SeyedAli/Remote-Sensing-Image-Classification-VIT
Browse files- README.md +64 -0
- all_results.json +13 -0
- config.json +42 -0
- eval_results.json +8 -0
- preprocessor_config.json +22 -0
- pytorch_model.bin +3 -0
- train_results.json +8 -0
- trainer_state.json +409 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
model-index:
|
| 9 |
+
- name: Remote-Sensing-Classification-image-classification
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# Remote-Sensing-Classification-image-classification
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.1593
|
| 21 |
+
- Accuracy: 0.9589
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 16
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: linear
|
| 46 |
+
- num_epochs: 4
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 52 |
+
| 0.3922 | 0.71 | 100 | 0.4227 | 0.8821 |
|
| 53 |
+
| 0.2986 | 1.43 | 200 | 0.3142 | 0.9089 |
|
| 54 |
+
| 0.1109 | 2.14 | 300 | 0.2056 | 0.9518 |
|
| 55 |
+
| 0.0864 | 2.86 | 400 | 0.2472 | 0.9375 |
|
| 56 |
+
| 0.0397 | 3.57 | 500 | 0.1593 | 0.9589 |
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
### Framework versions
|
| 60 |
+
|
| 61 |
+
- Transformers 4.34.1
|
| 62 |
+
- Pytorch 2.1.0+cu118
|
| 63 |
+
- Datasets 2.14.5
|
| 64 |
+
- Tokenizers 0.14.1
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9589285714285715,
|
| 4 |
+
"eval_loss": 0.15928371250629425,
|
| 5 |
+
"eval_runtime": 10.4548,
|
| 6 |
+
"eval_samples_per_second": 53.564,
|
| 7 |
+
"eval_steps_per_second": 6.695,
|
| 8 |
+
"total_flos": 6.943593421681459e+17,
|
| 9 |
+
"train_loss": 0.23941477111407689,
|
| 10 |
+
"train_runtime": 355.1929,
|
| 11 |
+
"train_samples_per_second": 25.226,
|
| 12 |
+
"train_steps_per_second": 1.577
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "field",
|
| 13 |
+
"1": "forest",
|
| 14 |
+
"2": "grass",
|
| 15 |
+
"3": "industry",
|
| 16 |
+
"4": "parking",
|
| 17 |
+
"5": "resident",
|
| 18 |
+
"6": "river or lake"
|
| 19 |
+
},
|
| 20 |
+
"image_size": 224,
|
| 21 |
+
"initializer_range": 0.02,
|
| 22 |
+
"intermediate_size": 3072,
|
| 23 |
+
"label2id": {
|
| 24 |
+
"field": "0",
|
| 25 |
+
"forest": "1",
|
| 26 |
+
"grass": "2",
|
| 27 |
+
"industry": "3",
|
| 28 |
+
"parking": "4",
|
| 29 |
+
"resident": "5",
|
| 30 |
+
"river or lake": "6"
|
| 31 |
+
},
|
| 32 |
+
"layer_norm_eps": 1e-12,
|
| 33 |
+
"model_type": "vit",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_channels": 3,
|
| 36 |
+
"num_hidden_layers": 12,
|
| 37 |
+
"patch_size": 16,
|
| 38 |
+
"problem_type": "single_label_classification",
|
| 39 |
+
"qkv_bias": true,
|
| 40 |
+
"torch_dtype": "float32",
|
| 41 |
+
"transformers_version": "4.34.1"
|
| 42 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9589285714285715,
|
| 4 |
+
"eval_loss": 0.15928371250629425,
|
| 5 |
+
"eval_runtime": 10.4548,
|
| 6 |
+
"eval_samples_per_second": 53.564,
|
| 7 |
+
"eval_steps_per_second": 6.695
|
| 8 |
+
}
|
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": "ViTFeatureExtractor",
|
| 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 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2645713d76c85eeb70bd052d962df17e6692b35246219ddf8872b10ccbea02d9
|
| 3 |
+
size 343284522
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 6.943593421681459e+17,
|
| 4 |
+
"train_loss": 0.23941477111407689,
|
| 5 |
+
"train_runtime": 355.1929,
|
| 6 |
+
"train_samples_per_second": 25.226,
|
| 7 |
+
"train_steps_per_second": 1.577
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,409 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.15928371250629425,
|
| 3 |
+
"best_model_checkpoint": "./Remote-Sensing-Classification-image-classification/checkpoint-500",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 560,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 0.00019642857142857144,
|
| 14 |
+
"loss": 1.7404,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.14,
|
| 19 |
+
"learning_rate": 0.00019285714285714286,
|
| 20 |
+
"loss": 1.2945,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.21,
|
| 25 |
+
"learning_rate": 0.0001892857142857143,
|
| 26 |
+
"loss": 0.9461,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.29,
|
| 31 |
+
"learning_rate": 0.00018571428571428572,
|
| 32 |
+
"loss": 0.6914,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.36,
|
| 37 |
+
"learning_rate": 0.00018214285714285714,
|
| 38 |
+
"loss": 0.5938,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.43,
|
| 43 |
+
"learning_rate": 0.0001785714285714286,
|
| 44 |
+
"loss": 0.5389,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.5,
|
| 49 |
+
"learning_rate": 0.000175,
|
| 50 |
+
"loss": 0.522,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.57,
|
| 55 |
+
"learning_rate": 0.00017142857142857143,
|
| 56 |
+
"loss": 0.5365,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.64,
|
| 61 |
+
"learning_rate": 0.00016785714285714288,
|
| 62 |
+
"loss": 0.4732,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.71,
|
| 67 |
+
"learning_rate": 0.00016428571428571428,
|
| 68 |
+
"loss": 0.3922,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.71,
|
| 73 |
+
"eval_accuracy": 0.8821428571428571,
|
| 74 |
+
"eval_loss": 0.4226740300655365,
|
| 75 |
+
"eval_runtime": 9.2568,
|
| 76 |
+
"eval_samples_per_second": 60.496,
|
| 77 |
+
"eval_steps_per_second": 7.562,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.79,
|
| 82 |
+
"learning_rate": 0.00016071428571428573,
|
| 83 |
+
"loss": 0.3948,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.86,
|
| 88 |
+
"learning_rate": 0.00015714285714285716,
|
| 89 |
+
"loss": 0.5197,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.93,
|
| 94 |
+
"learning_rate": 0.0001535714285714286,
|
| 95 |
+
"loss": 0.2765,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.0,
|
| 100 |
+
"learning_rate": 0.00015000000000000001,
|
| 101 |
+
"loss": 0.3043,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.07,
|
| 106 |
+
"learning_rate": 0.00014642857142857141,
|
| 107 |
+
"loss": 0.136,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.14,
|
| 112 |
+
"learning_rate": 0.00014285714285714287,
|
| 113 |
+
"loss": 0.2342,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.21,
|
| 118 |
+
"learning_rate": 0.0001392857142857143,
|
| 119 |
+
"loss": 0.2023,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.29,
|
| 124 |
+
"learning_rate": 0.00013571428571428572,
|
| 125 |
+
"loss": 0.2333,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.36,
|
| 130 |
+
"learning_rate": 0.00013214285714285715,
|
| 131 |
+
"loss": 0.1803,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.43,
|
| 136 |
+
"learning_rate": 0.00012857142857142858,
|
| 137 |
+
"loss": 0.2986,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.43,
|
| 142 |
+
"eval_accuracy": 0.9089285714285714,
|
| 143 |
+
"eval_loss": 0.31418919563293457,
|
| 144 |
+
"eval_runtime": 13.8936,
|
| 145 |
+
"eval_samples_per_second": 40.306,
|
| 146 |
+
"eval_steps_per_second": 5.038,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.5,
|
| 151 |
+
"learning_rate": 0.000125,
|
| 152 |
+
"loss": 0.1593,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.57,
|
| 157 |
+
"learning_rate": 0.00012142857142857143,
|
| 158 |
+
"loss": 0.1688,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.64,
|
| 163 |
+
"learning_rate": 0.00011785714285714287,
|
| 164 |
+
"loss": 0.1953,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.71,
|
| 169 |
+
"learning_rate": 0.00011428571428571428,
|
| 170 |
+
"loss": 0.1932,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.79,
|
| 175 |
+
"learning_rate": 0.00011071428571428572,
|
| 176 |
+
"loss": 0.1448,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.86,
|
| 181 |
+
"learning_rate": 0.00010714285714285715,
|
| 182 |
+
"loss": 0.2531,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.93,
|
| 187 |
+
"learning_rate": 0.00010357142857142859,
|
| 188 |
+
"loss": 0.1877,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.0,
|
| 193 |
+
"learning_rate": 0.0001,
|
| 194 |
+
"loss": 0.1615,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.07,
|
| 199 |
+
"learning_rate": 9.642857142857143e-05,
|
| 200 |
+
"loss": 0.1966,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.14,
|
| 205 |
+
"learning_rate": 9.285714285714286e-05,
|
| 206 |
+
"loss": 0.1109,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.14,
|
| 211 |
+
"eval_accuracy": 0.9517857142857142,
|
| 212 |
+
"eval_loss": 0.20564210414886475,
|
| 213 |
+
"eval_runtime": 15.261,
|
| 214 |
+
"eval_samples_per_second": 36.695,
|
| 215 |
+
"eval_steps_per_second": 4.587,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.21,
|
| 220 |
+
"learning_rate": 8.92857142857143e-05,
|
| 221 |
+
"loss": 0.0765,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.29,
|
| 226 |
+
"learning_rate": 8.571428571428571e-05,
|
| 227 |
+
"loss": 0.0653,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.36,
|
| 232 |
+
"learning_rate": 8.214285714285714e-05,
|
| 233 |
+
"loss": 0.03,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.43,
|
| 238 |
+
"learning_rate": 7.857142857142858e-05,
|
| 239 |
+
"loss": 0.0783,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.5,
|
| 244 |
+
"learning_rate": 7.500000000000001e-05,
|
| 245 |
+
"loss": 0.0706,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.57,
|
| 250 |
+
"learning_rate": 7.142857142857143e-05,
|
| 251 |
+
"loss": 0.0635,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.64,
|
| 256 |
+
"learning_rate": 6.785714285714286e-05,
|
| 257 |
+
"loss": 0.0281,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.71,
|
| 262 |
+
"learning_rate": 6.428571428571429e-05,
|
| 263 |
+
"loss": 0.0636,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.79,
|
| 268 |
+
"learning_rate": 6.0714285714285715e-05,
|
| 269 |
+
"loss": 0.043,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.86,
|
| 274 |
+
"learning_rate": 5.714285714285714e-05,
|
| 275 |
+
"loss": 0.0864,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 2.86,
|
| 280 |
+
"eval_accuracy": 0.9375,
|
| 281 |
+
"eval_loss": 0.2472143918275833,
|
| 282 |
+
"eval_runtime": 8.8207,
|
| 283 |
+
"eval_samples_per_second": 63.487,
|
| 284 |
+
"eval_steps_per_second": 7.936,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.93,
|
| 289 |
+
"learning_rate": 5.3571428571428575e-05,
|
| 290 |
+
"loss": 0.1104,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 3.0,
|
| 295 |
+
"learning_rate": 5e-05,
|
| 296 |
+
"loss": 0.0393,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.07,
|
| 301 |
+
"learning_rate": 4.642857142857143e-05,
|
| 302 |
+
"loss": 0.0235,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 3.14,
|
| 307 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 308 |
+
"loss": 0.0191,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.21,
|
| 313 |
+
"learning_rate": 3.928571428571429e-05,
|
| 314 |
+
"loss": 0.0236,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.29,
|
| 319 |
+
"learning_rate": 3.571428571428572e-05,
|
| 320 |
+
"loss": 0.0448,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.36,
|
| 325 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 326 |
+
"loss": 0.0401,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.43,
|
| 331 |
+
"learning_rate": 2.857142857142857e-05,
|
| 332 |
+
"loss": 0.0186,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.5,
|
| 337 |
+
"learning_rate": 2.5e-05,
|
| 338 |
+
"loss": 0.0197,
|
| 339 |
+
"step": 490
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.57,
|
| 343 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 344 |
+
"loss": 0.0397,
|
| 345 |
+
"step": 500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.57,
|
| 349 |
+
"eval_accuracy": 0.9589285714285715,
|
| 350 |
+
"eval_loss": 0.15928371250629425,
|
| 351 |
+
"eval_runtime": 9.0766,
|
| 352 |
+
"eval_samples_per_second": 61.697,
|
| 353 |
+
"eval_steps_per_second": 7.712,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 3.64,
|
| 358 |
+
"learning_rate": 1.785714285714286e-05,
|
| 359 |
+
"loss": 0.0291,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 3.71,
|
| 364 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 365 |
+
"loss": 0.0182,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 3.79,
|
| 370 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 371 |
+
"loss": 0.0452,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.86,
|
| 376 |
+
"learning_rate": 7.142857142857143e-06,
|
| 377 |
+
"loss": 0.0173,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.93,
|
| 382 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 383 |
+
"loss": 0.0165,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 4.0,
|
| 388 |
+
"learning_rate": 0.0,
|
| 389 |
+
"loss": 0.0166,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 4.0,
|
| 394 |
+
"step": 560,
|
| 395 |
+
"total_flos": 6.943593421681459e+17,
|
| 396 |
+
"train_loss": 0.23941477111407689,
|
| 397 |
+
"train_runtime": 355.1929,
|
| 398 |
+
"train_samples_per_second": 25.226,
|
| 399 |
+
"train_steps_per_second": 1.577
|
| 400 |
+
}
|
| 401 |
+
],
|
| 402 |
+
"logging_steps": 10,
|
| 403 |
+
"max_steps": 560,
|
| 404 |
+
"num_train_epochs": 4,
|
| 405 |
+
"save_steps": 100,
|
| 406 |
+
"total_flos": 6.943593421681459e+17,
|
| 407 |
+
"trial_name": null,
|
| 408 |
+
"trial_params": null
|
| 409 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0aba89a8a6a87a7e39c00b1530c7d0d9b8e55c1e3d5f6772e893b7b01aaca8e9
|
| 3 |
+
size 4600
|