Hugging Face's logo

prithivMLmods
/
siglip2-x256p32-explicit-content

Image Classification
Transformers
Safetensors
English
siglip
siglip2
384
explicit-content
adult-content
classification
Model card Files
xet
Community
prithivMLmods commited on
Commit
7ba9785
·
verified ·
1 Parent(s): 55ce3c6

Upload folder using huggingface_hub

Browse files
Files changed (20) hide show
  1. checkpoint-1656/config.json +52 -0
  2. checkpoint-1656/model.safetensors +3 -0
  3. checkpoint-1656/optimizer.pt +3 -0
  4. checkpoint-1656/preprocessor_config.json +24 -0
  5. checkpoint-1656/rng_state.pth +3 -0
  6. checkpoint-1656/scheduler.pt +3 -0
  7. checkpoint-1656/trainer_state.json +75 -0
  8. checkpoint-1656/training_args.bin +3 -0
  9. checkpoint-828/config.json +52 -0
  10. checkpoint-828/model.safetensors +3 -0
  11. checkpoint-828/optimizer.pt +3 -0
  12. checkpoint-828/preprocessor_config.json +24 -0
  13. checkpoint-828/rng_state.pth +3 -0
  14. checkpoint-828/scheduler.pt +3 -0
  15. checkpoint-828/trainer_state.json +51 -0
  16. checkpoint-828/training_args.bin +3 -0
  17. config.json +52 -0
  18. model.safetensors +3 -0
  19. preprocessor_config.json +24 -0
  20. training_args.bin +3 -0
checkpoint-1656/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Anime Picture",
7
+ "1": "Hentai Picture",
8
+ "2": "Normal or Safe",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime Picture": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai Picture": 1,
17
+ "Normal or Safe": 2,
18
+ "Pornography": 3
19
+ },
20
+ "model_type": "siglip",
21
+ "problem_type": "single_label_classification",
22
+ "text_config": {
23
+ "attention_dropout": 0.0,
24
+ "hidden_act": "gelu_pytorch_tanh",
25
+ "hidden_size": 768,
26
+ "intermediate_size": 3072,
27
+ "layer_norm_eps": 1e-06,
28
+ "max_position_embeddings": 64,
29
+ "model_type": "siglip_text_model",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "projection_size": 768,
33
+ "torch_dtype": "float32",
34
+ "vocab_size": 256000
35
+ },
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.50.0",
38
+ "vision_config": {
39
+ "attention_dropout": 0.0,
40
+ "hidden_act": "gelu_pytorch_tanh",
41
+ "hidden_size": 768,
42
+ "image_size": 256,
43
+ "intermediate_size": 3072,
44
+ "layer_norm_eps": 1e-06,
45
+ "model_type": "siglip_vision_model",
46
+ "num_attention_heads": 12,
47
+ "num_channels": 3,
48
+ "num_hidden_layers": 12,
49
+ "patch_size": 32,
50
+ "torch_dtype": "float32"
51
+ }
52
+ }
checkpoint-1656/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26fdf4b6b7d682bb4bad93cef5112ed0e59b003cecd8cfbe1859ee04b18bca06
3
+ size 378249612
checkpoint-1656/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7045310aa000f0d07aec0edffe499d09debcfb90911c81d2a8eb823e3e463fe5
3
+ size 699918970
checkpoint-1656/preprocessor_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "SiglipImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "processor_class": "SiglipProcessor",
18
+ "resample": 2,
19
+ "rescale_factor": 0.00392156862745098,
20
+ "size": {
21
+ "height": 256,
22
+ "width": 256
23
+ }
24
+ }
checkpoint-1656/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66b395f57c7ddce57e9ef1719606f78ac267efd46ac9e0cacd7f473229ff44b2
3
+ size 14244
checkpoint-1656/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:348fd74b7d8c6958010fc0224d4cd95363f5a0f9aef9317146ccb791a1b2fade
3
+ size 1064
checkpoint-1656/trainer_state.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1656,
3
+ "best_metric": 0.17462851107120514,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1656",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1656,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.6038647342995169,
14
+ "grad_norm": 43.964569091796875,
15
+ "learning_rate": 1.4396014943960149e-06,
16
+ "loss": 0.5559,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.927651701091266,
22
+ "eval_loss": 0.2081761658191681,
23
+ "eval_model_preparation_time": 0.0027,
24
+ "eval_runtime": 405.243,
25
+ "eval_samples_per_second": 65.351,
26
+ "eval_steps_per_second": 8.17,
27
+ "step": 828
28
+ },
29
+ {
30
+ "epoch": 1.2077294685990339,
31
+ "grad_norm": 9.05398941040039,
32
+ "learning_rate": 8.169364881693648e-07,
33
+ "loss": 0.2458,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 1.8115942028985508,
38
+ "grad_norm": 24.472911834716797,
39
+ "learning_rate": 1.942714819427148e-07,
40
+ "loss": 0.2142,
41
+ "step": 1500
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.9408677264660348,
46
+ "eval_loss": 0.17462851107120514,
47
+ "eval_model_preparation_time": 0.0027,
48
+ "eval_runtime": 399.5276,
49
+ "eval_samples_per_second": 66.286,
50
+ "eval_steps_per_second": 8.287,
51
+ "step": 1656
52
+ }
53
+ ],
54
+ "logging_steps": 500,
55
+ "max_steps": 1656,
56
+ "num_input_tokens_seen": 0,
57
+ "num_train_epochs": 2,
58
+ "save_steps": 500,
59
+ "stateful_callbacks": {
60
+ "TrainerControl": {
61
+ "args": {
62
+ "should_epoch_stop": false,
63
+ "should_evaluate": false,
64
+ "should_log": false,
65
+ "should_save": true,
66
+ "should_training_stop": true
67
+ },
68
+ "attributes": {}
69
+ }
70
+ },
71
+ "total_flos": 5.904914979883844e+18,
72
+ "train_batch_size": 32,
73
+ "trial_name": null,
74
+ "trial_params": null
75
+ }
checkpoint-1656/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b82b36bb262e454f8e5c5fe3efc83cadeb46f932ed1ecf2e64afcef391dc64f0
3
+ size 5304
checkpoint-828/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Anime Picture",
7
+ "1": "Hentai Picture",
8
+ "2": "Normal or Safe",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime Picture": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai Picture": 1,
17
+ "Normal or Safe": 2,
18
+ "Pornography": 3
19
+ },
20
+ "model_type": "siglip",
21
+ "problem_type": "single_label_classification",
22
+ "text_config": {
23
+ "attention_dropout": 0.0,
24
+ "hidden_act": "gelu_pytorch_tanh",
25
+ "hidden_size": 768,
26
+ "intermediate_size": 3072,
27
+ "layer_norm_eps": 1e-06,
28
+ "max_position_embeddings": 64,
29
+ "model_type": "siglip_text_model",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "projection_size": 768,
33
+ "torch_dtype": "float32",
34
+ "vocab_size": 256000
35
+ },
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.50.0",
38
+ "vision_config": {
39
+ "attention_dropout": 0.0,
40
+ "hidden_act": "gelu_pytorch_tanh",
41
+ "hidden_size": 768,
42
+ "image_size": 256,
43
+ "intermediate_size": 3072,
44
+ "layer_norm_eps": 1e-06,
45
+ "model_type": "siglip_vision_model",
46
+ "num_attention_heads": 12,
47
+ "num_channels": 3,
48
+ "num_hidden_layers": 12,
49
+ "patch_size": 32,
50
+ "torch_dtype": "float32"
51
+ }
52
+ }
checkpoint-828/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d85a4b1acfb94bf87af4075b6da7e7e6ccd97954f0cb61c891a34caf0a39c84b
3
+ size 378249612
checkpoint-828/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4b2a1282f0ae11368d3a0a7b6f29b4418ec3fae228a789c7625a91762e444a0
3
+ size 699918970
checkpoint-828/preprocessor_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "SiglipImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "processor_class": "SiglipProcessor",
18
+ "resample": 2,
19
+ "rescale_factor": 0.00392156862745098,
20
+ "size": {
21
+ "height": 256,
22
+ "width": 256
23
+ }
24
+ }
checkpoint-828/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ef17702a0725d2ce6b58c8dbb73765893aabc8ef91b4de533167fc5ea45b4b6
3
+ size 14244
checkpoint-828/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:258bf86f5a91e7ff8fc05994dd1f4509153fccdb55be793bf481c807217ca37b
3
+ size 1064
checkpoint-828/trainer_state.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 828,
3
+ "best_metric": 0.2081761658191681,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-828",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 828,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.6038647342995169,
14
+ "grad_norm": 43.964569091796875,
15
+ "learning_rate": 1.4396014943960149e-06,
16
+ "loss": 0.5559,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.927651701091266,
22
+ "eval_loss": 0.2081761658191681,
23
+ "eval_model_preparation_time": 0.0027,
24
+ "eval_runtime": 405.243,
25
+ "eval_samples_per_second": 65.351,
26
+ "eval_steps_per_second": 8.17,
27
+ "step": 828
28
+ }
29
+ ],
30
+ "logging_steps": 500,
31
+ "max_steps": 1656,
32
+ "num_input_tokens_seen": 0,
33
+ "num_train_epochs": 2,
34
+ "save_steps": 500,
35
+ "stateful_callbacks": {
36
+ "TrainerControl": {
37
+ "args": {
38
+ "should_epoch_stop": false,
39
+ "should_evaluate": false,
40
+ "should_log": false,
41
+ "should_save": true,
42
+ "should_training_stop": false
43
+ },
44
+ "attributes": {}
45
+ }
46
+ },
47
+ "total_flos": 2.952457489941922e+18,
48
+ "train_batch_size": 32,
49
+ "trial_name": null,
50
+ "trial_params": null
51
+ }
checkpoint-828/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b82b36bb262e454f8e5c5fe3efc83cadeb46f932ed1ecf2e64afcef391dc64f0
3
+ size 5304
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Anime Picture",
7
+ "1": "Hentai Picture",
8
+ "2": "Normal or Safe",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime Picture": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai Picture": 1,
17
+ "Normal or Safe": 2,
18
+ "Pornography": 3
19
+ },
20
+ "model_type": "siglip",
21
+ "problem_type": "single_label_classification",
22
+ "text_config": {
23
+ "attention_dropout": 0.0,
24
+ "hidden_act": "gelu_pytorch_tanh",
25
+ "hidden_size": 768,
26
+ "intermediate_size": 3072,
27
+ "layer_norm_eps": 1e-06,
28
+ "max_position_embeddings": 64,
29
+ "model_type": "siglip_text_model",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "projection_size": 768,
33
+ "torch_dtype": "float32",
34
+ "vocab_size": 256000
35
+ },
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.50.0",
38
+ "vision_config": {
39
+ "attention_dropout": 0.0,
40
+ "hidden_act": "gelu_pytorch_tanh",
41
+ "hidden_size": 768,
42
+ "image_size": 256,
43
+ "intermediate_size": 3072,
44
+ "layer_norm_eps": 1e-06,
45
+ "model_type": "siglip_vision_model",
46
+ "num_attention_heads": 12,
47
+ "num_channels": 3,
48
+ "num_hidden_layers": 12,
49
+ "patch_size": 32,
50
+ "torch_dtype": "float32"
51
+ }
52
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26fdf4b6b7d682bb4bad93cef5112ed0e59b003cecd8cfbe1859ee04b18bca06
3
+ size 378249612
preprocessor_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "SiglipImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "processor_class": "SiglipProcessor",
18
+ "resample": 2,
19
+ "rescale_factor": 0.00392156862745098,
20
+ "size": {
21
+ "height": 256,
22
+ "width": 256
23
+ }
24
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b82b36bb262e454f8e5c5fe3efc83cadeb46f932ed1ecf2e64afcef391dc64f0
3
+ size 5304