Hugging Face's logo

prithivMLmods
/
Image-Guard-ckpt-3312

Image Classification
Transformers
Safetensors
English
siglip
text-generation-inference
siglip2
image-filter
safe-image-moderation
adult-content-filter
content-safety
anime-detection
ai-safety
Model card Files
xet
Community
prithivMLmods commited on
Commit
4c8b634
·
verified ·
1 Parent(s): f1f1ffb

Upload folder using huggingface_hub

Browse files
Files changed (28) 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-2484/config.json +52 -0
  10. checkpoint-2484/model.safetensors +3 -0
  11. checkpoint-2484/optimizer.pt +3 -0
  12. checkpoint-2484/preprocessor_config.json +24 -0
  13. checkpoint-2484/rng_state.pth +3 -0
  14. checkpoint-2484/scheduler.pt +3 -0
  15. checkpoint-2484/trainer_state.json +92 -0
  16. checkpoint-2484/training_args.bin +3 -0
  17. checkpoint-3312/config.json +52 -0
  18. checkpoint-3312/model.safetensors +3 -0
  19. checkpoint-3312/optimizer.pt +3 -0
  20. checkpoint-3312/preprocessor_config.json +24 -0
  21. checkpoint-3312/rng_state.pth +3 -0
  22. checkpoint-3312/scheduler.pt +3 -0
  23. checkpoint-3312/trainer_state.json +116 -0
  24. checkpoint-3312/training_args.bin +3 -0
  25. config.json +52 -0
  26. model.safetensors +3 -0
  27. preprocessor_config.json +24 -0
  28. 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-SFW",
7
+ "1": "Hentai",
8
+ "2": "Normal-SFW",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime-SFW": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai": 1,
17
+ "Normal-SFW": 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": 224,
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": 16,
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 sha256:3cc80dd4b9283b9f59fe84914774307565dbab4c268b0ef6d24375b3f8fe2199
3
+ size 371577220
checkpoint-1656/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8414dc5ecaf49ab2dcbc196c40c5a4705db0e00bd61c4ffdffa6224ab7d06ed9
3
+ size 686574667
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": 224,
22
+ "width": 224
23
+ }
24
+ }
checkpoint-1656/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28084d42c87bbada89e294ecf25b2845eb7ec8ea4a2670824c66be70cb1719a3
3
+ size 14645
checkpoint-1656/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe832b07084f2c7df03d6102ca9410d5d3719dfa5b4997b9b76d95904031cc5e
3
+ size 1465
checkpoint-1656/trainer_state.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1656,
3
+ "best_metric": 0.4855363368988037,
4
+ "best_model_checkpoint": "Image-Guard-2.0/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": 6.096656322479248,
15
+ "learning_rate": 0.0001724095646842428,
16
+ "loss": 0.9437,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.6948608541328399,
22
+ "eval_loss": 0.7906858325004578,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 498.3996,
25
+ "eval_samples_per_second": 53.136,
26
+ "eval_steps_per_second": 6.643,
27
+ "step": 828
28
+ },
29
+ {
30
+ "epoch": 1.2077294685990339,
31
+ "grad_norm": 6.137750625610352,
32
+ "learning_rate": 0.00014175352544451258,
33
+ "loss": 0.7179,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 1.8115942028985508,
38
+ "grad_norm": 8.953537940979004,
39
+ "learning_rate": 0.00011109748620478235,
40
+ "loss": 0.638,
41
+ "step": 1500
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.8164482875807122,
46
+ "eval_loss": 0.4855363368988037,
47
+ "eval_model_preparation_time": 0.0023,
48
+ "eval_runtime": 499.4314,
49
+ "eval_samples_per_second": 53.026,
50
+ "eval_steps_per_second": 6.63,
51
+ "step": 1656
52
+ }
53
+ ],
54
+ "logging_steps": 500,
55
+ "max_steps": 3312,
56
+ "num_input_tokens_seen": 0,
57
+ "num_train_epochs": 4,
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": false
67
+ },
68
+ "attributes": {}
69
+ }
70
+ },
71
+ "total_flos": 4.436303952083448e+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 sha256:23a5276d4ee853f292e1cf086068008dce11bbb85b7eb61befcc23226fb81438
3
+ size 5713
checkpoint-2484/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Anime-SFW",
7
+ "1": "Hentai",
8
+ "2": "Normal-SFW",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime-SFW": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai": 1,
17
+ "Normal-SFW": 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": 224,
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": 16,
50
+ "torch_dtype": "float32"
51
+ }
52
+ }
checkpoint-2484/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:667cfd12beac9102851f57c0acedac9b7b415dbbf8e3bdfbb5fc70e71f914b8a
3
+ size 371577220
checkpoint-2484/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0df9bf352e7cc9eaf192276483870d74ebf4048f88efb86f9843540bb1b88616
3
+ size 686574667
checkpoint-2484/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": 224,
22
+ "width": 224
23
+ }
24
+ }
checkpoint-2484/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c33ee76f763942463e21d381009e61f5b2de5891def23619d96beadf3f79235f
3
+ size 14645
checkpoint-2484/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a1bf925656d7c590ef87de6874a77500252d5d5c15b1c8066244db657c28095
3
+ size 1465
checkpoint-2484/trainer_state.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2484,
3
+ "best_metric": 0.36775290966033936,
4
+ "best_model_checkpoint": "Image-Guard-2.0/checkpoint-2484",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2484,
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": 6.096656322479248,
15
+ "learning_rate": 0.0001724095646842428,
16
+ "loss": 0.9437,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.6948608541328399,
22
+ "eval_loss": 0.7906858325004578,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 498.3996,
25
+ "eval_samples_per_second": 53.136,
26
+ "eval_steps_per_second": 6.643,
27
+ "step": 828
28
+ },
29
+ {
30
+ "epoch": 1.2077294685990339,
31
+ "grad_norm": 6.137750625610352,
32
+ "learning_rate": 0.00014175352544451258,
33
+ "loss": 0.7179,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 1.8115942028985508,
38
+ "grad_norm": 8.953537940979004,
39
+ "learning_rate": 0.00011109748620478235,
40
+ "loss": 0.638,
41
+ "step": 1500
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.8164482875807122,
46
+ "eval_loss": 0.4855363368988037,
47
+ "eval_model_preparation_time": 0.0023,
48
+ "eval_runtime": 499.4314,
49
+ "eval_samples_per_second": 53.026,
50
+ "eval_steps_per_second": 6.63,
51
+ "step": 1656
52
+ },
53
+ {
54
+ "epoch": 2.4154589371980677,
55
+ "grad_norm": 7.226722240447998,
56
+ "learning_rate": 8.044144696505211e-05,
57
+ "loss": 0.526,
58
+ "step": 2000
59
+ },
60
+ {
61
+ "epoch": 3.0,
62
+ "eval_accuracy": 0.8618358947249178,
63
+ "eval_loss": 0.36775290966033936,
64
+ "eval_model_preparation_time": 0.0023,
65
+ "eval_runtime": 500.3065,
66
+ "eval_samples_per_second": 52.934,
67
+ "eval_steps_per_second": 6.618,
68
+ "step": 2484
69
+ }
70
+ ],
71
+ "logging_steps": 500,
72
+ "max_steps": 3312,
73
+ "num_input_tokens_seen": 0,
74
+ "num_train_epochs": 4,
75
+ "save_steps": 500,
76
+ "stateful_callbacks": {
77
+ "TrainerControl": {
78
+ "args": {
79
+ "should_epoch_stop": false,
80
+ "should_evaluate": false,
81
+ "should_log": false,
82
+ "should_save": true,
83
+ "should_training_stop": false
84
+ },
85
+ "attributes": {}
86
+ }
87
+ },
88
+ "total_flos": 6.654455928125172e+18,
89
+ "train_batch_size": 32,
90
+ "trial_name": null,
91
+ "trial_params": null
92
+ }
checkpoint-2484/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23a5276d4ee853f292e1cf086068008dce11bbb85b7eb61befcc23226fb81438
3
+ size 5713
checkpoint-3312/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Anime-SFW",
7
+ "1": "Hentai",
8
+ "2": "Normal-SFW",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime-SFW": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai": 1,
17
+ "Normal-SFW": 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": 224,
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": 16,
50
+ "torch_dtype": "float32"
51
+ }
52
+ }
checkpoint-3312/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e2ce468aa096b2f90aaa65c26a9cf35a4d569c970929c355e917e4bd11d03ba
3
+ size 371577220
checkpoint-3312/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f51d230dd96d8a0bfd6e37da8b67fc7eb0028a64106fea7feb19f3950e0d898
3
+ size 686574667
checkpoint-3312/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": 224,
22
+ "width": 224
23
+ }
24
+ }
checkpoint-3312/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afa650eaf5f28bb150e12307b561d61ce6ce87d145ddf52a5eb9665f34eab375
3
+ size 14645
checkpoint-3312/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f0bb21237f97eca32a2c098f63d123dccc7572edc12ac4d2e9f8327ca8fd5cd
3
+ size 1465
checkpoint-3312/trainer_state.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 3312,
3
+ "best_metric": 0.28185972571372986,
4
+ "best_model_checkpoint": "Image-Guard-2.0/checkpoint-3312",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3312,
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": 6.096656322479248,
15
+ "learning_rate": 0.0001724095646842428,
16
+ "loss": 0.9437,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.6948608541328399,
22
+ "eval_loss": 0.7906858325004578,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 498.3996,
25
+ "eval_samples_per_second": 53.136,
26
+ "eval_steps_per_second": 6.643,
27
+ "step": 828
28
+ },
29
+ {
30
+ "epoch": 1.2077294685990339,
31
+ "grad_norm": 6.137750625610352,
32
+ "learning_rate": 0.00014175352544451258,
33
+ "loss": 0.7179,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 1.8115942028985508,
38
+ "grad_norm": 8.953537940979004,
39
+ "learning_rate": 0.00011109748620478235,
40
+ "loss": 0.638,
41
+ "step": 1500
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.8164482875807122,
46
+ "eval_loss": 0.4855363368988037,
47
+ "eval_model_preparation_time": 0.0023,
48
+ "eval_runtime": 499.4314,
49
+ "eval_samples_per_second": 53.026,
50
+ "eval_steps_per_second": 6.63,
51
+ "step": 1656
52
+ },
53
+ {
54
+ "epoch": 2.4154589371980677,
55
+ "grad_norm": 7.226722240447998,
56
+ "learning_rate": 8.044144696505211e-05,
57
+ "loss": 0.526,
58
+ "step": 2000
59
+ },
60
+ {
61
+ "epoch": 3.0,
62
+ "eval_accuracy": 0.8618358947249178,
63
+ "eval_loss": 0.36775290966033936,
64
+ "eval_model_preparation_time": 0.0023,
65
+ "eval_runtime": 500.3065,
66
+ "eval_samples_per_second": 52.934,
67
+ "eval_steps_per_second": 6.618,
68
+ "step": 2484
69
+ },
70
+ {
71
+ "epoch": 3.0193236714975846,
72
+ "grad_norm": 6.0987372398376465,
73
+ "learning_rate": 4.9785407725321894e-05,
74
+ "loss": 0.4704,
75
+ "step": 2500
76
+ },
77
+ {
78
+ "epoch": 3.6231884057971016,
79
+ "grad_norm": 6.0757341384887695,
80
+ "learning_rate": 1.9129368485591662e-05,
81
+ "loss": 0.3812,
82
+ "step": 3000
83
+ },
84
+ {
85
+ "epoch": 4.0,
86
+ "eval_accuracy": 0.8942340369293509,
87
+ "eval_loss": 0.28185972571372986,
88
+ "eval_model_preparation_time": 0.0023,
89
+ "eval_runtime": 502.9996,
90
+ "eval_samples_per_second": 52.65,
91
+ "eval_steps_per_second": 6.583,
92
+ "step": 3312
93
+ }
94
+ ],
95
+ "logging_steps": 500,
96
+ "max_steps": 3312,
97
+ "num_input_tokens_seen": 0,
98
+ "num_train_epochs": 4,
99
+ "save_steps": 500,
100
+ "stateful_callbacks": {
101
+ "TrainerControl": {
102
+ "args": {
103
+ "should_epoch_stop": false,
104
+ "should_evaluate": false,
105
+ "should_log": false,
106
+ "should_save": true,
107
+ "should_training_stop": true
108
+ },
109
+ "attributes": {}
110
+ }
111
+ },
112
+ "total_flos": 8.872607904166896e+18,
113
+ "train_batch_size": 32,
114
+ "trial_name": null,
115
+ "trial_params": null
116
+ }
checkpoint-3312/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23a5276d4ee853f292e1cf086068008dce11bbb85b7eb61befcc23226fb81438
3
+ size 5713
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Anime-SFW",
7
+ "1": "Hentai",
8
+ "2": "Normal-SFW",
9
+ "3": "Pornography",
10
+ "4": "Enticing or Sensual"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Anime-SFW": 0,
15
+ "Enticing or Sensual": 4,
16
+ "Hentai": 1,
17
+ "Normal-SFW": 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": 224,
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": 16,
50
+ "torch_dtype": "float32"
51
+ }
52
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e2ce468aa096b2f90aaa65c26a9cf35a4d569c970929c355e917e4bd11d03ba
3
+ size 371577220
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": 224,
22
+ "width": 224
23
+ }
24
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23a5276d4ee853f292e1cf086068008dce11bbb85b7eb61befcc23226fb81438
3
+ size 5713