Hugging Face's logo

djramirezp
/
vit-face-classification-quiz2

Image Classification
Transformers
Safetensors
Spanish
vit
vision-transformer
binary-classification
deepfake-detection
Model card Files
xet
 Community
djramirezp commited on
Commit
3ac1355
·
verified ·
1 Parent(s): 2107cb5

Publicacion del mejor modelo (Parte 9.2)

Browse files
Files changed (16) hide show
  1. README.md +10 -10
  2. checkpoint-1458/config.json +33 -0
  3. checkpoint-1458/model.safetensors +3 -0
  4. checkpoint-1458/optimizer.pt +3 -0
  5. checkpoint-1458/rng_state.pth +3 -0
  6. checkpoint-1458/scheduler.pt +3 -0
  7. checkpoint-1458/trainer_state.json +157 -0
  8. checkpoint-1458/training_args.bin +3 -0
  9. checkpoint-1944/config.json +33 -0
  10. checkpoint-1944/model.safetensors +3 -0
  11. checkpoint-1944/optimizer.pt +3 -0
  12. checkpoint-1944/rng_state.pth +3 -0
  13. checkpoint-1944/scheduler.pt +3 -0
  14. checkpoint-1944/trainer_state.json +195 -0
  15. checkpoint-1944/training_args.bin +3 -0
  16. model.safetensors +1 -1
README.md CHANGED
@@ -36,15 +36,15 @@ La seleccion del mejor checkpoint se hizo con base en la metrica F1 de validacio
36
  ## Resultados
37
 
38
  ### Validacion
39
- - loss: 0.014208527281880379
40
- - accuracy: 0.9963436928702011
41
- - precision: 0.9963706766866571
42
- - recall: 0.9963436928702011
43
- - f1: 0.996343863976966
44
 
45
  ### Test
46
- - loss: 0.03636797145009041
47
- - accuracy: 0.9920071047957372
48
- - precision: 0.9920799653809312
49
- - recall: 0.9920071047957372
50
- - f1: 0.9920051158081001
 
36
  ## Resultados
37
 
38
  ### Validacion
39
+ - loss: 0.013790015131235123
40
+ - accuracy: 0.9981718464351006
41
+ - precision: 0.9981786173742297
42
+ - recall: 0.9981718464351006
43
+ - f1: 0.998171895325199
44
 
45
  ### Test
46
+ - loss: 0.02344433404505253
47
+ - accuracy: 0.9928952042628775
48
+ - precision: 0.9929184199081565
49
+ - recall: 0.9928952042628775
50
+ - f1: 0.9928942615297829
checkpoint-1458/config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ViTForImageClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.0,
6
+ "dtype": "float32",
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "FAKE",
13
+ "1": "REAL"
14
+ },
15
+ "image_size": 224,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "label2id": {
19
+ "FAKE": 0,
20
+ "REAL": 1
21
+ },
22
+ "layer_norm_eps": 1e-12,
23
+ "model_type": "vit",
24
+ "num_attention_heads": 12,
25
+ "num_channels": 3,
26
+ "num_hidden_layers": 12,
27
+ "patch_size": 16,
28
+ "pooler_act": "tanh",
29
+ "pooler_output_size": 768,
30
+ "problem_type": "single_label_classification",
31
+ "qkv_bias": true,
32
+ "transformers_version": "4.57.6"
33
+ }
checkpoint-1458/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:beebd5c23873379020d1d77c08eed45cbf341d3ff81518acd3023d137e749cec
3
+ size 343223968
checkpoint-1458/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ac5b7f0da38c65296236d98bffd507ca0d41487b39154f50582b8e0ece15634
3
+ size 686563322
checkpoint-1458/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04480b41443f43abe08b007d9f81b1cde1f8f5a62079008c358d77853cadc4d8
3
+ size 13990
checkpoint-1458/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af521141aecce918e88c427f624bce7d96ec3ab9e09628e2a4e7bacd6eefc9fb
3
+ size 1064
checkpoint-1458/trainer_state.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1458,
3
+ "best_metric": 0.998171895325199,
4
+ "best_model_checkpoint": "vit-finetuned-face-classification/checkpoint-1458",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1458,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 0.5751039385795593,
15
+ "learning_rate": 1.9462550028587764e-05,
16
+ "loss": 0.2933,
17
+ "step": 243
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.9872029250457038,
22
+ "eval_f1": 0.9872035238293116,
23
+ "eval_loss": 0.07286947220563889,
24
+ "eval_precision": 0.9875274885409214,
25
+ "eval_recall": 0.9872029250457038,
26
+ "eval_runtime": 6.2204,
27
+ "eval_samples_per_second": 87.937,
28
+ "eval_steps_per_second": 5.627,
29
+ "step": 243
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "grad_norm": 0.10031592100858688,
34
+ "learning_rate": 1.668381932532876e-05,
35
+ "loss": 0.0499,
36
+ "step": 486
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_accuracy": 0.9908592321755028,
41
+ "eval_f1": 0.9908598432036901,
42
+ "eval_loss": 0.030813397839665413,
43
+ "eval_precision": 0.9910260345080665,
44
+ "eval_recall": 0.9908592321755028,
45
+ "eval_runtime": 6.5614,
46
+ "eval_samples_per_second": 83.367,
47
+ "eval_steps_per_second": 5.334,
48
+ "step": 486
49
+ },
50
+ {
51
+ "epoch": 3.0,
52
+ "grad_norm": 0.07250971347093582,
53
+ "learning_rate": 1.3905088622069755e-05,
54
+ "loss": 0.0214,
55
+ "step": 729
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_accuracy": 0.9926873857404022,
60
+ "eval_f1": 0.9926878745776567,
61
+ "eval_loss": 0.031054198741912842,
62
+ "eval_precision": 0.9927945302716784,
63
+ "eval_recall": 0.9926873857404022,
64
+ "eval_runtime": 6.9981,
65
+ "eval_samples_per_second": 78.164,
66
+ "eval_steps_per_second": 5.001,
67
+ "step": 729
68
+ },
69
+ {
70
+ "epoch": 4.0,
71
+ "grad_norm": 0.04455922171473503,
72
+ "learning_rate": 1.1126357918810751e-05,
73
+ "loss": 0.0133,
74
+ "step": 972
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_accuracy": 0.9926873857404022,
79
+ "eval_f1": 0.9926878745776567,
80
+ "eval_loss": 0.032945409417152405,
81
+ "eval_precision": 0.9927945302716784,
82
+ "eval_recall": 0.9926873857404022,
83
+ "eval_runtime": 7.5929,
84
+ "eval_samples_per_second": 72.041,
85
+ "eval_steps_per_second": 4.61,
86
+ "step": 972
87
+ },
88
+ {
89
+ "epoch": 5.0,
90
+ "grad_norm": 0.03358669579029083,
91
+ "learning_rate": 8.347627215551745e-06,
92
+ "loss": 0.0065,
93
+ "step": 1215
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_accuracy": 0.9963436928702011,
98
+ "eval_f1": 0.996343863976966,
99
+ "eval_loss": 0.018969673663377762,
100
+ "eval_precision": 0.9963706766866571,
101
+ "eval_recall": 0.9963436928702011,
102
+ "eval_runtime": 7.8628,
103
+ "eval_samples_per_second": 69.568,
104
+ "eval_steps_per_second": 4.451,
105
+ "step": 1215
106
+ },
107
+ {
108
+ "epoch": 6.0,
109
+ "grad_norm": 0.04800612851977348,
110
+ "learning_rate": 5.56889651229274e-06,
111
+ "loss": 0.0046,
112
+ "step": 1458
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_accuracy": 0.9981718464351006,
117
+ "eval_f1": 0.998171895325199,
118
+ "eval_loss": 0.013790015131235123,
119
+ "eval_precision": 0.9981786173742297,
120
+ "eval_recall": 0.9981718464351006,
121
+ "eval_runtime": 7.6607,
122
+ "eval_samples_per_second": 71.403,
123
+ "eval_steps_per_second": 4.569,
124
+ "step": 1458
125
+ }
126
+ ],
127
+ "logging_steps": 500,
128
+ "max_steps": 1944,
129
+ "num_input_tokens_seen": 0,
130
+ "num_train_epochs": 8,
131
+ "save_steps": 500,
132
+ "stateful_callbacks": {
133
+ "EarlyStoppingCallback": {
134
+ "args": {
135
+ "early_stopping_patience": 2,
136
+ "early_stopping_threshold": 0.001
137
+ },
138
+ "attributes": {
139
+ "early_stopping_patience_counter": 0
140
+ }
141
+ },
142
+ "TrainerControl": {
143
+ "args": {
144
+ "should_epoch_stop": false,
145
+ "should_evaluate": false,
146
+ "should_log": false,
147
+ "should_save": true,
148
+ "should_training_stop": false
149
+ },
150
+ "attributes": {}
151
+ }
152
+ },
153
+ "total_flos": 1.805873325978452e+18,
154
+ "train_batch_size": 16,
155
+ "trial_name": null,
156
+ "trial_params": null
157
+ }
checkpoint-1458/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:996910b20e8895000999ca655730e24555f2998d3c6e626c04f395fdf89b53f1
3
+ size 5432
checkpoint-1944/config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ViTForImageClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.0,
6
+ "dtype": "float32",
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "FAKE",
13
+ "1": "REAL"
14
+ },
15
+ "image_size": 224,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "label2id": {
19
+ "FAKE": 0,
20
+ "REAL": 1
21
+ },
22
+ "layer_norm_eps": 1e-12,
23
+ "model_type": "vit",
24
+ "num_attention_heads": 12,
25
+ "num_channels": 3,
26
+ "num_hidden_layers": 12,
27
+ "patch_size": 16,
28
+ "pooler_act": "tanh",
29
+ "pooler_output_size": 768,
30
+ "problem_type": "single_label_classification",
31
+ "qkv_bias": true,
32
+ "transformers_version": "4.57.6"
33
+ }
checkpoint-1944/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e1441a6c921bbc231de6514f8654ec93a05f3fe6128f183e9567fcd53c36700
3
+ size 343223968
checkpoint-1944/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6d550722a56d3afa29fbea530d66f6f399b512717382b2f66b575b69f64b4a5
3
+ size 686563322
checkpoint-1944/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:825a4ce620b0097568150a1e62af629205132971f8a774ff41ee85b48b76817e
3
+ size 13990
checkpoint-1944/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16f6d6265ac68d00cc25ae49ae289ed5d2554ffdeaae60f22349b03a85f3d557
3
+ size 1064
checkpoint-1944/trainer_state.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1458,
3
+ "best_metric": 0.998171895325199,
4
+ "best_model_checkpoint": "vit-finetuned-face-classification/checkpoint-1458",
5
+ "epoch": 8.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1944,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 0.5751039385795593,
15
+ "learning_rate": 1.9462550028587764e-05,
16
+ "loss": 0.2933,
17
+ "step": 243
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.9872029250457038,
22
+ "eval_f1": 0.9872035238293116,
23
+ "eval_loss": 0.07286947220563889,
24
+ "eval_precision": 0.9875274885409214,
25
+ "eval_recall": 0.9872029250457038,
26
+ "eval_runtime": 6.2204,
27
+ "eval_samples_per_second": 87.937,
28
+ "eval_steps_per_second": 5.627,
29
+ "step": 243
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "grad_norm": 0.10031592100858688,
34
+ "learning_rate": 1.668381932532876e-05,
35
+ "loss": 0.0499,
36
+ "step": 486
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_accuracy": 0.9908592321755028,
41
+ "eval_f1": 0.9908598432036901,
42
+ "eval_loss": 0.030813397839665413,
43
+ "eval_precision": 0.9910260345080665,
44
+ "eval_recall": 0.9908592321755028,
45
+ "eval_runtime": 6.5614,
46
+ "eval_samples_per_second": 83.367,
47
+ "eval_steps_per_second": 5.334,
48
+ "step": 486
49
+ },
50
+ {
51
+ "epoch": 3.0,
52
+ "grad_norm": 0.07250971347093582,
53
+ "learning_rate": 1.3905088622069755e-05,
54
+ "loss": 0.0214,
55
+ "step": 729
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_accuracy": 0.9926873857404022,
60
+ "eval_f1": 0.9926878745776567,
61
+ "eval_loss": 0.031054198741912842,
62
+ "eval_precision": 0.9927945302716784,
63
+ "eval_recall": 0.9926873857404022,
64
+ "eval_runtime": 6.9981,
65
+ "eval_samples_per_second": 78.164,
66
+ "eval_steps_per_second": 5.001,
67
+ "step": 729
68
+ },
69
+ {
70
+ "epoch": 4.0,
71
+ "grad_norm": 0.04455922171473503,
72
+ "learning_rate": 1.1126357918810751e-05,
73
+ "loss": 0.0133,
74
+ "step": 972
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_accuracy": 0.9926873857404022,
79
+ "eval_f1": 0.9926878745776567,
80
+ "eval_loss": 0.032945409417152405,
81
+ "eval_precision": 0.9927945302716784,
82
+ "eval_recall": 0.9926873857404022,
83
+ "eval_runtime": 7.5929,
84
+ "eval_samples_per_second": 72.041,
85
+ "eval_steps_per_second": 4.61,
86
+ "step": 972
87
+ },
88
+ {
89
+ "epoch": 5.0,
90
+ "grad_norm": 0.03358669579029083,
91
+ "learning_rate": 8.347627215551745e-06,
92
+ "loss": 0.0065,
93
+ "step": 1215
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_accuracy": 0.9963436928702011,
98
+ "eval_f1": 0.996343863976966,
99
+ "eval_loss": 0.018969673663377762,
100
+ "eval_precision": 0.9963706766866571,
101
+ "eval_recall": 0.9963436928702011,
102
+ "eval_runtime": 7.8628,
103
+ "eval_samples_per_second": 69.568,
104
+ "eval_steps_per_second": 4.451,
105
+ "step": 1215
106
+ },
107
+ {
108
+ "epoch": 6.0,
109
+ "grad_norm": 0.04800612851977348,
110
+ "learning_rate": 5.56889651229274e-06,
111
+ "loss": 0.0046,
112
+ "step": 1458
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_accuracy": 0.9981718464351006,
117
+ "eval_f1": 0.998171895325199,
118
+ "eval_loss": 0.013790015131235123,
119
+ "eval_precision": 0.9981786173742297,
120
+ "eval_recall": 0.9981718464351006,
121
+ "eval_runtime": 7.6607,
122
+ "eval_samples_per_second": 71.403,
123
+ "eval_steps_per_second": 4.569,
124
+ "step": 1458
125
+ },
126
+ {
127
+ "epoch": 7.0,
128
+ "grad_norm": 0.026266049593687057,
129
+ "learning_rate": 2.7901658090337335e-06,
130
+ "loss": 0.0035,
131
+ "step": 1701
132
+ },
133
+ {
134
+ "epoch": 7.0,
135
+ "eval_accuracy": 0.9981718464351006,
136
+ "eval_f1": 0.998171895325199,
137
+ "eval_loss": 0.014415347948670387,
138
+ "eval_precision": 0.9981786173742297,
139
+ "eval_recall": 0.9981718464351006,
140
+ "eval_runtime": 8.3197,
141
+ "eval_samples_per_second": 65.748,
142
+ "eval_steps_per_second": 4.207,
143
+ "step": 1701
144
+ },
145
+ {
146
+ "epoch": 8.0,
147
+ "grad_norm": 0.027159975841641426,
148
+ "learning_rate": 1.1435105774728416e-08,
149
+ "loss": 0.0032,
150
+ "step": 1944
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.9981718464351006,
155
+ "eval_f1": 0.998171895325199,
156
+ "eval_loss": 0.01470749918371439,
157
+ "eval_precision": 0.9981786173742297,
158
+ "eval_recall": 0.9981718464351006,
159
+ "eval_runtime": 8.3236,
160
+ "eval_samples_per_second": 65.717,
161
+ "eval_steps_per_second": 4.205,
162
+ "step": 1944
163
+ }
164
+ ],
165
+ "logging_steps": 500,
166
+ "max_steps": 1944,
167
+ "num_input_tokens_seen": 0,
168
+ "num_train_epochs": 8,
169
+ "save_steps": 500,
170
+ "stateful_callbacks": {
171
+ "EarlyStoppingCallback": {
172
+ "args": {
173
+ "early_stopping_patience": 2,
174
+ "early_stopping_threshold": 0.001
175
+ },
176
+ "attributes": {
177
+ "early_stopping_patience_counter": 2
178
+ }
179
+ },
180
+ "TrainerControl": {
181
+ "args": {
182
+ "should_epoch_stop": false,
183
+ "should_evaluate": false,
184
+ "should_log": false,
185
+ "should_save": true,
186
+ "should_training_stop": true
187
+ },
188
+ "attributes": {}
189
+ }
190
+ },
191
+ "total_flos": 2.4078311013046026e+18,
192
+ "train_batch_size": 16,
193
+ "trial_name": null,
194
+ "trial_params": null
195
+ }
checkpoint-1944/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:996910b20e8895000999ca655730e24555f2998d3c6e626c04f395fdf89b53f1
3
+ size 5432
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5016fc6ba2598ee6c01c32dadd8a5ca9cd5796421ce14eb80bf7072cacc56f5e
3
  size 343223968
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:beebd5c23873379020d1d77c08eed45cbf341d3ff81518acd3023d137e749cec
3
  size 343223968