Hugging Face's logo

prithivMLmods
/
shoe-type-detection

Image Classification
Transformers
Safetensors
English
siglip
SigLIP2
Ballet Flat
Boat
Sneaker
Clog
Brogue
Model card Files
xet
Community
prithivMLmods commited on
Commit
3b12afe
·
verified ·
1 Parent(s): 921e029

Upload folder using huggingface_hub

Browse files
Files changed (36) hide show
  1. checkpoint-1252/config.json +52 -0
  2. checkpoint-1252/model.safetensors +3 -0
  3. checkpoint-1252/optimizer.pt +3 -0
  4. checkpoint-1252/preprocessor_config.json +24 -0
  5. checkpoint-1252/rng_state.pth +3 -0
  6. checkpoint-1252/scheduler.pt +3 -0
  7. checkpoint-1252/trainer_state.json +88 -0
  8. checkpoint-1252/training_args.bin +3 -0
  9. checkpoint-313/config.json +52 -0
  10. checkpoint-313/model.safetensors +3 -0
  11. checkpoint-313/optimizer.pt +3 -0
  12. checkpoint-313/preprocessor_config.json +24 -0
  13. checkpoint-313/rng_state.pth +3 -0
  14. checkpoint-313/scheduler.pt +3 -0
  15. checkpoint-313/trainer_state.json +44 -0
  16. checkpoint-313/training_args.bin +3 -0
  17. checkpoint-626/config.json +52 -0
  18. checkpoint-626/model.safetensors +3 -0
  19. checkpoint-626/optimizer.pt +3 -0
  20. checkpoint-626/preprocessor_config.json +24 -0
  21. checkpoint-626/rng_state.pth +3 -0
  22. checkpoint-626/scheduler.pt +3 -0
  23. checkpoint-626/trainer_state.json +61 -0
  24. checkpoint-626/training_args.bin +3 -0
  25. checkpoint-939/config.json +52 -0
  26. checkpoint-939/model.safetensors +3 -0
  27. checkpoint-939/optimizer.pt +3 -0
  28. checkpoint-939/preprocessor_config.json +24 -0
  29. checkpoint-939/rng_state.pth +3 -0
  30. checkpoint-939/scheduler.pt +3 -0
  31. checkpoint-939/trainer_state.json +71 -0
  32. checkpoint-939/training_args.bin +3 -0
  33. config.json +52 -0
  34. model.safetensors +3 -0
  35. preprocessor_config.json +24 -0
  36. training_args.bin +3 -0
checkpoint-1252/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Ballet Flat",
7
+ "1": "Boat",
8
+ "2": "Brogue",
9
+ "3": "Clog",
10
+ "4": "Sneaker"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Ballet Flat": 0,
15
+ "Boat": 1,
16
+ "Brogue": 2,
17
+ "Clog": 3,
18
+ "Sneaker": 4
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-1252/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed196cd97d8cd53631fe5cc3bef25d82df207b6a39be69a4c538de8e053a7f69
3
+ size 371577220
checkpoint-1252/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06064266012c16f73addc904d9075409a365e94564c7e4196fa5bf4adcc93273
3
+ size 686574202
checkpoint-1252/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-1252/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f5e9a9d3a4762802edd624d291d88431d1e6157ed174ab238b4cf8c0e5c29e0
3
+ size 14244
checkpoint-1252/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f01aabf88b7321d4217f3a4e020a66e6581ba7b92fe33dc4d135c7a17fa80ef
3
+ size 1064
checkpoint-1252/trainer_state.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1252,
3
+ "best_metric": 0.25282034277915955,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1252",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1252,
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
+ "eval_accuracy": 0.6746,
15
+ "eval_loss": 0.8624296188354492,
16
+ "eval_model_preparation_time": 0.0023,
17
+ "eval_runtime": 129.6436,
18
+ "eval_samples_per_second": 77.135,
19
+ "eval_steps_per_second": 9.642,
20
+ "step": 313
21
+ },
22
+ {
23
+ "epoch": 1.5974440894568689,
24
+ "grad_norm": 10.701898574829102,
25
+ "learning_rate": 0.00012512479201331117,
26
+ "loss": 1.0067,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.7695,
32
+ "eval_loss": 0.7205082774162292,
33
+ "eval_model_preparation_time": 0.0023,
34
+ "eval_runtime": 129.3402,
35
+ "eval_samples_per_second": 77.315,
36
+ "eval_steps_per_second": 9.664,
37
+ "step": 626
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "eval_accuracy": 0.8651,
42
+ "eval_loss": 0.3978084921836853,
43
+ "eval_model_preparation_time": 0.0023,
44
+ "eval_runtime": 129.5132,
45
+ "eval_samples_per_second": 77.212,
46
+ "eval_steps_per_second": 9.652,
47
+ "step": 939
48
+ },
49
+ {
50
+ "epoch": 3.194888178913738,
51
+ "grad_norm": 5.851517677307129,
52
+ "learning_rate": 4.193011647254576e-05,
53
+ "loss": 0.5959,
54
+ "step": 1000
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "eval_accuracy": 0.9197,
59
+ "eval_loss": 0.25282034277915955,
60
+ "eval_model_preparation_time": 0.0023,
61
+ "eval_runtime": 129.286,
62
+ "eval_samples_per_second": 77.348,
63
+ "eval_steps_per_second": 9.668,
64
+ "step": 1252
65
+ }
66
+ ],
67
+ "logging_steps": 500,
68
+ "max_steps": 1252,
69
+ "num_input_tokens_seen": 0,
70
+ "num_train_epochs": 4,
71
+ "save_steps": 500,
72
+ "stateful_callbacks": {
73
+ "TrainerControl": {
74
+ "args": {
75
+ "should_epoch_stop": false,
76
+ "should_evaluate": false,
77
+ "should_log": false,
78
+ "should_save": true,
79
+ "should_training_stop": true
80
+ },
81
+ "attributes": {}
82
+ }
83
+ },
84
+ "total_flos": 3.35030317719552e+18,
85
+ "train_batch_size": 32,
86
+ "trial_name": null,
87
+ "trial_params": null
88
+ }
checkpoint-1252/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:204006bafac16586eef5c520416ddc231ccdf5a76f9bffe88756655e3676d429
3
+ size 5304
checkpoint-313/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Ballet Flat",
7
+ "1": "Boat",
8
+ "2": "Brogue",
9
+ "3": "Clog",
10
+ "4": "Sneaker"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Ballet Flat": 0,
15
+ "Boat": 1,
16
+ "Brogue": 2,
17
+ "Clog": 3,
18
+ "Sneaker": 4
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-313/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f975fe12c3e2cf3930973728e5655c620b64208e4bfc6e3355b607f5bfd20d5b
3
+ size 371577220
checkpoint-313/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a50bf0b25861b9dfe7eef6d1bd7ec4e986e4f2ea409d5a15b428c0442a181170
3
+ size 686574202
checkpoint-313/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-313/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5c71b2757636618af226054c8fcdcecbc3433ba8a10b80225be943f6ab9d97d
3
+ size 14244
checkpoint-313/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e980d13442f3e9d68bbed3b5a14d900c79975c2b3ff48e0fb31503570a0866db
3
+ size 1064
checkpoint-313/trainer_state.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 313,
3
+ "best_metric": 0.8624296188354492,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-313",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 313,
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
+ "eval_accuracy": 0.6746,
15
+ "eval_loss": 0.8624296188354492,
16
+ "eval_model_preparation_time": 0.0023,
17
+ "eval_runtime": 129.6436,
18
+ "eval_samples_per_second": 77.135,
19
+ "eval_steps_per_second": 9.642,
20
+ "step": 313
21
+ }
22
+ ],
23
+ "logging_steps": 500,
24
+ "max_steps": 1252,
25
+ "num_input_tokens_seen": 0,
26
+ "num_train_epochs": 4,
27
+ "save_steps": 500,
28
+ "stateful_callbacks": {
29
+ "TrainerControl": {
30
+ "args": {
31
+ "should_epoch_stop": false,
32
+ "should_evaluate": false,
33
+ "should_log": false,
34
+ "should_save": true,
35
+ "should_training_stop": false
36
+ },
37
+ "attributes": {}
38
+ }
39
+ },
40
+ "total_flos": 8.3757579429888e+17,
41
+ "train_batch_size": 32,
42
+ "trial_name": null,
43
+ "trial_params": null
44
+ }
checkpoint-313/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:204006bafac16586eef5c520416ddc231ccdf5a76f9bffe88756655e3676d429
3
+ size 5304
checkpoint-626/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Ballet Flat",
7
+ "1": "Boat",
8
+ "2": "Brogue",
9
+ "3": "Clog",
10
+ "4": "Sneaker"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Ballet Flat": 0,
15
+ "Boat": 1,
16
+ "Brogue": 2,
17
+ "Clog": 3,
18
+ "Sneaker": 4
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-626/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9c44abafe318a7c16eba83074f14812f0676120aedc22b323cf3212b5fb4109
3
+ size 371577220
checkpoint-626/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4605312f9b4a9673df13f2a82d0bd38b70cae67cbc24c80c70621144475ae334
3
+ size 686574202
checkpoint-626/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-626/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af82d6208b626c8a56c5a50c572147ca8e3f620beaa54c6797f5e663add1c49b
3
+ size 14244
checkpoint-626/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de38ae1b28dbb9f0908f34f460d947c64b8e923e6b1adba7f73b288689453190
3
+ size 1064
checkpoint-626/trainer_state.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 626,
3
+ "best_metric": 0.7205082774162292,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-626",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 626,
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
+ "eval_accuracy": 0.6746,
15
+ "eval_loss": 0.8624296188354492,
16
+ "eval_model_preparation_time": 0.0023,
17
+ "eval_runtime": 129.6436,
18
+ "eval_samples_per_second": 77.135,
19
+ "eval_steps_per_second": 9.642,
20
+ "step": 313
21
+ },
22
+ {
23
+ "epoch": 1.5974440894568689,
24
+ "grad_norm": 10.701898574829102,
25
+ "learning_rate": 0.00012512479201331117,
26
+ "loss": 1.0067,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.7695,
32
+ "eval_loss": 0.7205082774162292,
33
+ "eval_model_preparation_time": 0.0023,
34
+ "eval_runtime": 129.3402,
35
+ "eval_samples_per_second": 77.315,
36
+ "eval_steps_per_second": 9.664,
37
+ "step": 626
38
+ }
39
+ ],
40
+ "logging_steps": 500,
41
+ "max_steps": 1252,
42
+ "num_input_tokens_seen": 0,
43
+ "num_train_epochs": 4,
44
+ "save_steps": 500,
45
+ "stateful_callbacks": {
46
+ "TrainerControl": {
47
+ "args": {
48
+ "should_epoch_stop": false,
49
+ "should_evaluate": false,
50
+ "should_log": false,
51
+ "should_save": true,
52
+ "should_training_stop": false
53
+ },
54
+ "attributes": {}
55
+ }
56
+ },
57
+ "total_flos": 1.67515158859776e+18,
58
+ "train_batch_size": 32,
59
+ "trial_name": null,
60
+ "trial_params": null
61
+ }
checkpoint-626/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:204006bafac16586eef5c520416ddc231ccdf5a76f9bffe88756655e3676d429
3
+ size 5304
checkpoint-939/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Ballet Flat",
7
+ "1": "Boat",
8
+ "2": "Brogue",
9
+ "3": "Clog",
10
+ "4": "Sneaker"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Ballet Flat": 0,
15
+ "Boat": 1,
16
+ "Brogue": 2,
17
+ "Clog": 3,
18
+ "Sneaker": 4
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-939/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df7aada0732d5d90900655e313188f7a7ce1325094aebb71f9e7fa6611a0cd52
3
+ size 371577220
checkpoint-939/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:839cf6cf60eddf310a607608c685e9479b6a3d2ec5b3fc90ff054957fd8ba815
3
+ size 686574202
checkpoint-939/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-939/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d05a72f74a1918a13db18f7ea2a2a5156b0bd2708979b9d71e8efaef28e5a5a
3
+ size 14244
checkpoint-939/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a037d2072775755e939865d8ebf7ced3fd8bef56cdcb5a8b6ddc1ac53d8157bf
3
+ size 1064
checkpoint-939/trainer_state.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 939,
3
+ "best_metric": 0.3978084921836853,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-939",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 939,
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
+ "eval_accuracy": 0.6746,
15
+ "eval_loss": 0.8624296188354492,
16
+ "eval_model_preparation_time": 0.0023,
17
+ "eval_runtime": 129.6436,
18
+ "eval_samples_per_second": 77.135,
19
+ "eval_steps_per_second": 9.642,
20
+ "step": 313
21
+ },
22
+ {
23
+ "epoch": 1.5974440894568689,
24
+ "grad_norm": 10.701898574829102,
25
+ "learning_rate": 0.00012512479201331117,
26
+ "loss": 1.0067,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.7695,
32
+ "eval_loss": 0.7205082774162292,
33
+ "eval_model_preparation_time": 0.0023,
34
+ "eval_runtime": 129.3402,
35
+ "eval_samples_per_second": 77.315,
36
+ "eval_steps_per_second": 9.664,
37
+ "step": 626
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "eval_accuracy": 0.8651,
42
+ "eval_loss": 0.3978084921836853,
43
+ "eval_model_preparation_time": 0.0023,
44
+ "eval_runtime": 129.5132,
45
+ "eval_samples_per_second": 77.212,
46
+ "eval_steps_per_second": 9.652,
47
+ "step": 939
48
+ }
49
+ ],
50
+ "logging_steps": 500,
51
+ "max_steps": 1252,
52
+ "num_input_tokens_seen": 0,
53
+ "num_train_epochs": 4,
54
+ "save_steps": 500,
55
+ "stateful_callbacks": {
56
+ "TrainerControl": {
57
+ "args": {
58
+ "should_epoch_stop": false,
59
+ "should_evaluate": false,
60
+ "should_log": false,
61
+ "should_save": true,
62
+ "should_training_stop": false
63
+ },
64
+ "attributes": {}
65
+ }
66
+ },
67
+ "total_flos": 2.51272738289664e+18,
68
+ "train_batch_size": 32,
69
+ "trial_name": null,
70
+ "trial_params": null
71
+ }
checkpoint-939/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:204006bafac16586eef5c520416ddc231ccdf5a76f9bffe88756655e3676d429
3
+ size 5304
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Ballet Flat",
7
+ "1": "Boat",
8
+ "2": "Brogue",
9
+ "3": "Clog",
10
+ "4": "Sneaker"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Ballet Flat": 0,
15
+ "Boat": 1,
16
+ "Brogue": 2,
17
+ "Clog": 3,
18
+ "Sneaker": 4
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:ed196cd97d8cd53631fe5cc3bef25d82df207b6a39be69a4c538de8e053a7f69
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 sha256:204006bafac16586eef5c520416ddc231ccdf5a76f9bffe88756655e3676d429
3
+ size 5304