Hugging Face's logo

NekoJar
/
trainer_output

Image Classification
Transformers
Safetensors
vit
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
 Community
NekoJar commited on
Commit
e54e140
·
verified ·
1 Parent(s): 2ec6065

End of training

Browse files
Files changed (3) hide show
  1. README.md +16 -16
  2. all_results.json +6 -6
  3. eval_results.json +6 -6
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
- value: 0.682045102427268
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
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 imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.9329
37
- - Accuracy: 0.6820
38
 
39
  ## Model description
40
 
@@ -54,11 +54,11 @@ More information needed
54
 
55
  The following hyperparameters were used during training:
56
  - learning_rate: 3e-05
57
- - train_batch_size: 32
58
- - eval_batch_size: 32
59
  - seed: 42
60
  - gradient_accumulation_steps: 8
61
- - total_train_batch_size: 256
62
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
63
  - lr_scheduler_type: cosine
64
  - lr_scheduler_warmup_steps: 1000
@@ -68,16 +68,16 @@ The following hyperparameters were used during training:
68
 
69
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
  |:-------------:|:------:|:----:|:---------------:|:--------:|
71
- | 2.0769 | 0.9904 | 90 | 2.0618 | 0.2197 |
72
- | 1.9731 | 1.9904 | 180 | 1.8697 | 0.3720 |
73
- | 1.5738 | 2.9904 | 270 | 1.4953 | 0.5214 |
74
- | 1.275 | 3.9904 | 360 | 1.2592 | 0.5581 |
75
- | 1.1388 | 4.9904 | 450 | 1.1400 | 0.6044 |
76
- | 1.0925 | 5.9904 | 540 | 1.0855 | 0.6283 |
77
- | 1.0004 | 6.9904 | 630 | 1.0080 | 0.6552 |
78
- | 0.9632 | 7.9904 | 720 | 0.9834 | 0.6561 |
79
- | 0.9173 | 8.9904 | 810 | 0.9329 | 0.6820 |
80
- | 0.8631 | 9.9904 | 900 | 0.9259 | 0.6741 |
81
 
82
 
83
  ### Framework versions
 
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
+ value: 0.5126527801687037
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
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 imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 1.4766
37
+ - Accuracy: 0.5127
38
 
39
  ## Model description
40
 
 
54
 
55
  The following hyperparameters were used during training:
56
  - learning_rate: 3e-05
57
+ - train_batch_size: 128
58
+ - eval_batch_size: 128
59
  - seed: 42
60
  - gradient_accumulation_steps: 8
61
+ - total_train_batch_size: 1024
62
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
63
  - lr_scheduler_type: cosine
64
  - lr_scheduler_warmup_steps: 1000
 
68
 
69
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
  |:-------------:|:------:|:----:|:---------------:|:--------:|
71
+ | No log | 0.9670 | 22 | 2.0708 | 0.1449 |
72
+ | No log | 1.9670 | 44 | 2.0653 | 0.1663 |
73
+ | 2.0961 | 2.9670 | 66 | 2.0564 | 0.1931 |
74
+ | 2.0961 | 3.9670 | 88 | 2.0423 | 0.2350 |
75
+ | 2.0586 | 4.9670 | 110 | 2.0171 | 0.2823 |
76
+ | 2.0586 | 5.9670 | 132 | 1.9638 | 0.3305 |
77
+ | 1.9128 | 6.9670 | 154 | 1.8130 | 0.3968 |
78
+ | 1.9128 | 7.9670 | 176 | 1.6647 | 0.4278 |
79
+ | 1.9128 | 8.9670 | 198 | 1.5676 | 0.4844 |
80
+ | 1.6466 | 9.9670 | 220 | 1.4766 | 0.5127 |
81
 
82
 
83
  ### Framework versions
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 9.990371389270976,
3
- "eval_accuracy": 0.682045102427268,
4
- "eval_loss": 0.9329204559326172,
5
- "eval_runtime": 55.0992,
6
- "eval_samples_per_second": 105.428,
7
- "eval_steps_per_second": 3.303
8
  }
 
1
  {
2
+ "epoch": 9.967032967032967,
3
+ "eval_accuracy": 0.5126527801687037,
4
+ "eval_loss": 1.476570963859558,
5
+ "eval_runtime": 57.5461,
6
+ "eval_samples_per_second": 100.945,
7
+ "eval_steps_per_second": 0.799
8
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 9.990371389270976,
3
- "eval_accuracy": 0.682045102427268,
4
- "eval_loss": 0.9329204559326172,
5
- "eval_runtime": 55.0992,
6
- "eval_samples_per_second": 105.428,
7
- "eval_steps_per_second": 3.303
8
  }
 
1
  {
2
+ "epoch": 9.967032967032967,
3
+ "eval_accuracy": 0.5126527801687037,
4
+ "eval_loss": 1.476570963859558,
5
+ "eval_runtime": 57.5461,
6
+ "eval_samples_per_second": 100.945,
7
+ "eval_steps_per_second": 0.799
8
  }