Hugging Face's logo

mlahr
/
font-identifier

Image Classification
Transformers
Safetensors
resnet
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
 Community
1
mlahr commited on
Commit
7ba936d
·
verified ·
1 Parent(s): e66d09e

Model save

Browse files
Files changed (2) hide show
  1. README.md +58 -58
  2. model.safetensors +1 -1
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
- value: 0.9681818181818181
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 [microsoft/resnet-18](https://huggingface.co/microsoft/resnet-18) on the imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.1184
37
- - Accuracy: 0.9682
38
 
39
  ## Model description
40
 
@@ -59,70 +59,70 @@ The following hyperparameters were used during training:
59
  - seed: 42
60
  - gradient_accumulation_steps: 4
61
  - total_train_batch_size: 64
62
- - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
63
  - lr_scheduler_type: linear
64
  - lr_scheduler_warmup_ratio: 0.1
65
  - num_epochs: 50
66
 
67
  ### Training results
68
 
69
- | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
- |:-------------:|:-----:|:----:|:---------------:|:--------:|
71
- | 3.3213 | 1.0 | 14 | 3.2453 | 0.0682 |
72
- | 3.1711 | 2.0 | 28 | 3.0051 | 0.1273 |
73
- | 2.8729 | 3.0 | 42 | 2.6285 | 0.2091 |
74
- | 2.562 | 4.0 | 56 | 2.1600 | 0.3773 |
75
- | 1.8675 | 5.0 | 70 | 1.6392 | 0.5364 |
76
- | 1.6359 | 6.0 | 84 | 1.2267 | 0.6682 |
77
- | 1.3499 | 7.0 | 98 | 1.0588 | 0.6818 |
78
- | 1.076 | 8.0 | 112 | 0.8791 | 0.6909 |
79
- | 0.9811 | 9.0 | 126 | 0.7573 | 0.7545 |
80
- | 0.7309 | 10.0 | 140 | 0.6195 | 0.7818 |
81
- | 0.7776 | 11.0 | 154 | 0.5426 | 0.8 |
82
- | 0.7365 | 12.0 | 168 | 0.4029 | 0.8773 |
83
- | 0.5767 | 13.0 | 182 | 0.4418 | 0.8364 |
84
- | 0.5838 | 14.0 | 196 | 0.3538 | 0.8818 |
85
- | 0.4491 | 15.0 | 210 | 0.3834 | 0.8636 |
86
- | 0.5056 | 16.0 | 224 | 0.2701 | 0.9227 |
87
- | 0.4364 | 17.0 | 238 | 0.3142 | 0.8818 |
88
- | 0.364 | 18.0 | 252 | 0.2617 | 0.9136 |
89
- | 0.3845 | 19.0 | 266 | 0.3092 | 0.8818 |
90
- | 0.3873 | 20.0 | 280 | 0.2309 | 0.9136 |
91
- | 0.3397 | 21.0 | 294 | 0.2267 | 0.9182 |
92
- | 0.3731 | 22.0 | 308 | 0.2205 | 0.9136 |
93
- | 0.329 | 23.0 | 322 | 0.1516 | 0.95 |
94
- | 0.3041 | 24.0 | 336 | 0.2081 | 0.9318 |
95
- | 0.2996 | 25.0 | 350 | 0.1876 | 0.9273 |
96
- | 0.2825 | 26.0 | 364 | 0.2241 | 0.9273 |
97
- | 0.2929 | 27.0 | 378 | 0.2055 | 0.9318 |
98
- | 0.2574 | 28.0 | 392 | 0.1667 | 0.9318 |
99
- | 0.2662 | 29.0 | 406 | 0.1586 | 0.9545 |
100
- | 0.2391 | 30.0 | 420 | 0.1782 | 0.9273 |
101
- | 0.2642 | 31.0 | 434 | 0.1590 | 0.9409 |
102
- | 0.2323 | 32.0 | 448 | 0.1662 | 0.9364 |
103
- | 0.2261 | 33.0 | 462 | 0.1549 | 0.9455 |
104
- | 0.2116 | 34.0 | 476 | 0.1538 | 0.95 |
105
- | 0.211 | 35.0 | 490 | 0.1497 | 0.9636 |
106
- | 0.2472 | 36.0 | 504 | 0.1579 | 0.9591 |
107
- | 0.2185 | 37.0 | 518 | 0.1227 | 0.9636 |
108
- | 0.2123 | 38.0 | 532 | 0.1389 | 0.95 |
109
- | 0.1691 | 39.0 | 546 | 0.1040 | 0.9727 |
110
- | 0.1805 | 40.0 | 560 | 0.1445 | 0.9545 |
111
- | 0.1828 | 41.0 | 574 | 0.1349 | 0.9455 |
112
- | 0.2005 | 42.0 | 588 | 0.1418 | 0.9455 |
113
- | 0.1986 | 43.0 | 602 | 0.1613 | 0.9455 |
114
- | 0.2012 | 44.0 | 616 | 0.1206 | 0.9591 |
115
- | 0.1494 | 45.0 | 630 | 0.1405 | 0.9591 |
116
- | 0.1891 | 46.0 | 644 | 0.1122 | 0.9727 |
117
- | 0.2012 | 47.0 | 658 | 0.1215 | 0.9636 |
118
- | 0.181 | 48.0 | 672 | 0.1784 | 0.9455 |
119
- | 0.1757 | 49.0 | 686 | 0.1703 | 0.9364 |
120
- | 0.1603 | 50.0 | 700 | 0.1184 | 0.9682 |
121
 
122
 
123
  ### Framework versions
124
 
125
  - Transformers 4.53.3
126
- - Pytorch 2.2.2
127
- - Datasets 2.19.2
128
  - Tokenizers 0.21.2
 
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
+ value: 0.9207317073170732
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 [microsoft/resnet-18](https://huggingface.co/microsoft/resnet-18) on the imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.1982
37
+ - Accuracy: 0.9207
38
 
39
  ## Model description
40
 
 
59
  - seed: 42
60
  - gradient_accumulation_steps: 4
61
  - total_train_batch_size: 64
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: linear
64
  - lr_scheduler_warmup_ratio: 0.1
65
  - num_epochs: 50
66
 
67
  ### Training results
68
 
69
+ | Training Loss | Epoch | Step | Accuracy | Validation Loss |
70
+ |:-------------:|:-----:|:----:|:--------:|:---------------:|
71
+ | 5.578 | 1.0 | 154 | 0.0110 | 5.5483 |
72
+ | 5.1442 | 2.0 | 308 | 0.0630 | 4.9534 |
73
+ | 4.3147 | 3.0 | 462 | 0.1715 | 4.0494 |
74
+ | 3.6217 | 4.0 | 616 | 0.3154 | 3.1552 |
75
+ | 2.8758 | 5.0 | 770 | 0.4598 | 2.3913 |
76
+ | 2.5127 | 6.0 | 924 | 0.5736 | 1.8551 |
77
+ | 2.0255 | 7.0 | 1078 | 0.6480 | 1.4583 |
78
+ | 1.7498 | 8.0 | 1232 | 0.7301 | 1.1673 |
79
+ | 1.5827 | 9.0 | 1386 | 0.7545 | 0.9545 |
80
+ | 1.3878 | 10.0 | 1540 | 0.8 | 0.7889 |
81
+ | 1.2023 | 11.0 | 1694 | 0.8122 | 0.7008 |
82
+ | 1.0313 | 12.0 | 1848 | 0.8394 | 0.5885 |
83
+ | 0.9895 | 13.0 | 2002 | 0.8480 | 0.5341 |
84
+ | 0.8701 | 14.0 | 2156 | 0.8512 | 0.5032 |
85
+ | 0.7848 | 15.0 | 2310 | 0.8728 | 0.4448 |
86
+ | 0.8012 | 16.0 | 2464 | 0.8569 | 0.4308 |
87
+ | 0.6994 | 17.0 | 2618 | 0.8866 | 0.3686 |
88
+ | 0.6097 | 18.0 | 2772 | 0.8829 | 0.3650 |
89
+ | 0.5716 | 19.0 | 2926 | 0.8817 | 0.3683 |
90
+ | 0.5551 | 20.0 | 3080 | 0.8862 | 0.3425 |
91
+ | 0.6083 | 21.0 | 3234 | 0.9020 | 0.3025 |
92
+ | 0.5 | 22.0 | 3388 | 0.8980 | 0.3025 |
93
+ | 0.5356 | 23.0 | 3542 | 0.8931 | 0.2901 |
94
+ | 0.4927 | 24.0 | 3696 | 0.9057 | 0.2807 |
95
+ | 0.4471 | 25.0 | 3850 | 0.8972 | 0.2655 |
96
+ | 0.46 | 26.0 | 4004 | 0.9028 | 0.2563 |
97
+ | 0.4675 | 27.0 | 4158 | 0.9004 | 0.2587 |
98
+ | 0.3876 | 28.0 | 4312 | 0.9118 | 0.2395 |
99
+ | 0.3754 | 29.0 | 4466 | 0.9118 | 0.2324 |
100
+ | 0.3685 | 30.0 | 4620 | 0.2299 | 0.9110 |
101
+ | 0.4442 | 31.0 | 4774 | 0.2392 | 0.9093 |
102
+ | 0.3946 | 32.0 | 4928 | 0.2270 | 0.9085 |
103
+ | 0.3452 | 33.0 | 5082 | 0.2412 | 0.9089 |
104
+ | 0.4025 | 34.0 | 5236 | 0.2203 | 0.9138 |
105
+ | 0.3293 | 35.0 | 5390 | 0.2195 | 0.9102 |
106
+ | 0.3391 | 36.0 | 5544 | 0.2185 | 0.9154 |
107
+ | 0.3225 | 37.0 | 5698 | 0.2242 | 0.9085 |
108
+ | 0.2877 | 38.0 | 5852 | 0.2092 | 0.9183 |
109
+ | 0.3184 | 39.0 | 6006 | 0.2048 | 0.9175 |
110
+ | 0.2992 | 40.0 | 6160 | 0.2102 | 0.9163 |
111
+ | 0.2696 | 41.0 | 6314 | 0.2025 | 0.9114 |
112
+ | 0.3523 | 42.0 | 6468 | 0.2208 | 0.9077 |
113
+ | 0.3203 | 43.0 | 6622 | 0.2103 | 0.9195 |
114
+ | 0.2708 | 44.0 | 6776 | 0.2088 | 0.9150 |
115
+ | 0.2763 | 45.0 | 6930 | 0.1997 | 0.9191 |
116
+ | 0.2673 | 46.0 | 7084 | 0.1900 | 0.9195 |
117
+ | 0.2755 | 47.0 | 7238 | 0.1986 | 0.9154 |
118
+ | 0.2548 | 48.0 | 7392 | 0.2074 | 0.9138 |
119
+ | 0.2807 | 49.0 | 7546 | 0.1912 | 0.9211 |
120
+ | 0.2604 | 50.0 | 7700 | 0.1982 | 0.9207 |
121
 
122
 
123
  ### Framework versions
124
 
125
  - Transformers 4.53.3
126
+ - Pytorch 2.7.1
127
+ - Datasets 4.0.0
128
  - Tokenizers 0.21.2
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cde0c29e435556890041bf8fa8d3249e900a9ec395e12b477e2d195c36421f41
3
  size 45263012
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a2c647fc27b6750160c6c85e36d3214837b99f214dc2844c552d09f3fe63f22
3
  size 45263012