Hugging Face's logo

prithivMLmods
/
OpenSDI-SD1.5-SigLIP2

Image Classification
Transformers
Safetensors
English
siglip
OpenSDI
Spotting Diffusion-Generated Images in the Open World
SD1.5
AI-vs-Real
SigLIP2
Stable Diffusion v1-5
Model card Files
xet
prithivMLmods commited on
Commit
0d939d5
·
verified ·
1 Parent(s): a211121

Upload folder using huggingface_hub

Browse files
Files changed (36) hide show
  1. checkpoint-1250/config.json +46 -0
  2. checkpoint-1250/model.safetensors +3 -0
  3. checkpoint-1250/optimizer.pt +3 -0
  4. checkpoint-1250/preprocessor_config.json +24 -0
  5. checkpoint-1250/rng_state.pth +3 -0
  6. checkpoint-1250/scheduler.pt +3 -0
  7. checkpoint-1250/trainer_state.json +68 -0
  8. checkpoint-1250/training_args.bin +3 -0
  9. checkpoint-1875/config.json +46 -0
  10. checkpoint-1875/model.safetensors +3 -0
  11. checkpoint-1875/optimizer.pt +3 -0
  12. checkpoint-1875/preprocessor_config.json +24 -0
  13. checkpoint-1875/rng_state.pth +3 -0
  14. checkpoint-1875/scheduler.pt +3 -0
  15. checkpoint-1875/trainer_state.json +85 -0
  16. checkpoint-1875/training_args.bin +3 -0
  17. checkpoint-2500/config.json +46 -0
  18. checkpoint-2500/model.safetensors +3 -0
  19. checkpoint-2500/optimizer.pt +3 -0
  20. checkpoint-2500/preprocessor_config.json +24 -0
  21. checkpoint-2500/rng_state.pth +3 -0
  22. checkpoint-2500/scheduler.pt +3 -0
  23. checkpoint-2500/trainer_state.json +109 -0
  24. checkpoint-2500/training_args.bin +3 -0
  25. checkpoint-625/config.json +46 -0
  26. checkpoint-625/model.safetensors +3 -0
  27. checkpoint-625/optimizer.pt +3 -0
  28. checkpoint-625/preprocessor_config.json +24 -0
  29. checkpoint-625/rng_state.pth +3 -0
  30. checkpoint-625/scheduler.pt +3 -0
  31. checkpoint-625/trainer_state.json +51 -0
  32. checkpoint-625/training_args.bin +3 -0
  33. config.json +46 -0
  34. model.safetensors +3 -0
  35. preprocessor_config.json +24 -0
  36. training_args.bin +3 -0
checkpoint-1250/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Real_Image",
7
+ "1": "SD1.5_Generated"
8
+ },
9
+ "initializer_factor": 1.0,
10
+ "label2id": {
11
+ "Real_Image": 0,
12
+ "SD1.5_Generated": 1
13
+ },
14
+ "model_type": "siglip",
15
+ "problem_type": "single_label_classification",
16
+ "text_config": {
17
+ "attention_dropout": 0.0,
18
+ "hidden_act": "gelu_pytorch_tanh",
19
+ "hidden_size": 768,
20
+ "intermediate_size": 3072,
21
+ "layer_norm_eps": 1e-06,
22
+ "max_position_embeddings": 64,
23
+ "model_type": "siglip_text_model",
24
+ "num_attention_heads": 12,
25
+ "num_hidden_layers": 12,
26
+ "projection_size": 768,
27
+ "torch_dtype": "float32",
28
+ "vocab_size": 256000
29
+ },
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.50.0",
32
+ "vision_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "image_size": 224,
37
+ "intermediate_size": 3072,
38
+ "layer_norm_eps": 1e-06,
39
+ "model_type": "siglip_vision_model",
40
+ "num_attention_heads": 12,
41
+ "num_channels": 3,
42
+ "num_hidden_layers": 12,
43
+ "patch_size": 16,
44
+ "torch_dtype": "float32"
45
+ }
46
+ }
checkpoint-1250/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d19e9d82ca87749108a24eba8bdfcd2d23ca5b58be7bfd323c4cdcc0c17e285d
3
+ size 371567992
checkpoint-1250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0698568d73a654fe6b1e0b444730631971f05a03b27f9b9ab672958ff82dd768
3
+ size 686555770
checkpoint-1250/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-1250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8e47c81b882f33c5e2f4e777f5a233ce5a180fe1b81e4d6371f9de3fbb3170c
3
+ size 14244
checkpoint-1250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4813f8726a327ba28582e848e96c014bd40ce7363feac27e775fabad1f6373c3
3
+ size 1064
checkpoint-1250/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1250,
3
+ "best_metric": 0.34147578477859497,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1250",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1250,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8,
14
+ "grad_norm": 12.590618133544922,
15
+ "learning_rate": 1.63265306122449e-05,
16
+ "loss": 0.5408,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.7876,
22
+ "eval_loss": 0.4367133677005768,
23
+ "eval_model_preparation_time": 0.0039,
24
+ "eval_runtime": 439.5267,
25
+ "eval_samples_per_second": 45.503,
26
+ "eval_steps_per_second": 5.688,
27
+ "step": 625
28
+ },
29
+ {
30
+ "epoch": 1.6,
31
+ "grad_norm": 32.36686706542969,
32
+ "learning_rate": 1.2244897959183674e-05,
33
+ "loss": 0.415,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.8439,
39
+ "eval_loss": 0.34147578477859497,
40
+ "eval_model_preparation_time": 0.0039,
41
+ "eval_runtime": 436.8014,
42
+ "eval_samples_per_second": 45.787,
43
+ "eval_steps_per_second": 5.723,
44
+ "step": 1250
45
+ }
46
+ ],
47
+ "logging_steps": 500,
48
+ "max_steps": 2500,
49
+ "num_input_tokens_seen": 0,
50
+ "num_train_epochs": 4,
51
+ "save_steps": 500,
52
+ "stateful_callbacks": {
53
+ "TrainerControl": {
54
+ "args": {
55
+ "should_epoch_stop": false,
56
+ "should_evaluate": false,
57
+ "should_log": false,
58
+ "should_save": true,
59
+ "should_training_stop": false
60
+ },
61
+ "attributes": {}
62
+ }
63
+ },
64
+ "total_flos": 3.35021983285248e+18,
65
+ "train_batch_size": 32,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
checkpoint-1250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ddb587f2e0557170aa6e5a866a23675974cbc6f0ada6180b0165721a997b3b2
3
+ size 5304
checkpoint-1875/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Real_Image",
7
+ "1": "SD1.5_Generated"
8
+ },
9
+ "initializer_factor": 1.0,
10
+ "label2id": {
11
+ "Real_Image": 0,
12
+ "SD1.5_Generated": 1
13
+ },
14
+ "model_type": "siglip",
15
+ "problem_type": "single_label_classification",
16
+ "text_config": {
17
+ "attention_dropout": 0.0,
18
+ "hidden_act": "gelu_pytorch_tanh",
19
+ "hidden_size": 768,
20
+ "intermediate_size": 3072,
21
+ "layer_norm_eps": 1e-06,
22
+ "max_position_embeddings": 64,
23
+ "model_type": "siglip_text_model",
24
+ "num_attention_heads": 12,
25
+ "num_hidden_layers": 12,
26
+ "projection_size": 768,
27
+ "torch_dtype": "float32",
28
+ "vocab_size": 256000
29
+ },
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.50.0",
32
+ "vision_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "image_size": 224,
37
+ "intermediate_size": 3072,
38
+ "layer_norm_eps": 1e-06,
39
+ "model_type": "siglip_vision_model",
40
+ "num_attention_heads": 12,
41
+ "num_channels": 3,
42
+ "num_hidden_layers": 12,
43
+ "patch_size": 16,
44
+ "torch_dtype": "float32"
45
+ }
46
+ }
checkpoint-1875/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:385ff99a0b7d19eaec1f9c7ce3c60279f2a3739ea800eeb37b9a2c5048cbc113
3
+ size 371567992
checkpoint-1875/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6eaa3f308d23cebf8d01dfb096894e5d7b974b36ebd6618f5b6d3571a123e790
3
+ size 686555770
checkpoint-1875/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-1875/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72f3bf471f7d501498c848c5bc931fdb3b46189bbc2a8c7e4dca5a1e3e186cbb
3
+ size 14244
checkpoint-1875/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30982c4fd6c689e8edb988a00752fc2dbd10270c951931f0ef6bcd84a8183f28
3
+ size 1064
checkpoint-1875/trainer_state.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1875,
3
+ "best_metric": 0.2799457609653473,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1875",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1875,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8,
14
+ "grad_norm": 12.590618133544922,
15
+ "learning_rate": 1.63265306122449e-05,
16
+ "loss": 0.5408,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.7876,
22
+ "eval_loss": 0.4367133677005768,
23
+ "eval_model_preparation_time": 0.0039,
24
+ "eval_runtime": 439.5267,
25
+ "eval_samples_per_second": 45.503,
26
+ "eval_steps_per_second": 5.688,
27
+ "step": 625
28
+ },
29
+ {
30
+ "epoch": 1.6,
31
+ "grad_norm": 32.36686706542969,
32
+ "learning_rate": 1.2244897959183674e-05,
33
+ "loss": 0.415,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.8439,
39
+ "eval_loss": 0.34147578477859497,
40
+ "eval_model_preparation_time": 0.0039,
41
+ "eval_runtime": 436.8014,
42
+ "eval_samples_per_second": 45.787,
43
+ "eval_steps_per_second": 5.723,
44
+ "step": 1250
45
+ },
46
+ {
47
+ "epoch": 2.4,
48
+ "grad_norm": 10.255178451538086,
49
+ "learning_rate": 8.16326530612245e-06,
50
+ "loss": 0.3582,
51
+ "step": 1500
52
+ },
53
+ {
54
+ "epoch": 3.0,
55
+ "eval_accuracy": 0.8738,
56
+ "eval_loss": 0.2799457609653473,
57
+ "eval_model_preparation_time": 0.0039,
58
+ "eval_runtime": 441.0603,
59
+ "eval_samples_per_second": 45.345,
60
+ "eval_steps_per_second": 5.668,
61
+ "step": 1875
62
+ }
63
+ ],
64
+ "logging_steps": 500,
65
+ "max_steps": 2500,
66
+ "num_input_tokens_seen": 0,
67
+ "num_train_epochs": 4,
68
+ "save_steps": 500,
69
+ "stateful_callbacks": {
70
+ "TrainerControl": {
71
+ "args": {
72
+ "should_epoch_stop": false,
73
+ "should_evaluate": false,
74
+ "should_log": false,
75
+ "should_save": true,
76
+ "should_training_stop": false
77
+ },
78
+ "attributes": {}
79
+ }
80
+ },
81
+ "total_flos": 5.02532974927872e+18,
82
+ "train_batch_size": 32,
83
+ "trial_name": null,
84
+ "trial_params": null
85
+ }
checkpoint-1875/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ddb587f2e0557170aa6e5a866a23675974cbc6f0ada6180b0165721a997b3b2
3
+ size 5304
checkpoint-2500/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Real_Image",
7
+ "1": "SD1.5_Generated"
8
+ },
9
+ "initializer_factor": 1.0,
10
+ "label2id": {
11
+ "Real_Image": 0,
12
+ "SD1.5_Generated": 1
13
+ },
14
+ "model_type": "siglip",
15
+ "problem_type": "single_label_classification",
16
+ "text_config": {
17
+ "attention_dropout": 0.0,
18
+ "hidden_act": "gelu_pytorch_tanh",
19
+ "hidden_size": 768,
20
+ "intermediate_size": 3072,
21
+ "layer_norm_eps": 1e-06,
22
+ "max_position_embeddings": 64,
23
+ "model_type": "siglip_text_model",
24
+ "num_attention_heads": 12,
25
+ "num_hidden_layers": 12,
26
+ "projection_size": 768,
27
+ "torch_dtype": "float32",
28
+ "vocab_size": 256000
29
+ },
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.50.0",
32
+ "vision_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "image_size": 224,
37
+ "intermediate_size": 3072,
38
+ "layer_norm_eps": 1e-06,
39
+ "model_type": "siglip_vision_model",
40
+ "num_attention_heads": 12,
41
+ "num_channels": 3,
42
+ "num_hidden_layers": 12,
43
+ "patch_size": 16,
44
+ "torch_dtype": "float32"
45
+ }
46
+ }
checkpoint-2500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cbf1a92c45666035c1d684a1e28d66bf49a60ba5bcf3b0b154e0de5ac38e55a
3
+ size 371567992
checkpoint-2500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db1850282149eee8b7d34610d17dbee7f2c897bbf66d91079c38a29467083bd1
3
+ size 686555770
checkpoint-2500/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-2500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe60017b72836415a5929c7f857b37e46b93ad59a05951f9779b2e5fe3ac5c42
3
+ size 14244
checkpoint-2500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3d944f5855dedd93de5810fb83351c896ba8bcf8ae6abf8ac51fb2013da8b6d
3
+ size 1064
checkpoint-2500/trainer_state.json ADDED
@@ -0,0 +1,109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2500,
3
+ "best_metric": 0.19813919067382812,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-2500",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8,
14
+ "grad_norm": 12.590618133544922,
15
+ "learning_rate": 1.63265306122449e-05,
16
+ "loss": 0.5408,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.7876,
22
+ "eval_loss": 0.4367133677005768,
23
+ "eval_model_preparation_time": 0.0039,
24
+ "eval_runtime": 439.5267,
25
+ "eval_samples_per_second": 45.503,
26
+ "eval_steps_per_second": 5.688,
27
+ "step": 625
28
+ },
29
+ {
30
+ "epoch": 1.6,
31
+ "grad_norm": 32.36686706542969,
32
+ "learning_rate": 1.2244897959183674e-05,
33
+ "loss": 0.415,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.8439,
39
+ "eval_loss": 0.34147578477859497,
40
+ "eval_model_preparation_time": 0.0039,
41
+ "eval_runtime": 436.8014,
42
+ "eval_samples_per_second": 45.787,
43
+ "eval_steps_per_second": 5.723,
44
+ "step": 1250
45
+ },
46
+ {
47
+ "epoch": 2.4,
48
+ "grad_norm": 10.255178451538086,
49
+ "learning_rate": 8.16326530612245e-06,
50
+ "loss": 0.3582,
51
+ "step": 1500
52
+ },
53
+ {
54
+ "epoch": 3.0,
55
+ "eval_accuracy": 0.8738,
56
+ "eval_loss": 0.2799457609653473,
57
+ "eval_model_preparation_time": 0.0039,
58
+ "eval_runtime": 441.0603,
59
+ "eval_samples_per_second": 45.345,
60
+ "eval_steps_per_second": 5.668,
61
+ "step": 1875
62
+ },
63
+ {
64
+ "epoch": 3.2,
65
+ "grad_norm": 26.685434341430664,
66
+ "learning_rate": 4.081632653061225e-06,
67
+ "loss": 0.2901,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 4.0,
72
+ "grad_norm": 15.797358512878418,
73
+ "learning_rate": 0.0,
74
+ "loss": 0.2258,
75
+ "step": 2500
76
+ },
77
+ {
78
+ "epoch": 4.0,
79
+ "eval_accuracy": 0.9164,
80
+ "eval_loss": 0.19813919067382812,
81
+ "eval_model_preparation_time": 0.0039,
82
+ "eval_runtime": 436.7326,
83
+ "eval_samples_per_second": 45.795,
84
+ "eval_steps_per_second": 5.724,
85
+ "step": 2500
86
+ }
87
+ ],
88
+ "logging_steps": 500,
89
+ "max_steps": 2500,
90
+ "num_input_tokens_seen": 0,
91
+ "num_train_epochs": 4,
92
+ "save_steps": 500,
93
+ "stateful_callbacks": {
94
+ "TrainerControl": {
95
+ "args": {
96
+ "should_epoch_stop": false,
97
+ "should_evaluate": false,
98
+ "should_log": false,
99
+ "should_save": true,
100
+ "should_training_stop": true
101
+ },
102
+ "attributes": {}
103
+ }
104
+ },
105
+ "total_flos": 6.70043966570496e+18,
106
+ "train_batch_size": 32,
107
+ "trial_name": null,
108
+ "trial_params": null
109
+ }
checkpoint-2500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ddb587f2e0557170aa6e5a866a23675974cbc6f0ada6180b0165721a997b3b2
3
+ size 5304
checkpoint-625/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Real_Image",
7
+ "1": "SD1.5_Generated"
8
+ },
9
+ "initializer_factor": 1.0,
10
+ "label2id": {
11
+ "Real_Image": 0,
12
+ "SD1.5_Generated": 1
13
+ },
14
+ "model_type": "siglip",
15
+ "problem_type": "single_label_classification",
16
+ "text_config": {
17
+ "attention_dropout": 0.0,
18
+ "hidden_act": "gelu_pytorch_tanh",
19
+ "hidden_size": 768,
20
+ "intermediate_size": 3072,
21
+ "layer_norm_eps": 1e-06,
22
+ "max_position_embeddings": 64,
23
+ "model_type": "siglip_text_model",
24
+ "num_attention_heads": 12,
25
+ "num_hidden_layers": 12,
26
+ "projection_size": 768,
27
+ "torch_dtype": "float32",
28
+ "vocab_size": 256000
29
+ },
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.50.0",
32
+ "vision_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "image_size": 224,
37
+ "intermediate_size": 3072,
38
+ "layer_norm_eps": 1e-06,
39
+ "model_type": "siglip_vision_model",
40
+ "num_attention_heads": 12,
41
+ "num_channels": 3,
42
+ "num_hidden_layers": 12,
43
+ "patch_size": 16,
44
+ "torch_dtype": "float32"
45
+ }
46
+ }
checkpoint-625/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cceaf6eb9b02914f5cce11bf7c89c27c5b455fb1255aac709f338df5876f85af
3
+ size 371567992
checkpoint-625/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d9e688d02d37c95c57f5c914c3e3022db783c059971635eda89360ab434ade7
3
+ size 686555770
checkpoint-625/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-625/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b97d2b074f45d2c8636b033e7029690ff74569ff4607f775707cdd593a86f2d8
3
+ size 14244
checkpoint-625/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:167502d7d8363065efda5eb4cca98b2dccacc5791c3382362431a13f215602ff
3
+ size 1064
checkpoint-625/trainer_state.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 625,
3
+ "best_metric": 0.4367133677005768,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-625",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 625,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8,
14
+ "grad_norm": 12.590618133544922,
15
+ "learning_rate": 1.63265306122449e-05,
16
+ "loss": 0.5408,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.7876,
22
+ "eval_loss": 0.4367133677005768,
23
+ "eval_model_preparation_time": 0.0039,
24
+ "eval_runtime": 439.5267,
25
+ "eval_samples_per_second": 45.503,
26
+ "eval_steps_per_second": 5.688,
27
+ "step": 625
28
+ }
29
+ ],
30
+ "logging_steps": 500,
31
+ "max_steps": 2500,
32
+ "num_input_tokens_seen": 0,
33
+ "num_train_epochs": 4,
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": 1.67510991642624e+18,
48
+ "train_batch_size": 32,
49
+ "trial_name": null,
50
+ "trial_params": null
51
+ }
checkpoint-625/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ddb587f2e0557170aa6e5a866a23675974cbc6f0ada6180b0165721a997b3b2
3
+ size 5304
config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Real_Image",
7
+ "1": "SD1.5_Generated"
8
+ },
9
+ "initializer_factor": 1.0,
10
+ "label2id": {
11
+ "Real_Image": 0,
12
+ "SD1.5_Generated": 1
13
+ },
14
+ "model_type": "siglip",
15
+ "problem_type": "single_label_classification",
16
+ "text_config": {
17
+ "attention_dropout": 0.0,
18
+ "hidden_act": "gelu_pytorch_tanh",
19
+ "hidden_size": 768,
20
+ "intermediate_size": 3072,
21
+ "layer_norm_eps": 1e-06,
22
+ "max_position_embeddings": 64,
23
+ "model_type": "siglip_text_model",
24
+ "num_attention_heads": 12,
25
+ "num_hidden_layers": 12,
26
+ "projection_size": 768,
27
+ "torch_dtype": "float32",
28
+ "vocab_size": 256000
29
+ },
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.50.0",
32
+ "vision_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "image_size": 224,
37
+ "intermediate_size": 3072,
38
+ "layer_norm_eps": 1e-06,
39
+ "model_type": "siglip_vision_model",
40
+ "num_attention_heads": 12,
41
+ "num_channels": 3,
42
+ "num_hidden_layers": 12,
43
+ "patch_size": 16,
44
+ "torch_dtype": "float32"
45
+ }
46
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cbf1a92c45666035c1d684a1e28d66bf49a60ba5bcf3b0b154e0de5ac38e55a
3
+ size 371567992
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 sha256:5ddb587f2e0557170aa6e5a866a23675974cbc6f0ada6180b0165721a997b3b2
3
+ size 5304