Hugging Face's logo

prithivMLmods
/
Fashion-Mnist-SigLIP2

Image Classification
Transformers
Safetensors
English
siglip
fashion
mnist
siglip2
Model card Files
xet
 Community
prithivMLmods commited on
Commit
d93f333
·
verified ·
1 Parent(s): df48e9c

Upload folder using huggingface_hub

Browse files
Files changed (28) hide show
  1. checkpoint-1875/config.json +62 -0
  2. checkpoint-1875/model.safetensors +3 -0
  3. checkpoint-1875/optimizer.pt +3 -0
  4. checkpoint-1875/preprocessor_config.json +24 -0
  5. checkpoint-1875/rng_state.pth +3 -0
  6. checkpoint-1875/scheduler.pt +3 -0
  7. checkpoint-1875/trainer_state.json +65 -0
  8. checkpoint-1875/training_args.bin +3 -0
  9. checkpoint-3750/config.json +62 -0
  10. checkpoint-3750/model.safetensors +3 -0
  11. checkpoint-3750/optimizer.pt +3 -0
  12. checkpoint-3750/preprocessor_config.json +24 -0
  13. checkpoint-3750/rng_state.pth +3 -0
  14. checkpoint-3750/scheduler.pt +3 -0
  15. checkpoint-3750/trainer_state.json +103 -0
  16. checkpoint-3750/training_args.bin +3 -0
  17. checkpoint-5625/config.json +62 -0
  18. checkpoint-5625/model.safetensors +3 -0
  19. checkpoint-5625/optimizer.pt +3 -0
  20. checkpoint-5625/preprocessor_config.json +24 -0
  21. checkpoint-5625/rng_state.pth +3 -0
  22. checkpoint-5625/scheduler.pt +3 -0
  23. checkpoint-5625/trainer_state.json +141 -0
  24. checkpoint-5625/training_args.bin +3 -0
  25. config.json +62 -0
  26. model.safetensors +3 -0
  27. preprocessor_config.json +24 -0
  28. training_args.bin +3 -0
checkpoint-1875/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "T-shirt / top",
7
+ "1": "Trouser",
8
+ "2": "Pullover",
9
+ "3": "Dress",
10
+ "4": "Coat",
11
+ "5": "Sandal",
12
+ "6": "Shirt",
13
+ "7": "Sneaker",
14
+ "8": "Bag",
15
+ "9": "Ankle boot"
16
+ },
17
+ "initializer_factor": 1.0,
18
+ "label2id": {
19
+ "Ankle boot": 9,
20
+ "Bag": 8,
21
+ "Coat": 4,
22
+ "Dress": 3,
23
+ "Pullover": 2,
24
+ "Sandal": 5,
25
+ "Shirt": 6,
26
+ "Sneaker": 7,
27
+ "T-shirt / top": 0,
28
+ "Trouser": 1
29
+ },
30
+ "model_type": "siglip",
31
+ "problem_type": "single_label_classification",
32
+ "text_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "intermediate_size": 3072,
37
+ "layer_norm_eps": 1e-06,
38
+ "max_position_embeddings": 64,
39
+ "model_type": "siglip_text_model",
40
+ "num_attention_heads": 12,
41
+ "num_hidden_layers": 12,
42
+ "projection_size": 768,
43
+ "torch_dtype": "float32",
44
+ "vocab_size": 256000
45
+ },
46
+ "torch_dtype": "float32",
47
+ "transformers_version": "4.50.0.dev0",
48
+ "vision_config": {
49
+ "attention_dropout": 0.0,
50
+ "hidden_act": "gelu_pytorch_tanh",
51
+ "hidden_size": 768,
52
+ "image_size": 224,
53
+ "intermediate_size": 3072,
54
+ "layer_norm_eps": 1e-06,
55
+ "model_type": "siglip_vision_model",
56
+ "num_attention_heads": 12,
57
+ "num_channels": 3,
58
+ "num_hidden_layers": 12,
59
+ "patch_size": 16,
60
+ "torch_dtype": "float32"
61
+ }
62
+ }
checkpoint-1875/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de996bc92de10f2d1ef53b6678ea57f3c3ebcd31fb3d485729c6bb067e1af6bb
3
+ size 371592608
checkpoint-1875/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c75f03d71130e291a52f57b11c5ddc33e0fafeea0a7a621e1e4f19fc820b827
3
+ size 686604922
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 sha256:3f2b3ac1c991d659109219935d886c6afab3ba720ceb1fdeba6003e7056c2020
3
+ size 14244
checkpoint-1875/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7166565740237ef510ac3171e802c83d9f91a6182d5e5ae2db23019195ed5d7
3
+ size 1064
checkpoint-1875/trainer_state.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1875,
3
+ "best_metric": 0.46604594588279724,
4
+ "best_model_checkpoint": "Fashion-Mnist-SigLIP2/checkpoint-1875",
5
+ "epoch": 1.0,
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": 0.26666666666666666,
14
+ "grad_norm": 9.494325637817383,
15
+ "learning_rate": 0.00018385650224215247,
16
+ "loss": 0.9674,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.5333333333333333,
21
+ "grad_norm": 6.167793273925781,
22
+ "learning_rate": 0.00016591928251121076,
23
+ "loss": 0.6265,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.8,
28
+ "grad_norm": 5.353821754455566,
29
+ "learning_rate": 0.00014798206278026908,
30
+ "loss": 0.5464,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "eval_accuracy": 0.8283,
36
+ "eval_loss": 0.46604594588279724,
37
+ "eval_model_preparation_time": 0.003,
38
+ "eval_runtime": 687.6086,
39
+ "eval_samples_per_second": 87.259,
40
+ "eval_steps_per_second": 10.907,
41
+ "step": 1875
42
+ }
43
+ ],
44
+ "logging_steps": 500,
45
+ "max_steps": 5625,
46
+ "num_input_tokens_seen": 0,
47
+ "num_train_epochs": 3,
48
+ "save_steps": 500,
49
+ "stateful_callbacks": {
50
+ "TrainerControl": {
51
+ "args": {
52
+ "should_epoch_stop": false,
53
+ "should_evaluate": false,
54
+ "should_log": false,
55
+ "should_save": true,
56
+ "should_training_stop": false
57
+ },
58
+ "attributes": {}
59
+ }
60
+ },
61
+ "total_flos": 5.02566312665088e+18,
62
+ "train_batch_size": 32,
63
+ "trial_name": null,
64
+ "trial_params": null
65
+ }
checkpoint-1875/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2634502d14d50ae6e309c6720abc81b68a2cb5d4fd494874448126eedae2d19
3
+ size 5304
checkpoint-3750/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "T-shirt / top",
7
+ "1": "Trouser",
8
+ "2": "Pullover",
9
+ "3": "Dress",
10
+ "4": "Coat",
11
+ "5": "Sandal",
12
+ "6": "Shirt",
13
+ "7": "Sneaker",
14
+ "8": "Bag",
15
+ "9": "Ankle boot"
16
+ },
17
+ "initializer_factor": 1.0,
18
+ "label2id": {
19
+ "Ankle boot": 9,
20
+ "Bag": 8,
21
+ "Coat": 4,
22
+ "Dress": 3,
23
+ "Pullover": 2,
24
+ "Sandal": 5,
25
+ "Shirt": 6,
26
+ "Sneaker": 7,
27
+ "T-shirt / top": 0,
28
+ "Trouser": 1
29
+ },
30
+ "model_type": "siglip",
31
+ "problem_type": "single_label_classification",
32
+ "text_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "intermediate_size": 3072,
37
+ "layer_norm_eps": 1e-06,
38
+ "max_position_embeddings": 64,
39
+ "model_type": "siglip_text_model",
40
+ "num_attention_heads": 12,
41
+ "num_hidden_layers": 12,
42
+ "projection_size": 768,
43
+ "torch_dtype": "float32",
44
+ "vocab_size": 256000
45
+ },
46
+ "torch_dtype": "float32",
47
+ "transformers_version": "4.50.0.dev0",
48
+ "vision_config": {
49
+ "attention_dropout": 0.0,
50
+ "hidden_act": "gelu_pytorch_tanh",
51
+ "hidden_size": 768,
52
+ "image_size": 224,
53
+ "intermediate_size": 3072,
54
+ "layer_norm_eps": 1e-06,
55
+ "model_type": "siglip_vision_model",
56
+ "num_attention_heads": 12,
57
+ "num_channels": 3,
58
+ "num_hidden_layers": 12,
59
+ "patch_size": 16,
60
+ "torch_dtype": "float32"
61
+ }
62
+ }
checkpoint-3750/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4163d135e7dcb8f18e396ca545a9074fd0573a6663aca5e5667ca3098763be13
3
+ size 371592608
checkpoint-3750/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60c9f31ab0f5f3f2a30cd4739c3f5302376fe2e8764cc858a9255dedd9384e49
3
+ size 686604922
checkpoint-3750/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-3750/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7eee6d9676312234f08fff7b0b5a5a73dc76adcb975c0fe0faa1746456545b1d
3
+ size 14244
checkpoint-3750/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e9430fe0a14226de43a1b2aa321c66cdcac9d6fa43887f2ce6af41a0dfe0299
3
+ size 1064
checkpoint-3750/trainer_state.json ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 3750,
3
+ "best_metric": 0.31374436616897583,
4
+ "best_model_checkpoint": "Fashion-Mnist-SigLIP2/checkpoint-3750",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3750,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.26666666666666666,
14
+ "grad_norm": 9.494325637817383,
15
+ "learning_rate": 0.00018385650224215247,
16
+ "loss": 0.9674,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.5333333333333333,
21
+ "grad_norm": 6.167793273925781,
22
+ "learning_rate": 0.00016591928251121076,
23
+ "loss": 0.6265,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.8,
28
+ "grad_norm": 5.353821754455566,
29
+ "learning_rate": 0.00014798206278026908,
30
+ "loss": 0.5464,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "eval_accuracy": 0.8283,
36
+ "eval_loss": 0.46604594588279724,
37
+ "eval_model_preparation_time": 0.003,
38
+ "eval_runtime": 687.6086,
39
+ "eval_samples_per_second": 87.259,
40
+ "eval_steps_per_second": 10.907,
41
+ "step": 1875
42
+ },
43
+ {
44
+ "epoch": 1.0666666666666667,
45
+ "grad_norm": 3.131957530975342,
46
+ "learning_rate": 0.00013004484304932735,
47
+ "loss": 0.4901,
48
+ "step": 2000
49
+ },
50
+ {
51
+ "epoch": 1.3333333333333333,
52
+ "grad_norm": 3.037127733230591,
53
+ "learning_rate": 0.00011210762331838565,
54
+ "loss": 0.4272,
55
+ "step": 2500
56
+ },
57
+ {
58
+ "epoch": 1.6,
59
+ "grad_norm": 2.186342716217041,
60
+ "learning_rate": 9.417040358744396e-05,
61
+ "loss": 0.3883,
62
+ "step": 3000
63
+ },
64
+ {
65
+ "epoch": 1.8666666666666667,
66
+ "grad_norm": 5.017179489135742,
67
+ "learning_rate": 7.623318385650225e-05,
68
+ "loss": 0.3628,
69
+ "step": 3500
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "eval_accuracy": 0.8862833333333333,
74
+ "eval_loss": 0.31374436616897583,
75
+ "eval_model_preparation_time": 0.003,
76
+ "eval_runtime": 680.7322,
77
+ "eval_samples_per_second": 88.14,
78
+ "eval_steps_per_second": 11.018,
79
+ "step": 3750
80
+ }
81
+ ],
82
+ "logging_steps": 500,
83
+ "max_steps": 5625,
84
+ "num_input_tokens_seen": 0,
85
+ "num_train_epochs": 3,
86
+ "save_steps": 500,
87
+ "stateful_callbacks": {
88
+ "TrainerControl": {
89
+ "args": {
90
+ "should_epoch_stop": false,
91
+ "should_evaluate": false,
92
+ "should_log": false,
93
+ "should_save": true,
94
+ "should_training_stop": false
95
+ },
96
+ "attributes": {}
97
+ }
98
+ },
99
+ "total_flos": 1.005132625330176e+19,
100
+ "train_batch_size": 32,
101
+ "trial_name": null,
102
+ "trial_params": null
103
+ }
checkpoint-3750/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2634502d14d50ae6e309c6720abc81b68a2cb5d4fd494874448126eedae2d19
3
+ size 5304
checkpoint-5625/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "T-shirt / top",
7
+ "1": "Trouser",
8
+ "2": "Pullover",
9
+ "3": "Dress",
10
+ "4": "Coat",
11
+ "5": "Sandal",
12
+ "6": "Shirt",
13
+ "7": "Sneaker",
14
+ "8": "Bag",
15
+ "9": "Ankle boot"
16
+ },
17
+ "initializer_factor": 1.0,
18
+ "label2id": {
19
+ "Ankle boot": 9,
20
+ "Bag": 8,
21
+ "Coat": 4,
22
+ "Dress": 3,
23
+ "Pullover": 2,
24
+ "Sandal": 5,
25
+ "Shirt": 6,
26
+ "Sneaker": 7,
27
+ "T-shirt / top": 0,
28
+ "Trouser": 1
29
+ },
30
+ "model_type": "siglip",
31
+ "problem_type": "single_label_classification",
32
+ "text_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "intermediate_size": 3072,
37
+ "layer_norm_eps": 1e-06,
38
+ "max_position_embeddings": 64,
39
+ "model_type": "siglip_text_model",
40
+ "num_attention_heads": 12,
41
+ "num_hidden_layers": 12,
42
+ "projection_size": 768,
43
+ "torch_dtype": "float32",
44
+ "vocab_size": 256000
45
+ },
46
+ "torch_dtype": "float32",
47
+ "transformers_version": "4.50.0.dev0",
48
+ "vision_config": {
49
+ "attention_dropout": 0.0,
50
+ "hidden_act": "gelu_pytorch_tanh",
51
+ "hidden_size": 768,
52
+ "image_size": 224,
53
+ "intermediate_size": 3072,
54
+ "layer_norm_eps": 1e-06,
55
+ "model_type": "siglip_vision_model",
56
+ "num_attention_heads": 12,
57
+ "num_channels": 3,
58
+ "num_hidden_layers": 12,
59
+ "patch_size": 16,
60
+ "torch_dtype": "float32"
61
+ }
62
+ }
checkpoint-5625/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65b2eec77d2ba57a107ddbfa84686e14e84944fe4ab3058f042e39f412898a2e
3
+ size 371592608
checkpoint-5625/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a70bb8de9ab73b193b186bfa8f990374191fb02aae7a1d25cc44290c717801b
3
+ size 686604922
checkpoint-5625/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-5625/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05b6cad3062a5f01a32647505941328080beb3e6580efbe83538629396c48dde
3
+ size 14244
checkpoint-5625/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c636060e32831a2f047136ebca516c48b537dc20cc1ffb3fdc1b08dea5335af
3
+ size 1064
checkpoint-5625/trainer_state.json ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 5625,
3
+ "best_metric": 0.21983294188976288,
4
+ "best_model_checkpoint": "Fashion-Mnist-SigLIP2/checkpoint-5625",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 5625,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.26666666666666666,
14
+ "grad_norm": 9.494325637817383,
15
+ "learning_rate": 0.00018385650224215247,
16
+ "loss": 0.9674,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.5333333333333333,
21
+ "grad_norm": 6.167793273925781,
22
+ "learning_rate": 0.00016591928251121076,
23
+ "loss": 0.6265,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.8,
28
+ "grad_norm": 5.353821754455566,
29
+ "learning_rate": 0.00014798206278026908,
30
+ "loss": 0.5464,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "eval_accuracy": 0.8283,
36
+ "eval_loss": 0.46604594588279724,
37
+ "eval_model_preparation_time": 0.003,
38
+ "eval_runtime": 687.6086,
39
+ "eval_samples_per_second": 87.259,
40
+ "eval_steps_per_second": 10.907,
41
+ "step": 1875
42
+ },
43
+ {
44
+ "epoch": 1.0666666666666667,
45
+ "grad_norm": 3.131957530975342,
46
+ "learning_rate": 0.00013004484304932735,
47
+ "loss": 0.4901,
48
+ "step": 2000
49
+ },
50
+ {
51
+ "epoch": 1.3333333333333333,
52
+ "grad_norm": 3.037127733230591,
53
+ "learning_rate": 0.00011210762331838565,
54
+ "loss": 0.4272,
55
+ "step": 2500
56
+ },
57
+ {
58
+ "epoch": 1.6,
59
+ "grad_norm": 2.186342716217041,
60
+ "learning_rate": 9.417040358744396e-05,
61
+ "loss": 0.3883,
62
+ "step": 3000
63
+ },
64
+ {
65
+ "epoch": 1.8666666666666667,
66
+ "grad_norm": 5.017179489135742,
67
+ "learning_rate": 7.623318385650225e-05,
68
+ "loss": 0.3628,
69
+ "step": 3500
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "eval_accuracy": 0.8862833333333333,
74
+ "eval_loss": 0.31374436616897583,
75
+ "eval_model_preparation_time": 0.003,
76
+ "eval_runtime": 680.7322,
77
+ "eval_samples_per_second": 88.14,
78
+ "eval_steps_per_second": 11.018,
79
+ "step": 3750
80
+ },
81
+ {
82
+ "epoch": 2.1333333333333333,
83
+ "grad_norm": 3.4942686557769775,
84
+ "learning_rate": 5.8295964125560546e-05,
85
+ "loss": 0.3199,
86
+ "step": 4000
87
+ },
88
+ {
89
+ "epoch": 2.4,
90
+ "grad_norm": 6.02702522277832,
91
+ "learning_rate": 4.035874439461884e-05,
92
+ "loss": 0.2806,
93
+ "step": 4500
94
+ },
95
+ {
96
+ "epoch": 2.6666666666666665,
97
+ "grad_norm": 2.8055193424224854,
98
+ "learning_rate": 2.2421524663677132e-05,
99
+ "loss": 0.2633,
100
+ "step": 5000
101
+ },
102
+ {
103
+ "epoch": 2.9333333333333336,
104
+ "grad_norm": 3.3106279373168945,
105
+ "learning_rate": 4.484304932735426e-06,
106
+ "loss": 0.2339,
107
+ "step": 5500
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "eval_accuracy": 0.918,
112
+ "eval_loss": 0.21983294188976288,
113
+ "eval_model_preparation_time": 0.003,
114
+ "eval_runtime": 681.9131,
115
+ "eval_samples_per_second": 87.988,
116
+ "eval_steps_per_second": 10.998,
117
+ "step": 5625
118
+ }
119
+ ],
120
+ "logging_steps": 500,
121
+ "max_steps": 5625,
122
+ "num_input_tokens_seen": 0,
123
+ "num_train_epochs": 3,
124
+ "save_steps": 500,
125
+ "stateful_callbacks": {
126
+ "TrainerControl": {
127
+ "args": {
128
+ "should_epoch_stop": false,
129
+ "should_evaluate": false,
130
+ "should_log": false,
131
+ "should_save": true,
132
+ "should_training_stop": true
133
+ },
134
+ "attributes": {}
135
+ }
136
+ },
137
+ "total_flos": 1.507698937995264e+19,
138
+ "train_batch_size": 32,
139
+ "trial_name": null,
140
+ "trial_params": null
141
+ }
checkpoint-5625/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2634502d14d50ae6e309c6720abc81b68a2cb5d4fd494874448126eedae2d19
3
+ size 5304
config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "T-shirt / top",
7
+ "1": "Trouser",
8
+ "2": "Pullover",
9
+ "3": "Dress",
10
+ "4": "Coat",
11
+ "5": "Sandal",
12
+ "6": "Shirt",
13
+ "7": "Sneaker",
14
+ "8": "Bag",
15
+ "9": "Ankle boot"
16
+ },
17
+ "initializer_factor": 1.0,
18
+ "label2id": {
19
+ "Ankle boot": 9,
20
+ "Bag": 8,
21
+ "Coat": 4,
22
+ "Dress": 3,
23
+ "Pullover": 2,
24
+ "Sandal": 5,
25
+ "Shirt": 6,
26
+ "Sneaker": 7,
27
+ "T-shirt / top": 0,
28
+ "Trouser": 1
29
+ },
30
+ "model_type": "siglip",
31
+ "problem_type": "single_label_classification",
32
+ "text_config": {
33
+ "attention_dropout": 0.0,
34
+ "hidden_act": "gelu_pytorch_tanh",
35
+ "hidden_size": 768,
36
+ "intermediate_size": 3072,
37
+ "layer_norm_eps": 1e-06,
38
+ "max_position_embeddings": 64,
39
+ "model_type": "siglip_text_model",
40
+ "num_attention_heads": 12,
41
+ "num_hidden_layers": 12,
42
+ "projection_size": 768,
43
+ "torch_dtype": "float32",
44
+ "vocab_size": 256000
45
+ },
46
+ "torch_dtype": "float32",
47
+ "transformers_version": "4.50.0.dev0",
48
+ "vision_config": {
49
+ "attention_dropout": 0.0,
50
+ "hidden_act": "gelu_pytorch_tanh",
51
+ "hidden_size": 768,
52
+ "image_size": 224,
53
+ "intermediate_size": 3072,
54
+ "layer_norm_eps": 1e-06,
55
+ "model_type": "siglip_vision_model",
56
+ "num_attention_heads": 12,
57
+ "num_channels": 3,
58
+ "num_hidden_layers": 12,
59
+ "patch_size": 16,
60
+ "torch_dtype": "float32"
61
+ }
62
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65b2eec77d2ba57a107ddbfa84686e14e84944fe4ab3058f042e39f412898a2e
3
+ size 371592608
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:e2634502d14d50ae6e309c6720abc81b68a2cb5d4fd494874448126eedae2d19
3
+ size 5304