0,
|
| 358 |
+
"learning_rate": 1e-05,
|
| 359 |
+
"loss": 0.2247,
|
| 360 |
+
"step": 60792
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 24.0,
|
| 364 |
+
"eval_accuracy": 0.6692322873495165,
|
| 365 |
+
"eval_loss": 1.1826834678649902,
|
| 366 |
+
"eval_runtime": 459.776,
|
| 367 |
+
"eval_samples_per_second": 11.021,
|
| 368 |
+
"eval_steps_per_second": 1.379,
|
| 369 |
+
"step": 60792
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 25.0,
|
| 373 |
+
"learning_rate": 8.333333333333334e-06,
|
| 374 |
+
"loss": 0.2106,
|
| 375 |
+
"step": 63325
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 25.0,
|
| 379 |
+
"eval_accuracy": 0.6808762581409118,
|
| 380 |
+
"eval_loss": 1.1205148696899414,
|
| 381 |
+
"eval_runtime": 531.3791,
|
| 382 |
+
"eval_samples_per_second": 9.536,
|
| 383 |
+
"eval_steps_per_second": 1.193,
|
| 384 |
+
"step": 63325
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 26.0,
|
| 388 |
+
"learning_rate": 6.666666666666667e-06,
|
| 389 |
+
"loss": 0.1954,
|
| 390 |
+
"step": 65858
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 26.0,
|
| 394 |
+
"eval_accuracy": 0.6727846852180778,
|
| 395 |
+
"eval_loss": 1.1761797666549683,
|
| 396 |
+
"eval_runtime": 463.5443,
|
| 397 |
+
"eval_samples_per_second": 10.931,
|
| 398 |
+
"eval_steps_per_second": 1.368,
|
| 399 |
+
"step": 65858
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 27.0,
|
| 403 |
+
"learning_rate": 5e-06,
|
| 404 |
+
"loss": 0.1846,
|
| 405 |
+
"step": 68391
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 27.0,
|
| 409 |
+
"eval_accuracy": 0.6800868363923426,
|
| 410 |
+
"eval_loss": 1.0658093690872192,
|
| 411 |
+
"eval_runtime": 460.4726,
|
| 412 |
+
"eval_samples_per_second": 11.004,
|
| 413 |
+
"eval_steps_per_second": 1.377,
|
| 414 |
+
"step": 68391
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 28.0,
|
| 418 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 419 |
+
"loss": 0.1761,
|
| 420 |
+
"step": 70924
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 28.0,
|
| 424 |
+
"eval_accuracy": 0.6714031971580817,
|
| 425 |
+
"eval_loss": 1.2492362260818481,
|
| 426 |
+
"eval_runtime": 459.0542,
|
| 427 |
+
"eval_samples_per_second": 11.038,
|
| 428 |
+
"eval_steps_per_second": 1.381,
|
| 429 |
+
"step": 70924
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 29.0,
|
| 433 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 434 |
+
"loss": 0.1689,
|
| 435 |
+
"step": 73457
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 29.0,
|
| 439 |
+
"eval_accuracy": 0.6684428656009473,
|
| 440 |
+
"eval_loss": 1.0898562669754028,
|
| 441 |
+
"eval_runtime": 465.3023,
|
| 442 |
+
"eval_samples_per_second": 10.89,
|
| 443 |
+
"eval_steps_per_second": 1.363,
|
| 444 |
+
"step": 73457
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 30.0,
|
| 448 |
+
"learning_rate": 0.0,
|
| 449 |
+
"loss": 0.163,
|
| 450 |
+
"step": 75990
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 30.0,
|
| 454 |
+
"eval_accuracy": 0.6759423722123544,
|
| 455 |
+
"eval_loss": 1.1256953477859497,
|
| 456 |
+
"eval_runtime": 459.4721,
|
| 457 |
+
"eval_samples_per_second": 11.028,
|
| 458 |
+
"eval_steps_per_second": 1.38,
|
| 459 |
+
"step": 75990
|
| 460 |
+
}
|
| 461 |
+
],
|
| 462 |
+
"logging_steps": 500,
|
| 463 |
+
"max_steps": 75990,
|
| 464 |
+
"num_input_tokens_seen": 0,
|
| 465 |
+
"num_train_epochs": 30,
|
| 466 |
+
"save_steps": 500,
|
| 467 |
+
"total_flos": 3.503372757110489e+19,
|
| 468 |
+
"train_batch_size": 8,
|
| 469 |
+
"trial_name": null,
|
| 470 |
+
"trial_params": null
|
| 471 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0fa6160d92a99d193faea72b2c9e54c3a35ac04b040cb7bfdf9e8377262ad96
|
| 3 |
+
size 4792
|