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
59284c7
·
verified ·
1 Parent(s): 8157193

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-112/config.json +88 -0
  2. checkpoint-112/model.safetensors +3 -0
  3. checkpoint-112/optimizer.pt +3 -0
  4. checkpoint-112/rng_state.pth +3 -0
  5. checkpoint-112/scheduler.pt +3 -0
  6. checkpoint-112/trainer_state.json +183 -0
  7. checkpoint-112/training_args.bin +3 -0
  8. checkpoint-126/config.json +88 -0
  9. checkpoint-126/model.safetensors +3 -0
  10. checkpoint-126/optimizer.pt +3 -0
  11. checkpoint-126/rng_state.pth +3 -0
  12. checkpoint-126/scheduler.pt +3 -0
  13. checkpoint-126/trainer_state.json +199 -0
  14. checkpoint-126/training_args.bin +3 -0
  15. checkpoint-14/config.json +88 -0
  16. checkpoint-14/model.safetensors +3 -0
  17. checkpoint-14/optimizer.pt +3 -0
  18. checkpoint-14/rng_state.pth +3 -0
  19. checkpoint-14/scheduler.pt +3 -0
  20. checkpoint-14/trainer_state.json +50 -0
  21. checkpoint-14/training_args.bin +3 -0
  22. checkpoint-140/config.json +88 -0
  23. checkpoint-140/model.safetensors +3 -0
  24. checkpoint-140/optimizer.pt +3 -0
  25. checkpoint-140/rng_state.pth +3 -0
  26. checkpoint-140/scheduler.pt +3 -0
  27. checkpoint-140/trainer_state.json +222 -0
  28. checkpoint-140/training_args.bin +3 -0
  29. checkpoint-154/config.json +88 -0
  30. checkpoint-154/model.safetensors +3 -0
  31. checkpoint-154/optimizer.pt +3 -0
  32. checkpoint-154/rng_state.pth +3 -0
  33. checkpoint-154/scheduler.pt +3 -0
  34. checkpoint-154/trainer_state.json +238 -0
  35. checkpoint-154/training_args.bin +3 -0
  36. checkpoint-168/config.json +88 -0
  37. checkpoint-168/model.safetensors +3 -0
  38. checkpoint-168/optimizer.pt +3 -0
  39. checkpoint-168/rng_state.pth +3 -0
  40. checkpoint-168/scheduler.pt +3 -0
  41. checkpoint-168/trainer_state.json +254 -0
  42. checkpoint-168/training_args.bin +3 -0
  43. checkpoint-182/config.json +88 -0
  44. checkpoint-182/model.safetensors +3 -0
  45. checkpoint-182/optimizer.pt +3 -0
  46. checkpoint-182/rng_state.pth +3 -0
  47. checkpoint-182/scheduler.pt +3 -0
  48. checkpoint-182/trainer_state.json +277 -0
  49. checkpoint-182/training_args.bin +3 -0
  50. checkpoint-196/config.json +88 -0
checkpoint-112/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-112/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d50eaf903cc1b3effc3fa1b21c2dab0e7fe892c26d1327dba045066152cf28a3
3
+ size 44805376
checkpoint-112/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e1cdf3953f170e3f58eb6147e901bb3d799e51334b1229b8eab11a42eba11d9
3
+ size 89553018
checkpoint-112/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c71ba54f0646bd8898ac2647e53f2aebe77a1fb7add37d0e3732a9964a9ec9b
3
+ size 13990
checkpoint-112/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cafe8b04c0d7fda41811110fc07bf501996487f6b372108ae9857fe438b0444
3
+ size 1064
checkpoint-112/trainer_state.json ADDED
@@ -0,0 +1,183 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 112,
3
+ "best_metric": 0.6909090909090909,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-112",
5
+ "epoch": 8.0,
6
+ "eval_steps": 500,
7
+ "global_step": 112,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ },
28
+ {
29
+ "epoch": 1.4363636363636363,
30
+ "grad_norm": 9.360711097717285,
31
+ "learning_rate": 1.357142857142857e-05,
32
+ "loss": 3.1711,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.12727272727272726,
38
+ "eval_loss": 3.0051403045654297,
39
+ "eval_runtime": 2.0612,
40
+ "eval_samples_per_second": 106.735,
41
+ "eval_steps_per_second": 6.792,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 2.1454545454545455,
46
+ "grad_norm": 8.500679016113281,
47
+ "learning_rate": 2.0714285714285718e-05,
48
+ "loss": 2.983,
49
+ "step": 30
50
+ },
51
+ {
52
+ "epoch": 2.8727272727272726,
53
+ "grad_norm": 9.82868766784668,
54
+ "learning_rate": 2.785714285714286e-05,
55
+ "loss": 2.8729,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "eval_accuracy": 0.20909090909090908,
61
+ "eval_loss": 2.6284613609313965,
62
+ "eval_runtime": 2.0503,
63
+ "eval_samples_per_second": 107.302,
64
+ "eval_steps_per_second": 6.828,
65
+ "step": 42
66
+ },
67
+ {
68
+ "epoch": 3.581818181818182,
69
+ "grad_norm": 7.700014114379883,
70
+ "learning_rate": 3.5e-05,
71
+ "loss": 2.562,
72
+ "step": 50
73
+ },
74
+ {
75
+ "epoch": 4.0,
76
+ "eval_accuracy": 0.37727272727272726,
77
+ "eval_loss": 2.160020112991333,
78
+ "eval_runtime": 2.3625,
79
+ "eval_samples_per_second": 93.121,
80
+ "eval_steps_per_second": 5.926,
81
+ "step": 56
82
+ },
83
+ {
84
+ "epoch": 4.290909090909091,
85
+ "grad_norm": 7.527115821838379,
86
+ "learning_rate": 4.214285714285714e-05,
87
+ "loss": 2.2406,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "grad_norm": 5.579476833343506,
93
+ "learning_rate": 4.928571428571429e-05,
94
+ "loss": 1.8675,
95
+ "step": 70
96
+ },
97
+ {
98
+ "epoch": 5.0,
99
+ "eval_accuracy": 0.5363636363636364,
100
+ "eval_loss": 1.639161229133606,
101
+ "eval_runtime": 2.0879,
102
+ "eval_samples_per_second": 105.372,
103
+ "eval_steps_per_second": 6.705,
104
+ "step": 70
105
+ },
106
+ {
107
+ "epoch": 5.7272727272727275,
108
+ "grad_norm": 6.313197135925293,
109
+ "learning_rate": 4.928571428571429e-05,
110
+ "loss": 1.6359,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 6.0,
115
+ "eval_accuracy": 0.6681818181818182,
116
+ "eval_loss": 1.2266921997070312,
117
+ "eval_runtime": 2.1098,
118
+ "eval_samples_per_second": 104.274,
119
+ "eval_steps_per_second": 6.636,
120
+ "step": 84
121
+ },
122
+ {
123
+ "epoch": 6.4363636363636365,
124
+ "grad_norm": 5.8209004402160645,
125
+ "learning_rate": 4.84920634920635e-05,
126
+ "loss": 1.3499,
127
+ "step": 90
128
+ },
129
+ {
130
+ "epoch": 7.0,
131
+ "eval_accuracy": 0.6818181818181818,
132
+ "eval_loss": 1.0587564706802368,
133
+ "eval_runtime": 2.9116,
134
+ "eval_samples_per_second": 75.56,
135
+ "eval_steps_per_second": 4.808,
136
+ "step": 98
137
+ },
138
+ {
139
+ "epoch": 7.1454545454545455,
140
+ "grad_norm": 6.231616973876953,
141
+ "learning_rate": 4.7698412698412706e-05,
142
+ "loss": 1.2501,
143
+ "step": 100
144
+ },
145
+ {
146
+ "epoch": 7.872727272727273,
147
+ "grad_norm": 5.9844584465026855,
148
+ "learning_rate": 4.690476190476191e-05,
149
+ "loss": 1.076,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.6909090909090909,
155
+ "eval_loss": 0.8790816068649292,
156
+ "eval_runtime": 4.5841,
157
+ "eval_samples_per_second": 47.992,
158
+ "eval_steps_per_second": 3.054,
159
+ "step": 112
160
+ }
161
+ ],
162
+ "logging_steps": 10,
163
+ "max_steps": 700,
164
+ "num_input_tokens_seen": 0,
165
+ "num_train_epochs": 50,
166
+ "save_steps": 500,
167
+ "stateful_callbacks": {
168
+ "TrainerControl": {
169
+ "args": {
170
+ "should_epoch_stop": false,
171
+ "should_evaluate": false,
172
+ "should_log": false,
173
+ "should_save": true,
174
+ "should_training_stop": false
175
+ },
176
+ "attributes": {}
177
+ }
178
+ },
179
+ "total_flos": 7.113540645421056e+16,
180
+ "train_batch_size": 16,
181
+ "trial_name": null,
182
+ "trial_params": null
183
+ }
checkpoint-112/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-126/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-126/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0fcafc481fe312b26f9166559a5cd78ba68d154a07574aa461db33ae4c3c8e2
3
+ size 44805376
checkpoint-126/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7604f32daf7f9e84e695d7d7f0448189f1f3c08241f0e4fe1069f93d6edbad16
3
+ size 89553018
checkpoint-126/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08d0b15529f08e1f1286ee56f01d06fa36778d8027c24990355c44c99cb246d1
3
+ size 13990
checkpoint-126/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6c463e68e18848078b2bd871c08a210b46e64a4e7cc64000912e3bb50156f7e
3
+ size 1064
checkpoint-126/trainer_state.json ADDED
@@ -0,0 +1,199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 126,
3
+ "best_metric": 0.7545454545454545,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-126",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 126,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ },
28
+ {
29
+ "epoch": 1.4363636363636363,
30
+ "grad_norm": 9.360711097717285,
31
+ "learning_rate": 1.357142857142857e-05,
32
+ "loss": 3.1711,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.12727272727272726,
38
+ "eval_loss": 3.0051403045654297,
39
+ "eval_runtime": 2.0612,
40
+ "eval_samples_per_second": 106.735,
41
+ "eval_steps_per_second": 6.792,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 2.1454545454545455,
46
+ "grad_norm": 8.500679016113281,
47
+ "learning_rate": 2.0714285714285718e-05,
48
+ "loss": 2.983,
49
+ "step": 30
50
+ },
51
+ {
52
+ "epoch": 2.8727272727272726,
53
+ "grad_norm": 9.82868766784668,
54
+ "learning_rate": 2.785714285714286e-05,
55
+ "loss": 2.8729,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "eval_accuracy": 0.20909090909090908,
61
+ "eval_loss": 2.6284613609313965,
62
+ "eval_runtime": 2.0503,
63
+ "eval_samples_per_second": 107.302,
64
+ "eval_steps_per_second": 6.828,
65
+ "step": 42
66
+ },
67
+ {
68
+ "epoch": 3.581818181818182,
69
+ "grad_norm": 7.700014114379883,
70
+ "learning_rate": 3.5e-05,
71
+ "loss": 2.562,
72
+ "step": 50
73
+ },
74
+ {
75
+ "epoch": 4.0,
76
+ "eval_accuracy": 0.37727272727272726,
77
+ "eval_loss": 2.160020112991333,
78
+ "eval_runtime": 2.3625,
79
+ "eval_samples_per_second": 93.121,
80
+ "eval_steps_per_second": 5.926,
81
+ "step": 56
82
+ },
83
+ {
84
+ "epoch": 4.290909090909091,
85
+ "grad_norm": 7.527115821838379,
86
+ "learning_rate": 4.214285714285714e-05,
87
+ "loss": 2.2406,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "grad_norm": 5.579476833343506,
93
+ "learning_rate": 4.928571428571429e-05,
94
+ "loss": 1.8675,
95
+ "step": 70
96
+ },
97
+ {
98
+ "epoch": 5.0,
99
+ "eval_accuracy": 0.5363636363636364,
100
+ "eval_loss": 1.639161229133606,
101
+ "eval_runtime": 2.0879,
102
+ "eval_samples_per_second": 105.372,
103
+ "eval_steps_per_second": 6.705,
104
+ "step": 70
105
+ },
106
+ {
107
+ "epoch": 5.7272727272727275,
108
+ "grad_norm": 6.313197135925293,
109
+ "learning_rate": 4.928571428571429e-05,
110
+ "loss": 1.6359,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 6.0,
115
+ "eval_accuracy": 0.6681818181818182,
116
+ "eval_loss": 1.2266921997070312,
117
+ "eval_runtime": 2.1098,
118
+ "eval_samples_per_second": 104.274,
119
+ "eval_steps_per_second": 6.636,
120
+ "step": 84
121
+ },
122
+ {
123
+ "epoch": 6.4363636363636365,
124
+ "grad_norm": 5.8209004402160645,
125
+ "learning_rate": 4.84920634920635e-05,
126
+ "loss": 1.3499,
127
+ "step": 90
128
+ },
129
+ {
130
+ "epoch": 7.0,
131
+ "eval_accuracy": 0.6818181818181818,
132
+ "eval_loss": 1.0587564706802368,
133
+ "eval_runtime": 2.9116,
134
+ "eval_samples_per_second": 75.56,
135
+ "eval_steps_per_second": 4.808,
136
+ "step": 98
137
+ },
138
+ {
139
+ "epoch": 7.1454545454545455,
140
+ "grad_norm": 6.231616973876953,
141
+ "learning_rate": 4.7698412698412706e-05,
142
+ "loss": 1.2501,
143
+ "step": 100
144
+ },
145
+ {
146
+ "epoch": 7.872727272727273,
147
+ "grad_norm": 5.9844584465026855,
148
+ "learning_rate": 4.690476190476191e-05,
149
+ "loss": 1.076,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.6909090909090909,
155
+ "eval_loss": 0.8790816068649292,
156
+ "eval_runtime": 4.5841,
157
+ "eval_samples_per_second": 47.992,
158
+ "eval_steps_per_second": 3.054,
159
+ "step": 112
160
+ },
161
+ {
162
+ "epoch": 8.581818181818182,
163
+ "grad_norm": 5.29907751083374,
164
+ "learning_rate": 4.6111111111111115e-05,
165
+ "loss": 0.9811,
166
+ "step": 120
167
+ },
168
+ {
169
+ "epoch": 9.0,
170
+ "eval_accuracy": 0.7545454545454545,
171
+ "eval_loss": 0.7573429346084595,
172
+ "eval_runtime": 4.4185,
173
+ "eval_samples_per_second": 49.79,
174
+ "eval_steps_per_second": 3.168,
175
+ "step": 126
176
+ }
177
+ ],
178
+ "logging_steps": 10,
179
+ "max_steps": 700,
180
+ "num_input_tokens_seen": 0,
181
+ "num_train_epochs": 50,
182
+ "save_steps": 500,
183
+ "stateful_callbacks": {
184
+ "TrainerControl": {
185
+ "args": {
186
+ "should_epoch_stop": false,
187
+ "should_evaluate": false,
188
+ "should_log": false,
189
+ "should_save": true,
190
+ "should_training_stop": false
191
+ },
192
+ "attributes": {}
193
+ }
194
+ },
195
+ "total_flos": 8.002733226098688e+16,
196
+ "train_batch_size": 16,
197
+ "trial_name": null,
198
+ "trial_params": null
199
+ }
checkpoint-126/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-14/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-14/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17b801a1dadee481f447f87699b9993fb7c2af7032b5da5880d75383931f2ba8
3
+ size 44805376
checkpoint-14/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:764c756ea32b0ef4c0b253dc7957edfba7d6b17ddabfca32606dda72a89b4e53
3
+ size 89553018
checkpoint-14/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f31b8321cc5b493ab28d53ab857a0fc387b45c005ec910d38eab126e903578c5
3
+ size 13990
checkpoint-14/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eeff771eceba11ececa0a4a67b23421b70685132a4ed4ec5ec6b6cd2df076d18
3
+ size 1064
checkpoint-14/trainer_state.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 14,
3
+ "best_metric": 0.06818181818181818,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-14",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 14,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ }
28
+ ],
29
+ "logging_steps": 10,
30
+ "max_steps": 700,
31
+ "num_input_tokens_seen": 0,
32
+ "num_train_epochs": 50,
33
+ "save_steps": 500,
34
+ "stateful_callbacks": {
35
+ "TrainerControl": {
36
+ "args": {
37
+ "should_epoch_stop": false,
38
+ "should_evaluate": false,
39
+ "should_log": false,
40
+ "should_save": true,
41
+ "should_training_stop": false
42
+ },
43
+ "attributes": {}
44
+ }
45
+ },
46
+ "total_flos": 8891925806776320.0,
47
+ "train_batch_size": 16,
48
+ "trial_name": null,
49
+ "trial_params": null
50
+ }
checkpoint-14/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-140/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-140/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c904d0f15a96f314622f8ce9f71fda949f93a54202d1814cfa64964ba995f6f7
3
+ size 44805376
checkpoint-140/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a470cfbaa1eecc70cf1468db51ac1445bfa58d244222b7f72cbf19b57856fcac
3
+ size 89553018
checkpoint-140/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3d440455cc448424e4c6d4cab742fba6d6c19f6d704a5e22c29757362b0a072
3
+ size 13990
checkpoint-140/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c51865cefe2f0444539ed235523536c8c919b9eb21d351c1952f953fe87b0e4
3
+ size 1064
checkpoint-140/trainer_state.json ADDED
@@ -0,0 +1,222 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 140,
3
+ "best_metric": 0.7818181818181819,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-140",
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 140,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ },
28
+ {
29
+ "epoch": 1.4363636363636363,
30
+ "grad_norm": 9.360711097717285,
31
+ "learning_rate": 1.357142857142857e-05,
32
+ "loss": 3.1711,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.12727272727272726,
38
+ "eval_loss": 3.0051403045654297,
39
+ "eval_runtime": 2.0612,
40
+ "eval_samples_per_second": 106.735,
41
+ "eval_steps_per_second": 6.792,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 2.1454545454545455,
46
+ "grad_norm": 8.500679016113281,
47
+ "learning_rate": 2.0714285714285718e-05,
48
+ "loss": 2.983,
49
+ "step": 30
50
+ },
51
+ {
52
+ "epoch": 2.8727272727272726,
53
+ "grad_norm": 9.82868766784668,
54
+ "learning_rate": 2.785714285714286e-05,
55
+ "loss": 2.8729,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "eval_accuracy": 0.20909090909090908,
61
+ "eval_loss": 2.6284613609313965,
62
+ "eval_runtime": 2.0503,
63
+ "eval_samples_per_second": 107.302,
64
+ "eval_steps_per_second": 6.828,
65
+ "step": 42
66
+ },
67
+ {
68
+ "epoch": 3.581818181818182,
69
+ "grad_norm": 7.700014114379883,
70
+ "learning_rate": 3.5e-05,
71
+ "loss": 2.562,
72
+ "step": 50
73
+ },
74
+ {
75
+ "epoch": 4.0,
76
+ "eval_accuracy": 0.37727272727272726,
77
+ "eval_loss": 2.160020112991333,
78
+ "eval_runtime": 2.3625,
79
+ "eval_samples_per_second": 93.121,
80
+ "eval_steps_per_second": 5.926,
81
+ "step": 56
82
+ },
83
+ {
84
+ "epoch": 4.290909090909091,
85
+ "grad_norm": 7.527115821838379,
86
+ "learning_rate": 4.214285714285714e-05,
87
+ "loss": 2.2406,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "grad_norm": 5.579476833343506,
93
+ "learning_rate": 4.928571428571429e-05,
94
+ "loss": 1.8675,
95
+ "step": 70
96
+ },
97
+ {
98
+ "epoch": 5.0,
99
+ "eval_accuracy": 0.5363636363636364,
100
+ "eval_loss": 1.639161229133606,
101
+ "eval_runtime": 2.0879,
102
+ "eval_samples_per_second": 105.372,
103
+ "eval_steps_per_second": 6.705,
104
+ "step": 70
105
+ },
106
+ {
107
+ "epoch": 5.7272727272727275,
108
+ "grad_norm": 6.313197135925293,
109
+ "learning_rate": 4.928571428571429e-05,
110
+ "loss": 1.6359,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 6.0,
115
+ "eval_accuracy": 0.6681818181818182,
116
+ "eval_loss": 1.2266921997070312,
117
+ "eval_runtime": 2.1098,
118
+ "eval_samples_per_second": 104.274,
119
+ "eval_steps_per_second": 6.636,
120
+ "step": 84
121
+ },
122
+ {
123
+ "epoch": 6.4363636363636365,
124
+ "grad_norm": 5.8209004402160645,
125
+ "learning_rate": 4.84920634920635e-05,
126
+ "loss": 1.3499,
127
+ "step": 90
128
+ },
129
+ {
130
+ "epoch": 7.0,
131
+ "eval_accuracy": 0.6818181818181818,
132
+ "eval_loss": 1.0587564706802368,
133
+ "eval_runtime": 2.9116,
134
+ "eval_samples_per_second": 75.56,
135
+ "eval_steps_per_second": 4.808,
136
+ "step": 98
137
+ },
138
+ {
139
+ "epoch": 7.1454545454545455,
140
+ "grad_norm": 6.231616973876953,
141
+ "learning_rate": 4.7698412698412706e-05,
142
+ "loss": 1.2501,
143
+ "step": 100
144
+ },
145
+ {
146
+ "epoch": 7.872727272727273,
147
+ "grad_norm": 5.9844584465026855,
148
+ "learning_rate": 4.690476190476191e-05,
149
+ "loss": 1.076,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.6909090909090909,
155
+ "eval_loss": 0.8790816068649292,
156
+ "eval_runtime": 4.5841,
157
+ "eval_samples_per_second": 47.992,
158
+ "eval_steps_per_second": 3.054,
159
+ "step": 112
160
+ },
161
+ {
162
+ "epoch": 8.581818181818182,
163
+ "grad_norm": 5.29907751083374,
164
+ "learning_rate": 4.6111111111111115e-05,
165
+ "loss": 0.9811,
166
+ "step": 120
167
+ },
168
+ {
169
+ "epoch": 9.0,
170
+ "eval_accuracy": 0.7545454545454545,
171
+ "eval_loss": 0.7573429346084595,
172
+ "eval_runtime": 4.4185,
173
+ "eval_samples_per_second": 49.79,
174
+ "eval_steps_per_second": 3.168,
175
+ "step": 126
176
+ },
177
+ {
178
+ "epoch": 9.290909090909091,
179
+ "grad_norm": 5.92799711227417,
180
+ "learning_rate": 4.531746031746032e-05,
181
+ "loss": 0.8655,
182
+ "step": 130
183
+ },
184
+ {
185
+ "epoch": 10.0,
186
+ "grad_norm": 3.577127695083618,
187
+ "learning_rate": 4.4523809523809525e-05,
188
+ "loss": 0.7309,
189
+ "step": 140
190
+ },
191
+ {
192
+ "epoch": 10.0,
193
+ "eval_accuracy": 0.7818181818181819,
194
+ "eval_loss": 0.6194924712181091,
195
+ "eval_runtime": 4.0981,
196
+ "eval_samples_per_second": 53.683,
197
+ "eval_steps_per_second": 3.416,
198
+ "step": 140
199
+ }
200
+ ],
201
+ "logging_steps": 10,
202
+ "max_steps": 700,
203
+ "num_input_tokens_seen": 0,
204
+ "num_train_epochs": 50,
205
+ "save_steps": 500,
206
+ "stateful_callbacks": {
207
+ "TrainerControl": {
208
+ "args": {
209
+ "should_epoch_stop": false,
210
+ "should_evaluate": false,
211
+ "should_log": false,
212
+ "should_save": true,
213
+ "should_training_stop": false
214
+ },
215
+ "attributes": {}
216
+ }
217
+ },
218
+ "total_flos": 8.89192580677632e+16,
219
+ "train_batch_size": 16,
220
+ "trial_name": null,
221
+ "trial_params": null
222
+ }
checkpoint-140/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-154/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-154/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21a2bd22b841b7078ee33091014d4bf5a7146a7f35bd8ca1ffb0f9f01add2aed
3
+ size 44805376
checkpoint-154/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9addc9bac617b291b0abdff3447bff155953f1de9a772a685af01bbd6e8f233
3
+ size 89553018
checkpoint-154/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a29d3d61d48bbe3658978a32b0749a88ca50c99a737f8321cc85792846b1fe48
3
+ size 13990
checkpoint-154/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c163c3783012dd3c27dedcdcb77ea50be6a1acafde65225e3fdd7e016b7687b
3
+ size 1064
checkpoint-154/trainer_state.json ADDED
@@ -0,0 +1,238 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 154,
3
+ "best_metric": 0.8,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-154",
5
+ "epoch": 11.0,
6
+ "eval_steps": 500,
7
+ "global_step": 154,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ },
28
+ {
29
+ "epoch": 1.4363636363636363,
30
+ "grad_norm": 9.360711097717285,
31
+ "learning_rate": 1.357142857142857e-05,
32
+ "loss": 3.1711,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.12727272727272726,
38
+ "eval_loss": 3.0051403045654297,
39
+ "eval_runtime": 2.0612,
40
+ "eval_samples_per_second": 106.735,
41
+ "eval_steps_per_second": 6.792,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 2.1454545454545455,
46
+ "grad_norm": 8.500679016113281,
47
+ "learning_rate": 2.0714285714285718e-05,
48
+ "loss": 2.983,
49
+ "step": 30
50
+ },
51
+ {
52
+ "epoch": 2.8727272727272726,
53
+ "grad_norm": 9.82868766784668,
54
+ "learning_rate": 2.785714285714286e-05,
55
+ "loss": 2.8729,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "eval_accuracy": 0.20909090909090908,
61
+ "eval_loss": 2.6284613609313965,
62
+ "eval_runtime": 2.0503,
63
+ "eval_samples_per_second": 107.302,
64
+ "eval_steps_per_second": 6.828,
65
+ "step": 42
66
+ },
67
+ {
68
+ "epoch": 3.581818181818182,
69
+ "grad_norm": 7.700014114379883,
70
+ "learning_rate": 3.5e-05,
71
+ "loss": 2.562,
72
+ "step": 50
73
+ },
74
+ {
75
+ "epoch": 4.0,
76
+ "eval_accuracy": 0.37727272727272726,
77
+ "eval_loss": 2.160020112991333,
78
+ "eval_runtime": 2.3625,
79
+ "eval_samples_per_second": 93.121,
80
+ "eval_steps_per_second": 5.926,
81
+ "step": 56
82
+ },
83
+ {
84
+ "epoch": 4.290909090909091,
85
+ "grad_norm": 7.527115821838379,
86
+ "learning_rate": 4.214285714285714e-05,
87
+ "loss": 2.2406,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "grad_norm": 5.579476833343506,
93
+ "learning_rate": 4.928571428571429e-05,
94
+ "loss": 1.8675,
95
+ "step": 70
96
+ },
97
+ {
98
+ "epoch": 5.0,
99
+ "eval_accuracy": 0.5363636363636364,
100
+ "eval_loss": 1.639161229133606,
101
+ "eval_runtime": 2.0879,
102
+ "eval_samples_per_second": 105.372,
103
+ "eval_steps_per_second": 6.705,
104
+ "step": 70
105
+ },
106
+ {
107
+ "epoch": 5.7272727272727275,
108
+ "grad_norm": 6.313197135925293,
109
+ "learning_rate": 4.928571428571429e-05,
110
+ "loss": 1.6359,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 6.0,
115
+ "eval_accuracy": 0.6681818181818182,
116
+ "eval_loss": 1.2266921997070312,
117
+ "eval_runtime": 2.1098,
118
+ "eval_samples_per_second": 104.274,
119
+ "eval_steps_per_second": 6.636,
120
+ "step": 84
121
+ },
122
+ {
123
+ "epoch": 6.4363636363636365,
124
+ "grad_norm": 5.8209004402160645,
125
+ "learning_rate": 4.84920634920635e-05,
126
+ "loss": 1.3499,
127
+ "step": 90
128
+ },
129
+ {
130
+ "epoch": 7.0,
131
+ "eval_accuracy": 0.6818181818181818,
132
+ "eval_loss": 1.0587564706802368,
133
+ "eval_runtime": 2.9116,
134
+ "eval_samples_per_second": 75.56,
135
+ "eval_steps_per_second": 4.808,
136
+ "step": 98
137
+ },
138
+ {
139
+ "epoch": 7.1454545454545455,
140
+ "grad_norm": 6.231616973876953,
141
+ "learning_rate": 4.7698412698412706e-05,
142
+ "loss": 1.2501,
143
+ "step": 100
144
+ },
145
+ {
146
+ "epoch": 7.872727272727273,
147
+ "grad_norm": 5.9844584465026855,
148
+ "learning_rate": 4.690476190476191e-05,
149
+ "loss": 1.076,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.6909090909090909,
155
+ "eval_loss": 0.8790816068649292,
156
+ "eval_runtime": 4.5841,
157
+ "eval_samples_per_second": 47.992,
158
+ "eval_steps_per_second": 3.054,
159
+ "step": 112
160
+ },
161
+ {
162
+ "epoch": 8.581818181818182,
163
+ "grad_norm": 5.29907751083374,
164
+ "learning_rate": 4.6111111111111115e-05,
165
+ "loss": 0.9811,
166
+ "step": 120
167
+ },
168
+ {
169
+ "epoch": 9.0,
170
+ "eval_accuracy": 0.7545454545454545,
171
+ "eval_loss": 0.7573429346084595,
172
+ "eval_runtime": 4.4185,
173
+ "eval_samples_per_second": 49.79,
174
+ "eval_steps_per_second": 3.168,
175
+ "step": 126
176
+ },
177
+ {
178
+ "epoch": 9.290909090909091,
179
+ "grad_norm": 5.92799711227417,
180
+ "learning_rate": 4.531746031746032e-05,
181
+ "loss": 0.8655,
182
+ "step": 130
183
+ },
184
+ {
185
+ "epoch": 10.0,
186
+ "grad_norm": 3.577127695083618,
187
+ "learning_rate": 4.4523809523809525e-05,
188
+ "loss": 0.7309,
189
+ "step": 140
190
+ },
191
+ {
192
+ "epoch": 10.0,
193
+ "eval_accuracy": 0.7818181818181819,
194
+ "eval_loss": 0.6194924712181091,
195
+ "eval_runtime": 4.0981,
196
+ "eval_samples_per_second": 53.683,
197
+ "eval_steps_per_second": 3.416,
198
+ "step": 140
199
+ },
200
+ {
201
+ "epoch": 10.727272727272727,
202
+ "grad_norm": 5.252236843109131,
203
+ "learning_rate": 4.373015873015873e-05,
204
+ "loss": 0.7776,
205
+ "step": 150
206
+ },
207
+ {
208
+ "epoch": 11.0,
209
+ "eval_accuracy": 0.8,
210
+ "eval_loss": 0.5426079034805298,
211
+ "eval_runtime": 4.0014,
212
+ "eval_samples_per_second": 54.981,
213
+ "eval_steps_per_second": 3.499,
214
+ "step": 154
215
+ }
216
+ ],
217
+ "logging_steps": 10,
218
+ "max_steps": 700,
219
+ "num_input_tokens_seen": 0,
220
+ "num_train_epochs": 50,
221
+ "save_steps": 500,
222
+ "stateful_callbacks": {
223
+ "TrainerControl": {
224
+ "args": {
225
+ "should_epoch_stop": false,
226
+ "should_evaluate": false,
227
+ "should_log": false,
228
+ "should_save": true,
229
+ "should_training_stop": false
230
+ },
231
+ "attributes": {}
232
+ }
233
+ },
234
+ "total_flos": 9.781118387453952e+16,
235
+ "train_batch_size": 16,
236
+ "trial_name": null,
237
+ "trial_params": null
238
+ }
checkpoint-154/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-168/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-168/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:994db2ee4c064757cf621886f5d7442819b454425938e8b6c4f1205744c03228
3
+ size 44805376
checkpoint-168/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:742a6c62f4ba03d57b5e147a842189525839bf8928abdd2f4480aeaf6a2612ea
3
+ size 89553018
checkpoint-168/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c82a69e27a889a84e9dac29179c963d320dd413216cf009b463851e4cb2cec1
3
+ size 13990
checkpoint-168/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ec959d95150727eb55080b0fd5043f44374f20a9af18e119afa404cba10321d
3
+ size 1064
checkpoint-168/trainer_state.json ADDED
@@ -0,0 +1,254 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 168,
3
+ "best_metric": 0.8772727272727273,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-168",
5
+ "epoch": 12.0,
6
+ "eval_steps": 500,
7
+ "global_step": 168,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ },
28
+ {
29
+ "epoch": 1.4363636363636363,
30
+ "grad_norm": 9.360711097717285,
31
+ "learning_rate": 1.357142857142857e-05,
32
+ "loss": 3.1711,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.12727272727272726,
38
+ "eval_loss": 3.0051403045654297,
39
+ "eval_runtime": 2.0612,
40
+ "eval_samples_per_second": 106.735,
41
+ "eval_steps_per_second": 6.792,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 2.1454545454545455,
46
+ "grad_norm": 8.500679016113281,
47
+ "learning_rate": 2.0714285714285718e-05,
48
+ "loss": 2.983,
49
+ "step": 30
50
+ },
51
+ {
52
+ "epoch": 2.8727272727272726,
53
+ "grad_norm": 9.82868766784668,
54
+ "learning_rate": 2.785714285714286e-05,
55
+ "loss": 2.8729,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "eval_accuracy": 0.20909090909090908,
61
+ "eval_loss": 2.6284613609313965,
62
+ "eval_runtime": 2.0503,
63
+ "eval_samples_per_second": 107.302,
64
+ "eval_steps_per_second": 6.828,
65
+ "step": 42
66
+ },
67
+ {
68
+ "epoch": 3.581818181818182,
69
+ "grad_norm": 7.700014114379883,
70
+ "learning_rate": 3.5e-05,
71
+ "loss": 2.562,
72
+ "step": 50
73
+ },
74
+ {
75
+ "epoch": 4.0,
76
+ "eval_accuracy": 0.37727272727272726,
77
+ "eval_loss": 2.160020112991333,
78
+ "eval_runtime": 2.3625,
79
+ "eval_samples_per_second": 93.121,
80
+ "eval_steps_per_second": 5.926,
81
+ "step": 56
82
+ },
83
+ {
84
+ "epoch": 4.290909090909091,
85
+ "grad_norm": 7.527115821838379,
86
+ "learning_rate": 4.214285714285714e-05,
87
+ "loss": 2.2406,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "grad_norm": 5.579476833343506,
93
+ "learning_rate": 4.928571428571429e-05,
94
+ "loss": 1.8675,
95
+ "step": 70
96
+ },
97
+ {
98
+ "epoch": 5.0,
99
+ "eval_accuracy": 0.5363636363636364,
100
+ "eval_loss": 1.639161229133606,
101
+ "eval_runtime": 2.0879,
102
+ "eval_samples_per_second": 105.372,
103
+ "eval_steps_per_second": 6.705,
104
+ "step": 70
105
+ },
106
+ {
107
+ "epoch": 5.7272727272727275,
108
+ "grad_norm": 6.313197135925293,
109
+ "learning_rate": 4.928571428571429e-05,
110
+ "loss": 1.6359,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 6.0,
115
+ "eval_accuracy": 0.6681818181818182,
116
+ "eval_loss": 1.2266921997070312,
117
+ "eval_runtime": 2.1098,
118
+ "eval_samples_per_second": 104.274,
119
+ "eval_steps_per_second": 6.636,
120
+ "step": 84
121
+ },
122
+ {
123
+ "epoch": 6.4363636363636365,
124
+ "grad_norm": 5.8209004402160645,
125
+ "learning_rate": 4.84920634920635e-05,
126
+ "loss": 1.3499,
127
+ "step": 90
128
+ },
129
+ {
130
+ "epoch": 7.0,
131
+ "eval_accuracy": 0.6818181818181818,
132
+ "eval_loss": 1.0587564706802368,
133
+ "eval_runtime": 2.9116,
134
+ "eval_samples_per_second": 75.56,
135
+ "eval_steps_per_second": 4.808,
136
+ "step": 98
137
+ },
138
+ {
139
+ "epoch": 7.1454545454545455,
140
+ "grad_norm": 6.231616973876953,
141
+ "learning_rate": 4.7698412698412706e-05,
142
+ "loss": 1.2501,
143
+ "step": 100
144
+ },
145
+ {
146
+ "epoch": 7.872727272727273,
147
+ "grad_norm": 5.9844584465026855,
148
+ "learning_rate": 4.690476190476191e-05,
149
+ "loss": 1.076,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.6909090909090909,
155
+ "eval_loss": 0.8790816068649292,
156
+ "eval_runtime": 4.5841,
157
+ "eval_samples_per_second": 47.992,
158
+ "eval_steps_per_second": 3.054,
159
+ "step": 112
160
+ },
161
+ {
162
+ "epoch": 8.581818181818182,
163
+ "grad_norm": 5.29907751083374,
164
+ "learning_rate": 4.6111111111111115e-05,
165
+ "loss": 0.9811,
166
+ "step": 120
167
+ },
168
+ {
169
+ "epoch": 9.0,
170
+ "eval_accuracy": 0.7545454545454545,
171
+ "eval_loss": 0.7573429346084595,
172
+ "eval_runtime": 4.4185,
173
+ "eval_samples_per_second": 49.79,
174
+ "eval_steps_per_second": 3.168,
175
+ "step": 126
176
+ },
177
+ {
178
+ "epoch": 9.290909090909091,
179
+ "grad_norm": 5.92799711227417,
180
+ "learning_rate": 4.531746031746032e-05,
181
+ "loss": 0.8655,
182
+ "step": 130
183
+ },
184
+ {
185
+ "epoch": 10.0,
186
+ "grad_norm": 3.577127695083618,
187
+ "learning_rate": 4.4523809523809525e-05,
188
+ "loss": 0.7309,
189
+ "step": 140
190
+ },
191
+ {
192
+ "epoch": 10.0,
193
+ "eval_accuracy": 0.7818181818181819,
194
+ "eval_loss": 0.6194924712181091,
195
+ "eval_runtime": 4.0981,
196
+ "eval_samples_per_second": 53.683,
197
+ "eval_steps_per_second": 3.416,
198
+ "step": 140
199
+ },
200
+ {
201
+ "epoch": 10.727272727272727,
202
+ "grad_norm": 5.252236843109131,
203
+ "learning_rate": 4.373015873015873e-05,
204
+ "loss": 0.7776,
205
+ "step": 150
206
+ },
207
+ {
208
+ "epoch": 11.0,
209
+ "eval_accuracy": 0.8,
210
+ "eval_loss": 0.5426079034805298,
211
+ "eval_runtime": 4.0014,
212
+ "eval_samples_per_second": 54.981,
213
+ "eval_steps_per_second": 3.499,
214
+ "step": 154
215
+ },
216
+ {
217
+ "epoch": 11.436363636363636,
218
+ "grad_norm": 4.398848533630371,
219
+ "learning_rate": 4.2936507936507935e-05,
220
+ "loss": 0.7365,
221
+ "step": 160
222
+ },
223
+ {
224
+ "epoch": 12.0,
225
+ "eval_accuracy": 0.8772727272727273,
226
+ "eval_loss": 0.40285420417785645,
227
+ "eval_runtime": 4.3238,
228
+ "eval_samples_per_second": 50.882,
229
+ "eval_steps_per_second": 3.238,
230
+ "step": 168
231
+ }
232
+ ],
233
+ "logging_steps": 10,
234
+ "max_steps": 700,
235
+ "num_input_tokens_seen": 0,
236
+ "num_train_epochs": 50,
237
+ "save_steps": 500,
238
+ "stateful_callbacks": {
239
+ "TrainerControl": {
240
+ "args": {
241
+ "should_epoch_stop": false,
242
+ "should_evaluate": false,
243
+ "should_log": false,
244
+ "should_save": true,
245
+ "should_training_stop": false
246
+ },
247
+ "attributes": {}
248
+ }
249
+ },
250
+ "total_flos": 1.0670310968131584e+17,
251
+ "train_batch_size": 16,
252
+ "trial_name": null,
253
+ "trial_params": null
254
+ }
checkpoint-168/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-182/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }
checkpoint-182/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04783c62c1e951592bd47021cc8f40405c44c10b33b25eceb52fb8888d3e3562
3
+ size 44805376
checkpoint-182/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3b758606025c6d643604731a13ed47748090b30ceea5a9c49c397fbf18f32cd
3
+ size 89553018
checkpoint-182/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f53f0b8d547f2fb5901e4ba725558d6ccafdba1ecddebb259b1585640875d547
3
+ size 13990
checkpoint-182/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b6644b860068fa4350a2825c82e6f306f5a289d29410d5882eb740ab2cc559e
3
+ size 1064
checkpoint-182/trainer_state.json ADDED
@@ -0,0 +1,277 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 168,
3
+ "best_metric": 0.8772727272727273,
4
+ "best_model_checkpoint": "font-identifier/checkpoint-168",
5
+ "epoch": 13.0,
6
+ "eval_steps": 500,
7
+ "global_step": 182,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.7272727272727273,
14
+ "grad_norm": 9.430075645446777,
15
+ "learning_rate": 6.428571428571429e-06,
16
+ "loss": 3.3213,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.06818181818181818,
22
+ "eval_loss": 3.2453081607818604,
23
+ "eval_runtime": 2.4333,
24
+ "eval_samples_per_second": 90.411,
25
+ "eval_steps_per_second": 5.753,
26
+ "step": 14
27
+ },
28
+ {
29
+ "epoch": 1.4363636363636363,
30
+ "grad_norm": 9.360711097717285,
31
+ "learning_rate": 1.357142857142857e-05,
32
+ "loss": 3.1711,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.12727272727272726,
38
+ "eval_loss": 3.0051403045654297,
39
+ "eval_runtime": 2.0612,
40
+ "eval_samples_per_second": 106.735,
41
+ "eval_steps_per_second": 6.792,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 2.1454545454545455,
46
+ "grad_norm": 8.500679016113281,
47
+ "learning_rate": 2.0714285714285718e-05,
48
+ "loss": 2.983,
49
+ "step": 30
50
+ },
51
+ {
52
+ "epoch": 2.8727272727272726,
53
+ "grad_norm": 9.82868766784668,
54
+ "learning_rate": 2.785714285714286e-05,
55
+ "loss": 2.8729,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "eval_accuracy": 0.20909090909090908,
61
+ "eval_loss": 2.6284613609313965,
62
+ "eval_runtime": 2.0503,
63
+ "eval_samples_per_second": 107.302,
64
+ "eval_steps_per_second": 6.828,
65
+ "step": 42
66
+ },
67
+ {
68
+ "epoch": 3.581818181818182,
69
+ "grad_norm": 7.700014114379883,
70
+ "learning_rate": 3.5e-05,
71
+ "loss": 2.562,
72
+ "step": 50
73
+ },
74
+ {
75
+ "epoch": 4.0,
76
+ "eval_accuracy": 0.37727272727272726,
77
+ "eval_loss": 2.160020112991333,
78
+ "eval_runtime": 2.3625,
79
+ "eval_samples_per_second": 93.121,
80
+ "eval_steps_per_second": 5.926,
81
+ "step": 56
82
+ },
83
+ {
84
+ "epoch": 4.290909090909091,
85
+ "grad_norm": 7.527115821838379,
86
+ "learning_rate": 4.214285714285714e-05,
87
+ "loss": 2.2406,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "grad_norm": 5.579476833343506,
93
+ "learning_rate": 4.928571428571429e-05,
94
+ "loss": 1.8675,
95
+ "step": 70
96
+ },
97
+ {
98
+ "epoch": 5.0,
99
+ "eval_accuracy": 0.5363636363636364,
100
+ "eval_loss": 1.639161229133606,
101
+ "eval_runtime": 2.0879,
102
+ "eval_samples_per_second": 105.372,
103
+ "eval_steps_per_second": 6.705,
104
+ "step": 70
105
+ },
106
+ {
107
+ "epoch": 5.7272727272727275,
108
+ "grad_norm": 6.313197135925293,
109
+ "learning_rate": 4.928571428571429e-05,
110
+ "loss": 1.6359,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 6.0,
115
+ "eval_accuracy": 0.6681818181818182,
116
+ "eval_loss": 1.2266921997070312,
117
+ "eval_runtime": 2.1098,
118
+ "eval_samples_per_second": 104.274,
119
+ "eval_steps_per_second": 6.636,
120
+ "step": 84
121
+ },
122
+ {
123
+ "epoch": 6.4363636363636365,
124
+ "grad_norm": 5.8209004402160645,
125
+ "learning_rate": 4.84920634920635e-05,
126
+ "loss": 1.3499,
127
+ "step": 90
128
+ },
129
+ {
130
+ "epoch": 7.0,
131
+ "eval_accuracy": 0.6818181818181818,
132
+ "eval_loss": 1.0587564706802368,
133
+ "eval_runtime": 2.9116,
134
+ "eval_samples_per_second": 75.56,
135
+ "eval_steps_per_second": 4.808,
136
+ "step": 98
137
+ },
138
+ {
139
+ "epoch": 7.1454545454545455,
140
+ "grad_norm": 6.231616973876953,
141
+ "learning_rate": 4.7698412698412706e-05,
142
+ "loss": 1.2501,
143
+ "step": 100
144
+ },
145
+ {
146
+ "epoch": 7.872727272727273,
147
+ "grad_norm": 5.9844584465026855,
148
+ "learning_rate": 4.690476190476191e-05,
149
+ "loss": 1.076,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.6909090909090909,
155
+ "eval_loss": 0.8790816068649292,
156
+ "eval_runtime": 4.5841,
157
+ "eval_samples_per_second": 47.992,
158
+ "eval_steps_per_second": 3.054,
159
+ "step": 112
160
+ },
161
+ {
162
+ "epoch": 8.581818181818182,
163
+ "grad_norm": 5.29907751083374,
164
+ "learning_rate": 4.6111111111111115e-05,
165
+ "loss": 0.9811,
166
+ "step": 120
167
+ },
168
+ {
169
+ "epoch": 9.0,
170
+ "eval_accuracy": 0.7545454545454545,
171
+ "eval_loss": 0.7573429346084595,
172
+ "eval_runtime": 4.4185,
173
+ "eval_samples_per_second": 49.79,
174
+ "eval_steps_per_second": 3.168,
175
+ "step": 126
176
+ },
177
+ {
178
+ "epoch": 9.290909090909091,
179
+ "grad_norm": 5.92799711227417,
180
+ "learning_rate": 4.531746031746032e-05,
181
+ "loss": 0.8655,
182
+ "step": 130
183
+ },
184
+ {
185
+ "epoch": 10.0,
186
+ "grad_norm": 3.577127695083618,
187
+ "learning_rate": 4.4523809523809525e-05,
188
+ "loss": 0.7309,
189
+ "step": 140
190
+ },
191
+ {
192
+ "epoch": 10.0,
193
+ "eval_accuracy": 0.7818181818181819,
194
+ "eval_loss": 0.6194924712181091,
195
+ "eval_runtime": 4.0981,
196
+ "eval_samples_per_second": 53.683,
197
+ "eval_steps_per_second": 3.416,
198
+ "step": 140
199
+ },
200
+ {
201
+ "epoch": 10.727272727272727,
202
+ "grad_norm": 5.252236843109131,
203
+ "learning_rate": 4.373015873015873e-05,
204
+ "loss": 0.7776,
205
+ "step": 150
206
+ },
207
+ {
208
+ "epoch": 11.0,
209
+ "eval_accuracy": 0.8,
210
+ "eval_loss": 0.5426079034805298,
211
+ "eval_runtime": 4.0014,
212
+ "eval_samples_per_second": 54.981,
213
+ "eval_steps_per_second": 3.499,
214
+ "step": 154
215
+ },
216
+ {
217
+ "epoch": 11.436363636363636,
218
+ "grad_norm": 4.398848533630371,
219
+ "learning_rate": 4.2936507936507935e-05,
220
+ "loss": 0.7365,
221
+ "step": 160
222
+ },
223
+ {
224
+ "epoch": 12.0,
225
+ "eval_accuracy": 0.8772727272727273,
226
+ "eval_loss": 0.40285420417785645,
227
+ "eval_runtime": 4.3238,
228
+ "eval_samples_per_second": 50.882,
229
+ "eval_steps_per_second": 3.238,
230
+ "step": 168
231
+ },
232
+ {
233
+ "epoch": 12.145454545454545,
234
+ "grad_norm": 5.8070549964904785,
235
+ "learning_rate": 4.214285714285714e-05,
236
+ "loss": 0.6116,
237
+ "step": 170
238
+ },
239
+ {
240
+ "epoch": 12.872727272727273,
241
+ "grad_norm": 6.383784770965576,
242
+ "learning_rate": 4.134920634920635e-05,
243
+ "loss": 0.5767,
244
+ "step": 180
245
+ },
246
+ {
247
+ "epoch": 13.0,
248
+ "eval_accuracy": 0.8363636363636363,
249
+ "eval_loss": 0.4417766332626343,
250
+ "eval_runtime": 4.8355,
251
+ "eval_samples_per_second": 45.497,
252
+ "eval_steps_per_second": 2.895,
253
+ "step": 182
254
+ }
255
+ ],
256
+ "logging_steps": 10,
257
+ "max_steps": 700,
258
+ "num_input_tokens_seen": 0,
259
+ "num_train_epochs": 50,
260
+ "save_steps": 500,
261
+ "stateful_callbacks": {
262
+ "TrainerControl": {
263
+ "args": {
264
+ "should_epoch_stop": false,
265
+ "should_evaluate": false,
266
+ "should_log": false,
267
+ "should_save": true,
268
+ "should_training_stop": false
269
+ },
270
+ "attributes": {}
271
+ }
272
+ },
273
+ "total_flos": 1.1559503548809216e+17,
274
+ "train_batch_size": 16,
275
+ "trial_name": null,
276
+ "trial_params": null
277
+ }
checkpoint-182/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe902249061f5c6116674b4b3dc02d72fa3794a0c9100f91b35a0c0b09d98f6c
3
+ size 5304
checkpoint-196/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ResNetForImageClassification"
4
+ ],
5
+ "depths": [
6
+ 2,
7
+ 2,
8
+ 2,
9
+ 2
10
+ ],
11
+ "downsample_in_bottleneck": false,
12
+ "downsample_in_first_stage": false,
13
+ "embedding_size": 64,
14
+ "hidden_act": "relu",
15
+ "hidden_sizes": [
16
+ 64,
17
+ 128,
18
+ 256,
19
+ 512
20
+ ],
21
+ "id2label": {
22
+ "0": "Arial",
23
+ "1": "Arial Black",
24
+ "10": "Times New Roman",
25
+ "11": "Times New Roman Bold",
26
+ "12": "Times New Roman Bold Italic",
27
+ "13": "Times New Roman Italic",
28
+ "14": "Trebuchet MS",
29
+ "15": "Trebuchet MS Bold",
30
+ "16": "Trebuchet MS Bold Italic",
31
+ "17": "Trebuchet MS Italic",
32
+ "18": "Verdana",
33
+ "19": "Verdana Bold",
34
+ "2": "Arial Bold",
35
+ "20": "Verdana Bold Italic",
36
+ "21": "Verdana Italic",
37
+ "3": "Arial Bold Italic",
38
+ "4": "Avenir",
39
+ "5": "Courier",
40
+ "6": "Georgia",
41
+ "7": "Helvetica",
42
+ "8": "Tahoma",
43
+ "9": "Tahoma Bold"
44
+ },
45
+ "label2id": {
46
+ "Arial": "0",
47
+ "Arial Black": "1",
48
+ "Arial Bold": "2",
49
+ "Arial Bold Italic": "3",
50
+ "Avenir": "4",
51
+ "Courier": "5",
52
+ "Georgia": "6",
53
+ "Helvetica": "7",
54
+ "Tahoma": "8",
55
+ "Tahoma Bold": "9",
56
+ "Times New Roman": "10",
57
+ "Times New Roman Bold": "11",
58
+ "Times New Roman Bold Italic": "12",
59
+ "Times New Roman Italic": "13",
60
+ "Trebuchet MS": "14",
61
+ "Trebuchet MS Bold": "15",
62
+ "Trebuchet MS Bold Italic": "16",
63
+ "Trebuchet MS Italic": "17",
64
+ "Verdana": "18",
65
+ "Verdana Bold": "19",
66
+ "Verdana Bold Italic": "20",
67
+ "Verdana Italic": "21"
68
+ },
69
+ "layer_type": "basic",
70
+ "model_type": "resnet",
71
+ "num_channels": 3,
72
+ "out_features": [
73
+ "stage4"
74
+ ],
75
+ "out_indices": [
76
+ 4
77
+ ],
78
+ "problem_type": "single_label_classification",
79
+ "stage_names": [
80
+ "stem",
81
+ "stage1",
82
+ "stage2",
83
+ "stage3",
84
+ "stage4"
85
+ ],
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.3"
88
+ }