heisenberg3376
commited on
Commit
•
a29d4d3
1
Parent(s):
2b23533
Training in progress, step 100
Browse files- README.md +82 -0
- all_results.json +13 -0
- config.json +50 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +36 -0
- runs/Jul17_09-22-53_405903fcfe02/events.out.tfevents.1721208182.405903fcfe02.739.0 +3 -0
- runs/Jul17_09-26-03_405903fcfe02/events.out.tfevents.1721208370.405903fcfe02.739.1 +3 -0
- runs/Jul17_09-26-03_405903fcfe02/events.out.tfevents.1721208623.405903fcfe02.739.2 +3 -0
- runs/Jul17_09-32-17_405903fcfe02/events.out.tfevents.1721208742.405903fcfe02.739.3 +3 -0
- train_results.json +8 -0
- trainer_state.json +516 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: google/vit-base-patch16-224-in21k
|
4 |
+
tags:
|
5 |
+
- image-classification
|
6 |
+
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- imagefolder
|
9 |
+
metrics:
|
10 |
+
- accuracy
|
11 |
+
model-index:
|
12 |
+
- name: vit-base-food-items-v1
|
13 |
+
results:
|
14 |
+
- task:
|
15 |
+
name: Image Classification
|
16 |
+
type: image-classification
|
17 |
+
dataset:
|
18 |
+
name: beans
|
19 |
+
type: imagefolder
|
20 |
+
config: default
|
21 |
+
split: validation
|
22 |
+
args: default
|
23 |
+
metrics:
|
24 |
+
- name: Accuracy
|
25 |
+
type: accuracy
|
26 |
+
value: 0.9236363636363636
|
27 |
+
---
|
28 |
+
|
29 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
30 |
+
should probably proofread and complete it, then remove this comment. -->
|
31 |
+
|
32 |
+
# vit-base-food-items-v1
|
33 |
+
|
34 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
|
35 |
+
It achieves the following results on the evaluation set:
|
36 |
+
- Loss: 0.3363
|
37 |
+
- Accuracy: 0.9236
|
38 |
+
|
39 |
+
## Model description
|
40 |
+
|
41 |
+
More information needed
|
42 |
+
|
43 |
+
## Intended uses & limitations
|
44 |
+
|
45 |
+
More information needed
|
46 |
+
|
47 |
+
## Training and evaluation data
|
48 |
+
|
49 |
+
More information needed
|
50 |
+
|
51 |
+
## Training procedure
|
52 |
+
|
53 |
+
### Training hyperparameters
|
54 |
+
|
55 |
+
The following hyperparameters were used during training:
|
56 |
+
- learning_rate: 0.0002
|
57 |
+
- train_batch_size: 16
|
58 |
+
- eval_batch_size: 8
|
59 |
+
- seed: 42
|
60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
+
- lr_scheduler_type: linear
|
62 |
+
- num_epochs: 4
|
63 |
+
- mixed_precision_training: Native AMP
|
64 |
+
|
65 |
+
### Training results
|
66 |
+
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
68 |
+
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
69 |
+
| 0.4195 | 0.6579 | 100 | 0.5028 | 0.9055 |
|
70 |
+
| 0.1072 | 1.3158 | 200 | 0.3794 | 0.8945 |
|
71 |
+
| 0.0326 | 1.9737 | 300 | 0.3832 | 0.9055 |
|
72 |
+
| 0.0207 | 2.6316 | 400 | 0.3363 | 0.9236 |
|
73 |
+
| 0.0167 | 3.2895 | 500 | 0.3373 | 0.9236 |
|
74 |
+
| 0.0153 | 3.9474 | 600 | 0.3374 | 0.9236 |
|
75 |
+
|
76 |
+
|
77 |
+
### Framework versions
|
78 |
+
|
79 |
+
- Transformers 4.41.2
|
80 |
+
- Pytorch 2.3.0+cu121
|
81 |
+
- Datasets 2.20.0
|
82 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9236363636363636,
|
4 |
+
"eval_loss": 0.33629149198532104,
|
5 |
+
"eval_runtime": 7.1163,
|
6 |
+
"eval_samples_per_second": 77.287,
|
7 |
+
"eval_steps_per_second": 9.696,
|
8 |
+
"total_flos": 7.501829674622976e+17,
|
9 |
+
"train_loss": 0.22265003621578217,
|
10 |
+
"train_runtime": 237.6059,
|
11 |
+
"train_samples_per_second": 40.74,
|
12 |
+
"train_steps_per_second": 2.559
|
13 |
+
}
|
config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
3 |
+
"architectures": [
|
4 |
+
"ViTForImageClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.0,
|
7 |
+
"encoder_stride": 16,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.0,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"id2label": {
|
12 |
+
"0": "Bread",
|
13 |
+
"1": "Dairy product",
|
14 |
+
"10": "Vegetable-Fruit",
|
15 |
+
"2": "Dessert",
|
16 |
+
"3": "Egg",
|
17 |
+
"4": "Fried food",
|
18 |
+
"5": "Meat",
|
19 |
+
"6": "Noodles-Pasta",
|
20 |
+
"7": "Rice",
|
21 |
+
"8": "Seafood",
|
22 |
+
"9": "Soup"
|
23 |
+
},
|
24 |
+
"image_size": 224,
|
25 |
+
"initializer_range": 0.02,
|
26 |
+
"intermediate_size": 3072,
|
27 |
+
"label2id": {
|
28 |
+
"Bread": "0",
|
29 |
+
"Dairy product": "1",
|
30 |
+
"Dessert": "2",
|
31 |
+
"Egg": "3",
|
32 |
+
"Fried food": "4",
|
33 |
+
"Meat": "5",
|
34 |
+
"Noodles-Pasta": "6",
|
35 |
+
"Rice": "7",
|
36 |
+
"Seafood": "8",
|
37 |
+
"Soup": "9",
|
38 |
+
"Vegetable-Fruit": "10"
|
39 |
+
},
|
40 |
+
"layer_norm_eps": 1e-12,
|
41 |
+
"model_type": "vit",
|
42 |
+
"num_attention_heads": 12,
|
43 |
+
"num_channels": 3,
|
44 |
+
"num_hidden_layers": 12,
|
45 |
+
"patch_size": 16,
|
46 |
+
"problem_type": "single_label_classification",
|
47 |
+
"qkv_bias": true,
|
48 |
+
"torch_dtype": "float32",
|
49 |
+
"transformers_version": "4.41.2"
|
50 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9236363636363636,
|
4 |
+
"eval_loss": 0.33629149198532104,
|
5 |
+
"eval_runtime": 7.1163,
|
6 |
+
"eval_samples_per_second": 77.287,
|
7 |
+
"eval_steps_per_second": 9.696
|
8 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:535d09cd8b39917b646d5dd2b752011055935c5cadd188604fb079078c7ebfac
|
3 |
+
size 343251660
|
preprocessor_config.json
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_valid_processor_keys": [
|
3 |
+
"images",
|
4 |
+
"do_resize",
|
5 |
+
"size",
|
6 |
+
"resample",
|
7 |
+
"do_rescale",
|
8 |
+
"rescale_factor",
|
9 |
+
"do_normalize",
|
10 |
+
"image_mean",
|
11 |
+
"image_std",
|
12 |
+
"return_tensors",
|
13 |
+
"data_format",
|
14 |
+
"input_data_format"
|
15 |
+
],
|
16 |
+
"do_normalize": true,
|
17 |
+
"do_rescale": true,
|
18 |
+
"do_resize": true,
|
19 |
+
"image_mean": [
|
20 |
+
0.5,
|
21 |
+
0.5,
|
22 |
+
0.5
|
23 |
+
],
|
24 |
+
"image_processor_type": "ViTFeatureExtractor",
|
25 |
+
"image_std": [
|
26 |
+
0.5,
|
27 |
+
0.5,
|
28 |
+
0.5
|
29 |
+
],
|
30 |
+
"resample": 2,
|
31 |
+
"rescale_factor": 0.00392156862745098,
|
32 |
+
"size": {
|
33 |
+
"height": 224,
|
34 |
+
"width": 224
|
35 |
+
}
|
36 |
+
}
|
runs/Jul17_09-22-53_405903fcfe02/events.out.tfevents.1721208182.405903fcfe02.739.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:768ad43b38f92547fb617b93c5d0c7a0c72408da27a37423585966f76adce2e3
|
3 |
+
size 5199
|
runs/Jul17_09-26-03_405903fcfe02/events.out.tfevents.1721208370.405903fcfe02.739.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5d8c7db1cb1d871d01c41ce1fa5720d9c5fe736e177b473d84cc974aaeac0326
|
3 |
+
size 20097
|
runs/Jul17_09-26-03_405903fcfe02/events.out.tfevents.1721208623.405903fcfe02.739.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:17832bd2c862108bbb3a63f77ed3caf302a7feb532bfea3e073df4931cc17cc3
|
3 |
+
size 411
|
runs/Jul17_09-32-17_405903fcfe02/events.out.tfevents.1721208742.405903fcfe02.739.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5bc9ff6ab6be4a8fc3dceedf6dec83cb4b3c2bd8e9db315c267f3e2a5331f772
|
3 |
+
size 7679
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"total_flos": 7.501829674622976e+17,
|
4 |
+
"train_loss": 0.22265003621578217,
|
5 |
+
"train_runtime": 237.6059,
|
6 |
+
"train_samples_per_second": 40.74,
|
7 |
+
"train_steps_per_second": 2.559
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,516 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.33629149198532104,
|
3 |
+
"best_model_checkpoint": "vit-base-food-items-v1/checkpoint-400",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 608,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.06578947368421052,
|
13 |
+
"grad_norm": 2.054168224334717,
|
14 |
+
"learning_rate": 0.00019671052631578949,
|
15 |
+
"loss": 2.2227,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.13157894736842105,
|
20 |
+
"grad_norm": 2.418569326400757,
|
21 |
+
"learning_rate": 0.00019342105263157894,
|
22 |
+
"loss": 1.7988,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.19736842105263158,
|
27 |
+
"grad_norm": 2.0799572467803955,
|
28 |
+
"learning_rate": 0.00019013157894736844,
|
29 |
+
"loss": 1.3952,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.2631578947368421,
|
34 |
+
"grad_norm": 2.5012855529785156,
|
35 |
+
"learning_rate": 0.00018684210526315792,
|
36 |
+
"loss": 1.0071,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.32894736842105265,
|
41 |
+
"grad_norm": 1.610549807548523,
|
42 |
+
"learning_rate": 0.00018355263157894736,
|
43 |
+
"loss": 0.8514,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.39473684210526316,
|
48 |
+
"grad_norm": 2.7514488697052,
|
49 |
+
"learning_rate": 0.00018026315789473684,
|
50 |
+
"loss": 0.6752,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.4605263157894737,
|
55 |
+
"grad_norm": 5.107870101928711,
|
56 |
+
"learning_rate": 0.00017697368421052632,
|
57 |
+
"loss": 0.617,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.5263157894736842,
|
62 |
+
"grad_norm": 1.621307611465454,
|
63 |
+
"learning_rate": 0.0001736842105263158,
|
64 |
+
"loss": 0.4383,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.5921052631578947,
|
69 |
+
"grad_norm": 2.050955057144165,
|
70 |
+
"learning_rate": 0.00017039473684210527,
|
71 |
+
"loss": 0.4703,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.6578947368421053,
|
76 |
+
"grad_norm": 3.5689868927001953,
|
77 |
+
"learning_rate": 0.00016710526315789475,
|
78 |
+
"loss": 0.4195,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.6578947368421053,
|
83 |
+
"eval_accuracy": 0.9054545454545454,
|
84 |
+
"eval_loss": 0.5027927756309509,
|
85 |
+
"eval_runtime": 6.6566,
|
86 |
+
"eval_samples_per_second": 82.625,
|
87 |
+
"eval_steps_per_second": 10.366,
|
88 |
+
"step": 100
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.7236842105263158,
|
92 |
+
"grad_norm": 2.683819055557251,
|
93 |
+
"learning_rate": 0.00016381578947368422,
|
94 |
+
"loss": 0.3666,
|
95 |
+
"step": 110
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.7894736842105263,
|
99 |
+
"grad_norm": 2.7733426094055176,
|
100 |
+
"learning_rate": 0.0001605263157894737,
|
101 |
+
"loss": 0.3876,
|
102 |
+
"step": 120
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 0.8552631578947368,
|
106 |
+
"grad_norm": 3.341937303543091,
|
107 |
+
"learning_rate": 0.00015723684210526318,
|
108 |
+
"loss": 0.3778,
|
109 |
+
"step": 130
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.9210526315789473,
|
113 |
+
"grad_norm": 1.0890475511550903,
|
114 |
+
"learning_rate": 0.00015394736842105265,
|
115 |
+
"loss": 0.3368,
|
116 |
+
"step": 140
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.9868421052631579,
|
120 |
+
"grad_norm": 3.217635154724121,
|
121 |
+
"learning_rate": 0.0001506578947368421,
|
122 |
+
"loss": 0.2434,
|
123 |
+
"step": 150
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 1.0526315789473684,
|
127 |
+
"grad_norm": 1.1362298727035522,
|
128 |
+
"learning_rate": 0.00014736842105263158,
|
129 |
+
"loss": 0.1537,
|
130 |
+
"step": 160
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 1.118421052631579,
|
134 |
+
"grad_norm": 0.3043310344219208,
|
135 |
+
"learning_rate": 0.00014407894736842106,
|
136 |
+
"loss": 0.1786,
|
137 |
+
"step": 170
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 1.1842105263157894,
|
141 |
+
"grad_norm": 0.36744824051856995,
|
142 |
+
"learning_rate": 0.00014078947368421053,
|
143 |
+
"loss": 0.1534,
|
144 |
+
"step": 180
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.25,
|
148 |
+
"grad_norm": 0.4088458716869354,
|
149 |
+
"learning_rate": 0.0001375,
|
150 |
+
"loss": 0.1273,
|
151 |
+
"step": 190
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 1.3157894736842106,
|
155 |
+
"grad_norm": 0.22641144692897797,
|
156 |
+
"learning_rate": 0.00013421052631578948,
|
157 |
+
"loss": 0.1072,
|
158 |
+
"step": 200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.3157894736842106,
|
162 |
+
"eval_accuracy": 0.8945454545454545,
|
163 |
+
"eval_loss": 0.37944725155830383,
|
164 |
+
"eval_runtime": 6.6833,
|
165 |
+
"eval_samples_per_second": 82.295,
|
166 |
+
"eval_steps_per_second": 10.324,
|
167 |
+
"step": 200
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.381578947368421,
|
171 |
+
"grad_norm": 0.14886893332004547,
|
172 |
+
"learning_rate": 0.00013092105263157893,
|
173 |
+
"loss": 0.0846,
|
174 |
+
"step": 210
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 1.4473684210526316,
|
178 |
+
"grad_norm": 0.17389647662639618,
|
179 |
+
"learning_rate": 0.00012763157894736844,
|
180 |
+
"loss": 0.0789,
|
181 |
+
"step": 220
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 1.513157894736842,
|
185 |
+
"grad_norm": 0.12492559105157852,
|
186 |
+
"learning_rate": 0.00012434210526315791,
|
187 |
+
"loss": 0.0605,
|
188 |
+
"step": 230
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.5789473684210527,
|
192 |
+
"grad_norm": 0.14732375741004944,
|
193 |
+
"learning_rate": 0.00012105263157894738,
|
194 |
+
"loss": 0.0867,
|
195 |
+
"step": 240
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 1.6447368421052633,
|
199 |
+
"grad_norm": 0.1113506406545639,
|
200 |
+
"learning_rate": 0.00011776315789473684,
|
201 |
+
"loss": 0.0436,
|
202 |
+
"step": 250
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 1.7105263157894737,
|
206 |
+
"grad_norm": 0.09813081473112106,
|
207 |
+
"learning_rate": 0.00011447368421052632,
|
208 |
+
"loss": 0.0416,
|
209 |
+
"step": 260
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.776315789473684,
|
213 |
+
"grad_norm": 6.826725006103516,
|
214 |
+
"learning_rate": 0.0001111842105263158,
|
215 |
+
"loss": 0.0514,
|
216 |
+
"step": 270
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 1.8421052631578947,
|
220 |
+
"grad_norm": 0.10619573295116425,
|
221 |
+
"learning_rate": 0.00010789473684210527,
|
222 |
+
"loss": 0.0601,
|
223 |
+
"step": 280
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 1.9078947368421053,
|
227 |
+
"grad_norm": 0.13959018886089325,
|
228 |
+
"learning_rate": 0.00010460526315789475,
|
229 |
+
"loss": 0.0454,
|
230 |
+
"step": 290
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 1.973684210526316,
|
234 |
+
"grad_norm": 0.08468258380889893,
|
235 |
+
"learning_rate": 0.00010131578947368421,
|
236 |
+
"loss": 0.0326,
|
237 |
+
"step": 300
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.973684210526316,
|
241 |
+
"eval_accuracy": 0.9054545454545454,
|
242 |
+
"eval_loss": 0.38323774933815,
|
243 |
+
"eval_runtime": 6.0691,
|
244 |
+
"eval_samples_per_second": 90.622,
|
245 |
+
"eval_steps_per_second": 11.369,
|
246 |
+
"step": 300
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 2.039473684210526,
|
250 |
+
"grad_norm": 0.07823757082223892,
|
251 |
+
"learning_rate": 9.802631578947369e-05,
|
252 |
+
"loss": 0.0392,
|
253 |
+
"step": 310
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"epoch": 2.1052631578947367,
|
257 |
+
"grad_norm": 0.07656868547201157,
|
258 |
+
"learning_rate": 9.473684210526316e-05,
|
259 |
+
"loss": 0.0288,
|
260 |
+
"step": 320
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 2.1710526315789473,
|
264 |
+
"grad_norm": 0.07013211399316788,
|
265 |
+
"learning_rate": 9.144736842105264e-05,
|
266 |
+
"loss": 0.0313,
|
267 |
+
"step": 330
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 2.236842105263158,
|
271 |
+
"grad_norm": 0.07913695275783539,
|
272 |
+
"learning_rate": 8.81578947368421e-05,
|
273 |
+
"loss": 0.0378,
|
274 |
+
"step": 340
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 2.3026315789473686,
|
278 |
+
"grad_norm": 0.3869466483592987,
|
279 |
+
"learning_rate": 8.486842105263159e-05,
|
280 |
+
"loss": 0.0253,
|
281 |
+
"step": 350
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 2.3684210526315788,
|
285 |
+
"grad_norm": 0.06490592658519745,
|
286 |
+
"learning_rate": 8.157894736842105e-05,
|
287 |
+
"loss": 0.0241,
|
288 |
+
"step": 360
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 2.4342105263157894,
|
292 |
+
"grad_norm": 0.06631086021661758,
|
293 |
+
"learning_rate": 7.828947368421053e-05,
|
294 |
+
"loss": 0.0231,
|
295 |
+
"step": 370
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 2.5,
|
299 |
+
"grad_norm": 0.05489266291260719,
|
300 |
+
"learning_rate": 7.500000000000001e-05,
|
301 |
+
"loss": 0.0218,
|
302 |
+
"step": 380
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 2.5657894736842106,
|
306 |
+
"grad_norm": 0.07426982372999191,
|
307 |
+
"learning_rate": 7.171052631578947e-05,
|
308 |
+
"loss": 0.0215,
|
309 |
+
"step": 390
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 2.6315789473684212,
|
313 |
+
"grad_norm": 0.063384510576725,
|
314 |
+
"learning_rate": 6.842105263157895e-05,
|
315 |
+
"loss": 0.0207,
|
316 |
+
"step": 400
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 2.6315789473684212,
|
320 |
+
"eval_accuracy": 0.9236363636363636,
|
321 |
+
"eval_loss": 0.33629149198532104,
|
322 |
+
"eval_runtime": 6.0608,
|
323 |
+
"eval_samples_per_second": 90.746,
|
324 |
+
"eval_steps_per_second": 11.385,
|
325 |
+
"step": 400
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 2.6973684210526314,
|
329 |
+
"grad_norm": 0.05782260745763779,
|
330 |
+
"learning_rate": 6.513157894736842e-05,
|
331 |
+
"loss": 0.0201,
|
332 |
+
"step": 410
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 2.763157894736842,
|
336 |
+
"grad_norm": 0.05535552278161049,
|
337 |
+
"learning_rate": 6.18421052631579e-05,
|
338 |
+
"loss": 0.0194,
|
339 |
+
"step": 420
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 2.8289473684210527,
|
343 |
+
"grad_norm": 0.05756945163011551,
|
344 |
+
"learning_rate": 5.855263157894737e-05,
|
345 |
+
"loss": 0.0191,
|
346 |
+
"step": 430
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 2.8947368421052633,
|
350 |
+
"grad_norm": 0.05671467259526253,
|
351 |
+
"learning_rate": 5.526315789473685e-05,
|
352 |
+
"loss": 0.0188,
|
353 |
+
"step": 440
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 2.9605263157894735,
|
357 |
+
"grad_norm": 0.05619660019874573,
|
358 |
+
"learning_rate": 5.197368421052632e-05,
|
359 |
+
"loss": 0.0183,
|
360 |
+
"step": 450
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 3.026315789473684,
|
364 |
+
"grad_norm": 0.05277419090270996,
|
365 |
+
"learning_rate": 4.868421052631579e-05,
|
366 |
+
"loss": 0.0177,
|
367 |
+
"step": 460
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 3.0921052631578947,
|
371 |
+
"grad_norm": 0.05281645059585571,
|
372 |
+
"learning_rate": 4.539473684210527e-05,
|
373 |
+
"loss": 0.0174,
|
374 |
+
"step": 470
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 3.1578947368421053,
|
378 |
+
"grad_norm": 0.06867770105600357,
|
379 |
+
"learning_rate": 4.210526315789474e-05,
|
380 |
+
"loss": 0.017,
|
381 |
+
"step": 480
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 3.223684210526316,
|
385 |
+
"grad_norm": 0.047292064875364304,
|
386 |
+
"learning_rate": 3.8815789473684214e-05,
|
387 |
+
"loss": 0.0168,
|
388 |
+
"step": 490
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 3.2894736842105265,
|
392 |
+
"grad_norm": 0.043311525136232376,
|
393 |
+
"learning_rate": 3.5526315789473684e-05,
|
394 |
+
"loss": 0.0167,
|
395 |
+
"step": 500
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.2894736842105265,
|
399 |
+
"eval_accuracy": 0.9236363636363636,
|
400 |
+
"eval_loss": 0.33733832836151123,
|
401 |
+
"eval_runtime": 5.7257,
|
402 |
+
"eval_samples_per_second": 96.057,
|
403 |
+
"eval_steps_per_second": 12.051,
|
404 |
+
"step": 500
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 3.3552631578947367,
|
408 |
+
"grad_norm": 0.04796218127012253,
|
409 |
+
"learning_rate": 3.223684210526316e-05,
|
410 |
+
"loss": 0.0165,
|
411 |
+
"step": 510
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 3.4210526315789473,
|
415 |
+
"grad_norm": 0.048424966633319855,
|
416 |
+
"learning_rate": 2.8947368421052634e-05,
|
417 |
+
"loss": 0.0163,
|
418 |
+
"step": 520
|
419 |
+
},
|
420 |
+
{
|
421 |
+
"epoch": 3.486842105263158,
|
422 |
+
"grad_norm": 0.046178512275218964,
|
423 |
+
"learning_rate": 2.565789473684211e-05,
|
424 |
+
"loss": 0.0157,
|
425 |
+
"step": 530
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 3.5526315789473686,
|
429 |
+
"grad_norm": 0.04182315245270729,
|
430 |
+
"learning_rate": 2.236842105263158e-05,
|
431 |
+
"loss": 0.0156,
|
432 |
+
"step": 540
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 3.6184210526315788,
|
436 |
+
"grad_norm": 0.04811399057507515,
|
437 |
+
"learning_rate": 1.9078947368421056e-05,
|
438 |
+
"loss": 0.0157,
|
439 |
+
"step": 550
|
440 |
+
},
|
441 |
+
{
|
442 |
+
"epoch": 3.6842105263157894,
|
443 |
+
"grad_norm": 0.04523231461644173,
|
444 |
+
"learning_rate": 1.5789473684210526e-05,
|
445 |
+
"loss": 0.0157,
|
446 |
+
"step": 560
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 3.75,
|
450 |
+
"grad_norm": 0.04799880087375641,
|
451 |
+
"learning_rate": 1.25e-05,
|
452 |
+
"loss": 0.0155,
|
453 |
+
"step": 570
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 3.8157894736842106,
|
457 |
+
"grad_norm": 0.04668057709932327,
|
458 |
+
"learning_rate": 9.210526315789474e-06,
|
459 |
+
"loss": 0.0154,
|
460 |
+
"step": 580
|
461 |
+
},
|
462 |
+
{
|
463 |
+
"epoch": 3.8815789473684212,
|
464 |
+
"grad_norm": 0.044472016394138336,
|
465 |
+
"learning_rate": 5.921052631578948e-06,
|
466 |
+
"loss": 0.0154,
|
467 |
+
"step": 590
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 3.9473684210526314,
|
471 |
+
"grad_norm": 0.05030672252178192,
|
472 |
+
"learning_rate": 2.631578947368421e-06,
|
473 |
+
"loss": 0.0153,
|
474 |
+
"step": 600
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 3.9473684210526314,
|
478 |
+
"eval_accuracy": 0.9236363636363636,
|
479 |
+
"eval_loss": 0.33738574385643005,
|
480 |
+
"eval_runtime": 6.0053,
|
481 |
+
"eval_samples_per_second": 91.586,
|
482 |
+
"eval_steps_per_second": 11.49,
|
483 |
+
"step": 600
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 4.0,
|
487 |
+
"step": 608,
|
488 |
+
"total_flos": 7.501829674622976e+17,
|
489 |
+
"train_loss": 0.22265003621578217,
|
490 |
+
"train_runtime": 237.6059,
|
491 |
+
"train_samples_per_second": 40.74,
|
492 |
+
"train_steps_per_second": 2.559
|
493 |
+
}
|
494 |
+
],
|
495 |
+
"logging_steps": 10,
|
496 |
+
"max_steps": 608,
|
497 |
+
"num_input_tokens_seen": 0,
|
498 |
+
"num_train_epochs": 4,
|
499 |
+
"save_steps": 100,
|
500 |
+
"stateful_callbacks": {
|
501 |
+
"TrainerControl": {
|
502 |
+
"args": {
|
503 |
+
"should_epoch_stop": false,
|
504 |
+
"should_evaluate": false,
|
505 |
+
"should_log": false,
|
506 |
+
"should_save": true,
|
507 |
+
"should_training_stop": false
|
508 |
+
},
|
509 |
+
"attributes": {}
|
510 |
+
}
|
511 |
+
},
|
512 |
+
"total_flos": 7.501829674622976e+17,
|
513 |
+
"train_batch_size": 16,
|
514 |
+
"trial_name": null,
|
515 |
+
"trial_params": null
|
516 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cff28f95376c6aecb268700119787cbf2428019aa507742603920dd75428cdb8
|
3 |
+
size 5112
|