End of training
Browse files- README.md +108 -0
- all_results.json +19 -0
- config.json +38 -0
- eval_results.json +13 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- train_results.json +8 -0
- trainer_state.json +401 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: google/vit-base-patch16-224
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- imagefolder
|
| 9 |
+
metrics:
|
| 10 |
+
- precision
|
| 11 |
+
- recall
|
| 12 |
+
- f1
|
| 13 |
+
- accuracy
|
| 14 |
+
model-index:
|
| 15 |
+
- name: vit-base-patch16-224-finetuned-barkley
|
| 16 |
+
results:
|
| 17 |
+
- task:
|
| 18 |
+
name: Image Classification
|
| 19 |
+
type: image-classification
|
| 20 |
+
dataset:
|
| 21 |
+
name: imagefolder
|
| 22 |
+
type: imagefolder
|
| 23 |
+
config: default
|
| 24 |
+
split: train
|
| 25 |
+
args: default
|
| 26 |
+
metrics:
|
| 27 |
+
- name: Precision
|
| 28 |
+
type: precision
|
| 29 |
+
value: 0.9936145510835913
|
| 30 |
+
- name: Recall
|
| 31 |
+
type: recall
|
| 32 |
+
value: 0.993421052631579
|
| 33 |
+
- name: F1
|
| 34 |
+
type: f1
|
| 35 |
+
value: 0.993419541966282
|
| 36 |
+
- name: Accuracy
|
| 37 |
+
type: accuracy
|
| 38 |
+
value: 0.9939393939393939
|
| 39 |
+
---
|
| 40 |
+
|
| 41 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 42 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 43 |
+
|
| 44 |
+
# vit-base-patch16-224-finetuned-barkley
|
| 45 |
+
|
| 46 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
| 47 |
+
It achieves the following results on the evaluation set:
|
| 48 |
+
- Loss: 0.0340
|
| 49 |
+
- Precision: 0.9936
|
| 50 |
+
- Recall: 0.9934
|
| 51 |
+
- F1: 0.9934
|
| 52 |
+
- Accuracy: 0.9939
|
| 53 |
+
- Top1 Accuracy: 0.9934
|
| 54 |
+
- Error Rate: 0.0061
|
| 55 |
+
|
| 56 |
+
## Model description
|
| 57 |
+
|
| 58 |
+
More information needed
|
| 59 |
+
|
| 60 |
+
## Intended uses & limitations
|
| 61 |
+
|
| 62 |
+
More information needed
|
| 63 |
+
|
| 64 |
+
## Training and evaluation data
|
| 65 |
+
|
| 66 |
+
More information needed
|
| 67 |
+
|
| 68 |
+
## Training procedure
|
| 69 |
+
|
| 70 |
+
### Training hyperparameters
|
| 71 |
+
|
| 72 |
+
The following hyperparameters were used during training:
|
| 73 |
+
- learning_rate: 0.0002
|
| 74 |
+
- train_batch_size: 32
|
| 75 |
+
- eval_batch_size: 32
|
| 76 |
+
- seed: 42
|
| 77 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 78 |
+
- lr_scheduler_type: linear
|
| 79 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 80 |
+
- num_epochs: 30
|
| 81 |
+
- mixed_precision_training: Native AMP
|
| 82 |
+
|
| 83 |
+
### Training results
|
| 84 |
+
|
| 85 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | Top1 Accuracy | Error Rate |
|
| 86 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|:-------------:|:----------:|
|
| 87 |
+
| 1.7463 | 1.0 | 38 | 1.7013 | 0.2143 | 0.2171 | 0.1930 | 0.2186 | 0.2171 | 0.7814 |
|
| 88 |
+
| 1.5581 | 2.0 | 76 | 1.4481 | 0.3512 | 0.3487 | 0.3287 | 0.3682 | 0.3487 | 0.6318 |
|
| 89 |
+
| 1.2665 | 3.0 | 114 | 1.0585 | 0.7397 | 0.7237 | 0.7274 | 0.7294 | 0.7237 | 0.2706 |
|
| 90 |
+
| 0.8572 | 4.0 | 152 | 0.5839 | 0.9467 | 0.9408 | 0.9417 | 0.9449 | 0.9408 | 0.0551 |
|
| 91 |
+
| 0.4337 | 5.0 | 190 | 0.2339 | 0.9820 | 0.9803 | 0.9802 | 0.9818 | 0.9803 | 0.0182 |
|
| 92 |
+
| 0.1569 | 6.0 | 228 | 0.0949 | 0.9739 | 0.9737 | 0.9735 | 0.9756 | 0.9737 | 0.0244 |
|
| 93 |
+
| 0.0577 | 7.0 | 266 | 0.0434 | 0.9872 | 0.9868 | 0.9867 | 0.9879 | 0.9868 | 0.0121 |
|
| 94 |
+
| 0.0172 | 8.0 | 304 | 0.0380 | 0.9870 | 0.9868 | 0.9868 | 0.9877 | 0.9868 | 0.0123 |
|
| 95 |
+
| 0.0208 | 9.0 | 342 | 0.0530 | 0.9876 | 0.9868 | 0.9868 | 0.9879 | 0.9868 | 0.0121 |
|
| 96 |
+
| 0.0071 | 10.0 | 380 | 0.0987 | 0.9716 | 0.9671 | 0.9669 | 0.9697 | 0.9671 | 0.0303 |
|
| 97 |
+
| 0.0062 | 11.0 | 418 | 0.0340 | 0.9936 | 0.9934 | 0.9934 | 0.9939 | 0.9934 | 0.0061 |
|
| 98 |
+
| 0.0165 | 12.0 | 456 | 0.0649 | 0.9809 | 0.9803 | 0.9799 | 0.9818 | 0.9803 | 0.0182 |
|
| 99 |
+
| 0.0057 | 13.0 | 494 | 0.0375 | 0.9936 | 0.9934 | 0.9934 | 0.9939 | 0.9934 | 0.0061 |
|
| 100 |
+
| 0.0038 | 14.0 | 532 | 0.0377 | 0.9936 | 0.9934 | 0.9934 | 0.9939 | 0.9934 | 0.0061 |
|
| 101 |
+
|
| 102 |
+
|
| 103 |
+
### Framework versions
|
| 104 |
+
|
| 105 |
+
- Transformers 4.45.2
|
| 106 |
+
- Pytorch 2.3.1+cu121
|
| 107 |
+
- Datasets 3.0.1
|
| 108 |
+
- Tokenizers 0.20.1
|
all_results.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"eval_accuracy": 0.9939393939393939,
|
| 4 |
+
"eval_error_rate": 0.0060606060606061,
|
| 5 |
+
"eval_f1": 0.993419541966282,
|
| 6 |
+
"eval_loss": 0.03395003080368042,
|
| 7 |
+
"eval_model_preparation_time": 0.008,
|
| 8 |
+
"eval_precision": 0.9936145510835913,
|
| 9 |
+
"eval_recall": 0.993421052631579,
|
| 10 |
+
"eval_runtime": 62.1415,
|
| 11 |
+
"eval_samples_per_second": 2.446,
|
| 12 |
+
"eval_steps_per_second": 0.08,
|
| 13 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 14 |
+
"total_flos": 1.319259102551212e+18,
|
| 15 |
+
"train_loss": 0.4395476876008779,
|
| 16 |
+
"train_runtime": 10817.4486,
|
| 17 |
+
"train_samples_per_second": 3.372,
|
| 18 |
+
"train_steps_per_second": 0.105
|
| 19 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224",
|
| 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": "Iinstia bijuga",
|
| 13 |
+
"1": "Mangifera indica",
|
| 14 |
+
"2": "Pterocarpus indicus",
|
| 15 |
+
"3": "Roystonea regia",
|
| 16 |
+
"4": "Tabebuia"
|
| 17 |
+
},
|
| 18 |
+
"image_size": 224,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"label2id": {
|
| 22 |
+
"Iinstia bijuga": 0,
|
| 23 |
+
"Mangifera indica": 1,
|
| 24 |
+
"Pterocarpus indicus": 2,
|
| 25 |
+
"Roystonea regia": 3,
|
| 26 |
+
"Tabebuia": 4
|
| 27 |
+
},
|
| 28 |
+
"layer_norm_eps": 1e-12,
|
| 29 |
+
"model_type": "vit",
|
| 30 |
+
"num_attention_heads": 12,
|
| 31 |
+
"num_channels": 3,
|
| 32 |
+
"num_hidden_layers": 12,
|
| 33 |
+
"patch_size": 16,
|
| 34 |
+
"problem_type": "single_label_classification",
|
| 35 |
+
"qkv_bias": true,
|
| 36 |
+
"torch_dtype": "float32",
|
| 37 |
+
"transformers_version": "4.45.2"
|
| 38 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"eval_accuracy": 0.9939393939393939,
|
| 4 |
+
"eval_error_rate": 0.0060606060606061,
|
| 5 |
+
"eval_f1": 0.993419541966282,
|
| 6 |
+
"eval_loss": 0.03395003080368042,
|
| 7 |
+
"eval_precision": 0.9936145510835913,
|
| 8 |
+
"eval_recall": 0.993421052631579,
|
| 9 |
+
"eval_runtime": 62.1415,
|
| 10 |
+
"eval_samples_per_second": 2.446,
|
| 11 |
+
"eval_steps_per_second": 0.08,
|
| 12 |
+
"eval_top1_accuracy": 0.993421052631579
|
| 13 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d8b7d00aba59d6176b6560b26fd8fafe32180a89f3d001465ff90a2d2a48ba0
|
| 3 |
+
size 343233204
|
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": "ViTImageProcessorFast",
|
| 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 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 1.319259102551212e+18,
|
| 4 |
+
"train_loss": 0.4395476876008779,
|
| 5 |
+
"train_runtime": 10817.4486,
|
| 6 |
+
"train_samples_per_second": 3.372,
|
| 7 |
+
"train_steps_per_second": 0.105
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,401 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.03395003080368042,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-finetuned-barkley\\checkpoint-418",
|
| 4 |
+
"epoch": 14.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 532,
|
| 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 |
+
"train_accuracy": 0.21134868421052633
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"epoch": 1.0,
|
| 17 |
+
"grad_norm": 7.52012300491333,
|
| 18 |
+
"learning_rate": 1.1659108139509125e-06,
|
| 19 |
+
"loss": 1.7463,
|
| 20 |
+
"step": 38
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0,
|
| 24 |
+
"eval_accuracy": 0.21856060606060607,
|
| 25 |
+
"eval_error_rate": 0.781439393939394,
|
| 26 |
+
"eval_f1": 0.19295495348680888,
|
| 27 |
+
"eval_loss": 1.701319694519043,
|
| 28 |
+
"eval_precision": 0.2142622862160236,
|
| 29 |
+
"eval_recall": 0.21710526315789475,
|
| 30 |
+
"eval_runtime": 46.6936,
|
| 31 |
+
"eval_samples_per_second": 3.255,
|
| 32 |
+
"eval_steps_per_second": 0.107,
|
| 33 |
+
"eval_top1_accuracy": 0.21710526315789475,
|
| 34 |
+
"step": 38
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"train_accuracy": 0.30409356725146197
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.0,
|
| 42 |
+
"grad_norm": 6.449554920196533,
|
| 43 |
+
"learning_rate": 2.23574935501902e-06,
|
| 44 |
+
"loss": 1.5581,
|
| 45 |
+
"step": 76
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.0,
|
| 49 |
+
"eval_accuracy": 0.36818181818181817,
|
| 50 |
+
"eval_error_rate": 0.6318181818181818,
|
| 51 |
+
"eval_f1": 0.328691576484973,
|
| 52 |
+
"eval_loss": 1.4481072425842285,
|
| 53 |
+
"eval_precision": 0.3511731503217571,
|
| 54 |
+
"eval_recall": 0.34868421052631576,
|
| 55 |
+
"eval_runtime": 63.7667,
|
| 56 |
+
"eval_samples_per_second": 2.384,
|
| 57 |
+
"eval_steps_per_second": 0.078,
|
| 58 |
+
"eval_top1_accuracy": 0.34868421052631576,
|
| 59 |
+
"step": 76
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"train_accuracy": 0.5197368421052632
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 3.0,
|
| 67 |
+
"grad_norm": 5.767495155334473,
|
| 68 |
+
"learning_rate": 3.927960312504679e-06,
|
| 69 |
+
"loss": 1.2665,
|
| 70 |
+
"step": 114
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 3.0,
|
| 74 |
+
"eval_accuracy": 0.7293560606060607,
|
| 75 |
+
"eval_error_rate": 0.27064393939393927,
|
| 76 |
+
"eval_f1": 0.7273905235235443,
|
| 77 |
+
"eval_loss": 1.0585086345672607,
|
| 78 |
+
"eval_precision": 0.7397410035407469,
|
| 79 |
+
"eval_recall": 0.7236842105263158,
|
| 80 |
+
"eval_runtime": 47.8118,
|
| 81 |
+
"eval_samples_per_second": 3.179,
|
| 82 |
+
"eval_steps_per_second": 0.105,
|
| 83 |
+
"eval_top1_accuracy": 0.7236842105263158,
|
| 84 |
+
"step": 114
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"train_accuracy": 0.8062865497076024
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 4.0,
|
| 92 |
+
"grad_norm": 12.979792594909668,
|
| 93 |
+
"learning_rate": 6.113544042901594e-06,
|
| 94 |
+
"loss": 0.8572,
|
| 95 |
+
"step": 152
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 4.0,
|
| 99 |
+
"eval_accuracy": 0.9448863636363637,
|
| 100 |
+
"eval_error_rate": 0.05511363636363631,
|
| 101 |
+
"eval_f1": 0.9416857901577765,
|
| 102 |
+
"eval_loss": 0.58392333984375,
|
| 103 |
+
"eval_precision": 0.9467451523545707,
|
| 104 |
+
"eval_recall": 0.9407894736842105,
|
| 105 |
+
"eval_runtime": 46.5589,
|
| 106 |
+
"eval_samples_per_second": 3.265,
|
| 107 |
+
"eval_steps_per_second": 0.107,
|
| 108 |
+
"eval_top1_accuracy": 0.9407894736842105,
|
| 109 |
+
"step": 152
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"train_accuracy": 0.9466374269005848
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.0,
|
| 117 |
+
"grad_norm": 4.086988925933838,
|
| 118 |
+
"learning_rate": 8.62589039584572e-06,
|
| 119 |
+
"loss": 0.4337,
|
| 120 |
+
"step": 190
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.0,
|
| 124 |
+
"eval_accuracy": 0.9818181818181818,
|
| 125 |
+
"eval_error_rate": 0.018181818181818188,
|
| 126 |
+
"eval_f1": 0.9802351026820783,
|
| 127 |
+
"eval_loss": 0.23391342163085938,
|
| 128 |
+
"eval_precision": 0.9819548872180451,
|
| 129 |
+
"eval_recall": 0.9802631578947368,
|
| 130 |
+
"eval_runtime": 48.9157,
|
| 131 |
+
"eval_samples_per_second": 3.107,
|
| 132 |
+
"eval_steps_per_second": 0.102,
|
| 133 |
+
"eval_top1_accuracy": 0.9802631578947368,
|
| 134 |
+
"step": 190
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"train_accuracy": 0.9846491228070176
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 6.0,
|
| 142 |
+
"grad_norm": 1.9588254690170288,
|
| 143 |
+
"learning_rate": 1.1273479642392808e-05,
|
| 144 |
+
"loss": 0.1569,
|
| 145 |
+
"step": 228
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 6.0,
|
| 149 |
+
"eval_accuracy": 0.9755681818181818,
|
| 150 |
+
"eval_error_rate": 0.024431818181818166,
|
| 151 |
+
"eval_f1": 0.9734739448002901,
|
| 152 |
+
"eval_loss": 0.09490276873111725,
|
| 153 |
+
"eval_precision": 0.9738519848439453,
|
| 154 |
+
"eval_recall": 0.9736842105263158,
|
| 155 |
+
"eval_runtime": 62.9862,
|
| 156 |
+
"eval_samples_per_second": 2.413,
|
| 157 |
+
"eval_steps_per_second": 0.079,
|
| 158 |
+
"eval_top1_accuracy": 0.9736842105263158,
|
| 159 |
+
"step": 228
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"train_accuracy": 0.9890350877192983
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 7.0,
|
| 167 |
+
"grad_norm": 4.820840835571289,
|
| 168 |
+
"learning_rate": 1.3854482295832083e-05,
|
| 169 |
+
"loss": 0.0577,
|
| 170 |
+
"step": 266
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 7.0,
|
| 174 |
+
"eval_accuracy": 0.9878787878787879,
|
| 175 |
+
"eval_error_rate": 0.012121212121212088,
|
| 176 |
+
"eval_f1": 0.9867362170674966,
|
| 177 |
+
"eval_loss": 0.04335561394691467,
|
| 178 |
+
"eval_precision": 0.9872349657566376,
|
| 179 |
+
"eval_recall": 0.9868421052631579,
|
| 180 |
+
"eval_runtime": 60.8368,
|
| 181 |
+
"eval_samples_per_second": 2.498,
|
| 182 |
+
"eval_steps_per_second": 0.082,
|
| 183 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 184 |
+
"step": 266
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"train_accuracy": 0.9978070175438597
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 8.0,
|
| 192 |
+
"grad_norm": 0.17172254621982574,
|
| 193 |
+
"learning_rate": 1.6172144859969913e-05,
|
| 194 |
+
"loss": 0.0172,
|
| 195 |
+
"step": 304
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 8.0,
|
| 199 |
+
"eval_accuracy": 0.9876893939393939,
|
| 200 |
+
"eval_error_rate": 0.012310606060606077,
|
| 201 |
+
"eval_f1": 0.9868389880173071,
|
| 202 |
+
"eval_loss": 0.038041990250349045,
|
| 203 |
+
"eval_precision": 0.9870356037151703,
|
| 204 |
+
"eval_recall": 0.9868421052631579,
|
| 205 |
+
"eval_runtime": 52.1078,
|
| 206 |
+
"eval_samples_per_second": 2.917,
|
| 207 |
+
"eval_steps_per_second": 0.096,
|
| 208 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 209 |
+
"step": 304
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"train_accuracy": 0.993421052631579
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 9.0,
|
| 217 |
+
"grad_norm": 0.35163307189941406,
|
| 218 |
+
"learning_rate": 1.8049788627450628e-05,
|
| 219 |
+
"loss": 0.0208,
|
| 220 |
+
"step": 342
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 9.0,
|
| 224 |
+
"eval_accuracy": 0.9878787878787879,
|
| 225 |
+
"eval_error_rate": 0.012121212121212088,
|
| 226 |
+
"eval_f1": 0.986835875199362,
|
| 227 |
+
"eval_loss": 0.05302947387099266,
|
| 228 |
+
"eval_precision": 0.9876160990712075,
|
| 229 |
+
"eval_recall": 0.9868421052631579,
|
| 230 |
+
"eval_runtime": 62.895,
|
| 231 |
+
"eval_samples_per_second": 2.417,
|
| 232 |
+
"eval_steps_per_second": 0.079,
|
| 233 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 234 |
+
"step": 342
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"train_accuracy": 0.9985380116959064
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 10.0,
|
| 242 |
+
"grad_norm": 0.4090166985988617,
|
| 243 |
+
"learning_rate": 1.9344278148426573e-05,
|
| 244 |
+
"loss": 0.0071,
|
| 245 |
+
"step": 380
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 10.0,
|
| 249 |
+
"eval_accuracy": 0.9696969696969697,
|
| 250 |
+
"eval_error_rate": 0.030303030303030276,
|
| 251 |
+
"eval_f1": 0.9669489564967301,
|
| 252 |
+
"eval_loss": 0.09868758916854858,
|
| 253 |
+
"eval_precision": 0.9715504978662873,
|
| 254 |
+
"eval_recall": 0.9671052631578947,
|
| 255 |
+
"eval_runtime": 62.4085,
|
| 256 |
+
"eval_samples_per_second": 2.436,
|
| 257 |
+
"eval_steps_per_second": 0.08,
|
| 258 |
+
"eval_top1_accuracy": 0.9671052631578947,
|
| 259 |
+
"step": 380
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 11.0,
|
| 263 |
+
"train_accuracy": 0.9948830409356725
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 11.0,
|
| 267 |
+
"grad_norm": 0.16305039823055267,
|
| 268 |
+
"learning_rate": 1.995693264909124e-05,
|
| 269 |
+
"loss": 0.0062,
|
| 270 |
+
"step": 418
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 11.0,
|
| 274 |
+
"eval_accuracy": 0.9939393939393939,
|
| 275 |
+
"eval_error_rate": 0.0060606060606061,
|
| 276 |
+
"eval_f1": 0.993419541966282,
|
| 277 |
+
"eval_loss": 0.03395003080368042,
|
| 278 |
+
"eval_precision": 0.9936145510835913,
|
| 279 |
+
"eval_recall": 0.993421052631579,
|
| 280 |
+
"eval_runtime": 47.4116,
|
| 281 |
+
"eval_samples_per_second": 3.206,
|
| 282 |
+
"eval_steps_per_second": 0.105,
|
| 283 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 284 |
+
"step": 418
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 12.0,
|
| 288 |
+
"train_accuracy": 0.993421052631579
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 12.0,
|
| 292 |
+
"grad_norm": 3.5277631282806396,
|
| 293 |
+
"learning_rate": 1.9969660582486496e-05,
|
| 294 |
+
"loss": 0.0165,
|
| 295 |
+
"step": 456
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 12.0,
|
| 299 |
+
"eval_accuracy": 0.9818181818181818,
|
| 300 |
+
"eval_error_rate": 0.018181818181818188,
|
| 301 |
+
"eval_f1": 0.979943544279758,
|
| 302 |
+
"eval_loss": 0.06494485586881638,
|
| 303 |
+
"eval_precision": 0.9808553804296839,
|
| 304 |
+
"eval_recall": 0.9802631578947368,
|
| 305 |
+
"eval_runtime": 56.3797,
|
| 306 |
+
"eval_samples_per_second": 2.696,
|
| 307 |
+
"eval_steps_per_second": 0.089,
|
| 308 |
+
"eval_top1_accuracy": 0.9802631578947368,
|
| 309 |
+
"step": 456
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 13.0,
|
| 313 |
+
"train_accuracy": 0.9956140350877193
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 13.0,
|
| 317 |
+
"grad_norm": 0.053868379443883896,
|
| 318 |
+
"learning_rate": 1.980785357262109e-05,
|
| 319 |
+
"loss": 0.0057,
|
| 320 |
+
"step": 494
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 13.0,
|
| 324 |
+
"eval_accuracy": 0.9939393939393939,
|
| 325 |
+
"eval_error_rate": 0.0060606060606061,
|
| 326 |
+
"eval_f1": 0.993419541966282,
|
| 327 |
+
"eval_loss": 0.03750297427177429,
|
| 328 |
+
"eval_precision": 0.9936145510835913,
|
| 329 |
+
"eval_recall": 0.993421052631579,
|
| 330 |
+
"eval_runtime": 61.2882,
|
| 331 |
+
"eval_samples_per_second": 2.48,
|
| 332 |
+
"eval_steps_per_second": 0.082,
|
| 333 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 334 |
+
"step": 494
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 14.0,
|
| 338 |
+
"train_accuracy": 0.9985380116959064
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 14.0,
|
| 342 |
+
"grad_norm": 0.02169680781662464,
|
| 343 |
+
"learning_rate": 1.950863907947058e-05,
|
| 344 |
+
"loss": 0.0038,
|
| 345 |
+
"step": 532
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 14.0,
|
| 349 |
+
"eval_accuracy": 0.9939393939393939,
|
| 350 |
+
"eval_error_rate": 0.0060606060606061,
|
| 351 |
+
"eval_f1": 0.993419541966282,
|
| 352 |
+
"eval_loss": 0.037650275975465775,
|
| 353 |
+
"eval_precision": 0.9936145510835913,
|
| 354 |
+
"eval_recall": 0.993421052631579,
|
| 355 |
+
"eval_runtime": 64.509,
|
| 356 |
+
"eval_samples_per_second": 2.356,
|
| 357 |
+
"eval_steps_per_second": 0.078,
|
| 358 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 359 |
+
"step": 532
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 14.0,
|
| 363 |
+
"step": 532,
|
| 364 |
+
"total_flos": 1.319259102551212e+18,
|
| 365 |
+
"train_loss": 0.4395476876008779,
|
| 366 |
+
"train_runtime": 10817.4486,
|
| 367 |
+
"train_samples_per_second": 3.372,
|
| 368 |
+
"train_steps_per_second": 0.105
|
| 369 |
+
}
|
| 370 |
+
],
|
| 371 |
+
"logging_steps": 500,
|
| 372 |
+
"max_steps": 1140,
|
| 373 |
+
"num_input_tokens_seen": 0,
|
| 374 |
+
"num_train_epochs": 30,
|
| 375 |
+
"save_steps": 500,
|
| 376 |
+
"stateful_callbacks": {
|
| 377 |
+
"EarlyStoppingCallback": {
|
| 378 |
+
"args": {
|
| 379 |
+
"early_stopping_patience": 3,
|
| 380 |
+
"early_stopping_threshold": 0.0
|
| 381 |
+
},
|
| 382 |
+
"attributes": {
|
| 383 |
+
"early_stopping_patience_counter": 3
|
| 384 |
+
}
|
| 385 |
+
},
|
| 386 |
+
"TrainerControl": {
|
| 387 |
+
"args": {
|
| 388 |
+
"should_epoch_stop": false,
|
| 389 |
+
"should_evaluate": false,
|
| 390 |
+
"should_log": false,
|
| 391 |
+
"should_save": true,
|
| 392 |
+
"should_training_stop": true
|
| 393 |
+
},
|
| 394 |
+
"attributes": {}
|
| 395 |
+
}
|
| 396 |
+
},
|
| 397 |
+
"total_flos": 1.319259102551212e+18,
|
| 398 |
+
"train_batch_size": 32,
|
| 399 |
+
"trial_name": null,
|
| 400 |
+
"trial_params": null
|
| 401 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5eb43bcfd2e0e3cbc7cac889051ecb5a72ca97601937778cbe86b8e6db9d07d1
|
| 3 |
+
size 5176
|