hushem_1x_deit_small_sgd_00001_fold2
This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 1.5076
- Accuracy: 0.1778
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
Training Loss |
Epoch |
Step |
Validation Loss |
Accuracy |
No log |
1.0 |
6 |
1.5128 |
0.1778 |
1.5351 |
2.0 |
12 |
1.5126 |
0.1778 |
1.5351 |
3.0 |
18 |
1.5123 |
0.1778 |
1.521 |
4.0 |
24 |
1.5120 |
0.1778 |
1.5462 |
5.0 |
30 |
1.5118 |
0.1778 |
1.5462 |
6.0 |
36 |
1.5116 |
0.1778 |
1.5099 |
7.0 |
42 |
1.5113 |
0.1778 |
1.5099 |
8.0 |
48 |
1.5111 |
0.1778 |
1.5333 |
9.0 |
54 |
1.5109 |
0.1778 |
1.5206 |
10.0 |
60 |
1.5106 |
0.1778 |
1.5206 |
11.0 |
66 |
1.5105 |
0.1778 |
1.5227 |
12.0 |
72 |
1.5103 |
0.1778 |
1.5227 |
13.0 |
78 |
1.5101 |
0.1778 |
1.5256 |
14.0 |
84 |
1.5099 |
0.1778 |
1.5395 |
15.0 |
90 |
1.5097 |
0.1778 |
1.5395 |
16.0 |
96 |
1.5095 |
0.1778 |
1.5169 |
17.0 |
102 |
1.5094 |
0.1778 |
1.5169 |
18.0 |
108 |
1.5092 |
0.1778 |
1.5502 |
19.0 |
114 |
1.5091 |
0.1778 |
1.4882 |
20.0 |
120 |
1.5090 |
0.1778 |
1.4882 |
21.0 |
126 |
1.5088 |
0.1778 |
1.5202 |
22.0 |
132 |
1.5087 |
0.1778 |
1.5202 |
23.0 |
138 |
1.5086 |
0.1778 |
1.5139 |
24.0 |
144 |
1.5085 |
0.1778 |
1.4995 |
25.0 |
150 |
1.5084 |
0.1778 |
1.4995 |
26.0 |
156 |
1.5083 |
0.1778 |
1.5175 |
27.0 |
162 |
1.5082 |
0.1778 |
1.5175 |
28.0 |
168 |
1.5081 |
0.1778 |
1.5365 |
29.0 |
174 |
1.5081 |
0.1778 |
1.5232 |
30.0 |
180 |
1.5080 |
0.1778 |
1.5232 |
31.0 |
186 |
1.5079 |
0.1778 |
1.5236 |
32.0 |
192 |
1.5079 |
0.1778 |
1.5236 |
33.0 |
198 |
1.5078 |
0.1778 |
1.5292 |
34.0 |
204 |
1.5078 |
0.1778 |
1.544 |
35.0 |
210 |
1.5077 |
0.1778 |
1.544 |
36.0 |
216 |
1.5077 |
0.1778 |
1.4971 |
37.0 |
222 |
1.5077 |
0.1778 |
1.4971 |
38.0 |
228 |
1.5077 |
0.1778 |
1.4951 |
39.0 |
234 |
1.5076 |
0.1778 |
1.5452 |
40.0 |
240 |
1.5076 |
0.1778 |
1.5452 |
41.0 |
246 |
1.5076 |
0.1778 |
1.5473 |
42.0 |
252 |
1.5076 |
0.1778 |
1.5473 |
43.0 |
258 |
1.5076 |
0.1778 |
1.5095 |
44.0 |
264 |
1.5076 |
0.1778 |
1.495 |
45.0 |
270 |
1.5076 |
0.1778 |
1.495 |
46.0 |
276 |
1.5076 |
0.1778 |
1.5118 |
47.0 |
282 |
1.5076 |
0.1778 |
1.5118 |
48.0 |
288 |
1.5076 |
0.1778 |
1.493 |
49.0 |
294 |
1.5076 |
0.1778 |
1.528 |
50.0 |
300 |
1.5076 |
0.1778 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1