Hugging Face's logo

prithivMLmods
/
Age-Classification-SigLIP2

Image Classification
Transformers
Safetensors
English
siglip
Age
Detection
Siglip2
ViT
AutoImageProcessor
0-60+
Model card Files Community
prithivMLmods commited on
Commit
48a079e
·
verified ·
1 Parent(s): 48cd5fd

Upload folder using huggingface_hub

Browse files
Files changed (36) hide show
  1. checkpoint-1190/config.json +52 -0
  2. checkpoint-1190/model.safetensors +3 -0
  3. checkpoint-1190/optimizer.pt +3 -0
  4. checkpoint-1190/preprocessor_config.json +24 -0
  5. checkpoint-1190/rng_state.pth +3 -0
  6. checkpoint-1190/scheduler.pt +3 -0
  7. checkpoint-1190/trainer_state.json +68 -0
  8. checkpoint-1190/training_args.bin +3 -0
  9. checkpoint-1785/config.json +52 -0
  10. checkpoint-1785/model.safetensors +3 -0
  11. checkpoint-1785/optimizer.pt +3 -0
  12. checkpoint-1785/preprocessor_config.json +24 -0
  13. checkpoint-1785/rng_state.pth +3 -0
  14. checkpoint-1785/scheduler.pt +3 -0
  15. checkpoint-1785/trainer_state.json +85 -0
  16. checkpoint-1785/training_args.bin +3 -0
  17. checkpoint-2380/config.json +52 -0
  18. checkpoint-2380/model.safetensors +3 -0
  19. checkpoint-2380/optimizer.pt +3 -0
  20. checkpoint-2380/preprocessor_config.json +24 -0
  21. checkpoint-2380/rng_state.pth +3 -0
  22. checkpoint-2380/scheduler.pt +3 -0
  23. checkpoint-2380/trainer_state.json +102 -0
  24. checkpoint-2380/training_args.bin +3 -0
  25. checkpoint-595/config.json +52 -0
  26. checkpoint-595/model.safetensors +3 -0
  27. checkpoint-595/optimizer.pt +3 -0
  28. checkpoint-595/preprocessor_config.json +24 -0
  29. checkpoint-595/rng_state.pth +3 -0
  30. checkpoint-595/scheduler.pt +3 -0
  31. checkpoint-595/trainer_state.json +51 -0
  32. checkpoint-595/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-1190/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Child 0-12",
7
+ "1": "Teenager 13-20",
8
+ "2": "Adult 21-44",
9
+ "3": "Middle Age 45-64",
10
+ "4": "Aged 65+"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Adult 21-44": 2,
15
+ "Aged 65+": 4,
16
+ "Child 0-12": 0,
17
+ "Middle Age 45-64": 3,
18
+ "Teenager 13-20": 1
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-1190/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdbed4d0c56e63c0679efed6f6aa1bedc90e4e074f506db490aaf451dc37827f
3
+ size 371577220
checkpoint-1190/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adcf8263812f473893c324f99c1ab9cf08ba953f260687a010c886aeacd9fd3f
3
+ size 686574202
checkpoint-1190/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-1190/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2b8bb857b275788df3a4bfe6c58412c20f635cf16b03ce1fb0bccf24f56310b
3
+ size 14244
checkpoint-1190/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3a5cfa0c1d3b1032f6cf492006e0000c9bf6ffdfd4ddf3b7f0fe3b81372d8db
3
+ size 1064
checkpoint-1190/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1190,
3
+ "best_metric": 0.41642823815345764,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1190",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1190,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8403361344537815,
14
+ "grad_norm": 37.12611770629883,
15
+ "learning_rate": 4.034334763948498e-06,
16
+ "loss": 0.757,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8265671013883046,
22
+ "eval_loss": 0.44112637639045715,
23
+ "eval_model_preparation_time": 0.0026,
24
+ "eval_runtime": 308.0267,
25
+ "eval_samples_per_second": 61.735,
26
+ "eval_steps_per_second": 7.717,
27
+ "step": 595
28
+ },
29
+ {
30
+ "epoch": 1.680672268907563,
31
+ "grad_norm": 40.48983383178711,
32
+ "learning_rate": 2.96137339055794e-06,
33
+ "loss": 0.4737,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.8392406394615061,
39
+ "eval_loss": 0.41642823815345764,
40
+ "eval_model_preparation_time": 0.0026,
41
+ "eval_runtime": 305.306,
42
+ "eval_samples_per_second": 62.285,
43
+ "eval_steps_per_second": 7.786,
44
+ "step": 1190
45
+ }
46
+ ],
47
+ "logging_steps": 500,
48
+ "max_steps": 2380,
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": 3.1854682608775004e+18,
65
+ "train_batch_size": 32,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
checkpoint-1190/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8922fc74519c0ff839d6685cc5cbfe75096514003e98226b40754849ee63ae91
3
+ size 5304
checkpoint-1785/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Child 0-12",
7
+ "1": "Teenager 13-20",
8
+ "2": "Adult 21-44",
9
+ "3": "Middle Age 45-64",
10
+ "4": "Aged 65+"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Adult 21-44": 2,
15
+ "Aged 65+": 4,
16
+ "Child 0-12": 0,
17
+ "Middle Age 45-64": 3,
18
+ "Teenager 13-20": 1
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-1785/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7998d9c8e595a15dd57728ad944ec307473bb570693c42b981c36a63e74428c5
3
+ size 371577220
checkpoint-1785/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9e9be3d71ecc0f448cd0e4829ccb2a7c028ce42da5af60d5c0590fd0d2c586a
3
+ size 686574202
checkpoint-1785/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-1785/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35bad94d3860ee97520c76c204fc73a3a0ea97862ce78616f228143e0990d2e0
3
+ size 14244
checkpoint-1785/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df87a38d0129fab3acc45cf2fd95d5002dfb342ad943acf1aaa3f056aa94245e
3
+ size 1064
checkpoint-1785/trainer_state.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1785,
3
+ "best_metric": 0.2768375873565674,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1785",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1785,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8403361344537815,
14
+ "grad_norm": 37.12611770629883,
15
+ "learning_rate": 4.034334763948498e-06,
16
+ "loss": 0.757,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8265671013883046,
22
+ "eval_loss": 0.44112637639045715,
23
+ "eval_model_preparation_time": 0.0026,
24
+ "eval_runtime": 308.0267,
25
+ "eval_samples_per_second": 61.735,
26
+ "eval_steps_per_second": 7.717,
27
+ "step": 595
28
+ },
29
+ {
30
+ "epoch": 1.680672268907563,
31
+ "grad_norm": 40.48983383178711,
32
+ "learning_rate": 2.96137339055794e-06,
33
+ "loss": 0.4737,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.8392406394615061,
39
+ "eval_loss": 0.41642823815345764,
40
+ "eval_model_preparation_time": 0.0026,
41
+ "eval_runtime": 305.306,
42
+ "eval_samples_per_second": 62.285,
43
+ "eval_steps_per_second": 7.786,
44
+ "step": 1190
45
+ },
46
+ {
47
+ "epoch": 2.5210084033613445,
48
+ "grad_norm": 30.135860443115234,
49
+ "learning_rate": 1.8884120171673823e-06,
50
+ "loss": 0.3811,
51
+ "step": 1500
52
+ },
53
+ {
54
+ "epoch": 3.0,
55
+ "eval_accuracy": 0.889724442574674,
56
+ "eval_loss": 0.2768375873565674,
57
+ "eval_model_preparation_time": 0.0026,
58
+ "eval_runtime": 304.3734,
59
+ "eval_samples_per_second": 62.476,
60
+ "eval_steps_per_second": 7.809,
61
+ "step": 1785
62
+ }
63
+ ],
64
+ "logging_steps": 500,
65
+ "max_steps": 2380,
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": 4.778202391316251e+18,
82
+ "train_batch_size": 32,
83
+ "trial_name": null,
84
+ "trial_params": null
85
+ }
checkpoint-1785/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8922fc74519c0ff839d6685cc5cbfe75096514003e98226b40754849ee63ae91
3
+ size 5304
checkpoint-2380/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Child 0-12",
7
+ "1": "Teenager 13-20",
8
+ "2": "Adult 21-44",
9
+ "3": "Middle Age 45-64",
10
+ "4": "Aged 65+"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Adult 21-44": 2,
15
+ "Aged 65+": 4,
16
+ "Child 0-12": 0,
17
+ "Middle Age 45-64": 3,
18
+ "Teenager 13-20": 1
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-2380/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e0205ae229b38817996d56b5106c6843cfa6c93eb460b43e85f8a8d2fabf90a
3
+ size 371577220
checkpoint-2380/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d90a7de971b8d47a71322bdc35890b432b704217166b29da632b5b8874306ed
3
+ size 686574202
checkpoint-2380/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-2380/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da985cf6e4426c2b38657d65b035877c75e77b8c965bdcd45c9f6896bcf76cc5
3
+ size 14244
checkpoint-2380/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06cfbe3a1e268b684dbc619fed22806f8dfef6cf5a3eb4c171206ac697a9ebb9
3
+ size 1064
checkpoint-2380/trainer_state.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2380,
3
+ "best_metric": 0.23066656291484833,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-2380",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2380,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8403361344537815,
14
+ "grad_norm": 37.12611770629883,
15
+ "learning_rate": 4.034334763948498e-06,
16
+ "loss": 0.757,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8265671013883046,
22
+ "eval_loss": 0.44112637639045715,
23
+ "eval_model_preparation_time": 0.0026,
24
+ "eval_runtime": 308.0267,
25
+ "eval_samples_per_second": 61.735,
26
+ "eval_steps_per_second": 7.717,
27
+ "step": 595
28
+ },
29
+ {
30
+ "epoch": 1.680672268907563,
31
+ "grad_norm": 40.48983383178711,
32
+ "learning_rate": 2.96137339055794e-06,
33
+ "loss": 0.4737,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.8392406394615061,
39
+ "eval_loss": 0.41642823815345764,
40
+ "eval_model_preparation_time": 0.0026,
41
+ "eval_runtime": 305.306,
42
+ "eval_samples_per_second": 62.285,
43
+ "eval_steps_per_second": 7.786,
44
+ "step": 1190
45
+ },
46
+ {
47
+ "epoch": 2.5210084033613445,
48
+ "grad_norm": 30.135860443115234,
49
+ "learning_rate": 1.8884120171673823e-06,
50
+ "loss": 0.3811,
51
+ "step": 1500
52
+ },
53
+ {
54
+ "epoch": 3.0,
55
+ "eval_accuracy": 0.889724442574674,
56
+ "eval_loss": 0.2768375873565674,
57
+ "eval_model_preparation_time": 0.0026,
58
+ "eval_runtime": 304.3734,
59
+ "eval_samples_per_second": 62.476,
60
+ "eval_steps_per_second": 7.809,
61
+ "step": 1785
62
+ },
63
+ {
64
+ "epoch": 3.361344537815126,
65
+ "grad_norm": 35.56352233886719,
66
+ "learning_rate": 8.15450643776824e-07,
67
+ "loss": 0.2957,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 4.0,
72
+ "eval_accuracy": 0.9109171224232225,
73
+ "eval_loss": 0.23066656291484833,
74
+ "eval_model_preparation_time": 0.0026,
75
+ "eval_runtime": 300.2689,
76
+ "eval_samples_per_second": 63.33,
77
+ "eval_steps_per_second": 7.916,
78
+ "step": 2380
79
+ }
80
+ ],
81
+ "logging_steps": 500,
82
+ "max_steps": 2380,
83
+ "num_input_tokens_seen": 0,
84
+ "num_train_epochs": 4,
85
+ "save_steps": 500,
86
+ "stateful_callbacks": {
87
+ "TrainerControl": {
88
+ "args": {
89
+ "should_epoch_stop": false,
90
+ "should_evaluate": false,
91
+ "should_log": false,
92
+ "should_save": true,
93
+ "should_training_stop": true
94
+ },
95
+ "attributes": {}
96
+ }
97
+ },
98
+ "total_flos": 6.370936521755001e+18,
99
+ "train_batch_size": 32,
100
+ "trial_name": null,
101
+ "trial_params": null
102
+ }
checkpoint-2380/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8922fc74519c0ff839d6685cc5cbfe75096514003e98226b40754849ee63ae91
3
+ size 5304
checkpoint-595/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Child 0-12",
7
+ "1": "Teenager 13-20",
8
+ "2": "Adult 21-44",
9
+ "3": "Middle Age 45-64",
10
+ "4": "Aged 65+"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Adult 21-44": 2,
15
+ "Aged 65+": 4,
16
+ "Child 0-12": 0,
17
+ "Middle Age 45-64": 3,
18
+ "Teenager 13-20": 1
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-595/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb0709850f1a01f281d83881866f671cbfea400e90d9610dedda4f141583d4d5
3
+ size 371577220
checkpoint-595/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:358610b3e71c07f2a32fce49859c1de8ce3a74caa8c383439a52b224f0ff23e8
3
+ size 686574202
checkpoint-595/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-595/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5181ecb6c4a34a3fa80f328ab2cae756bf09870289fe50718f33f41161538f4d
3
+ size 14244
checkpoint-595/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:191911990f5710fba30b6855987490935f90910ddb2c7f431527798eefd3d96a
3
+ size 1064
checkpoint-595/trainer_state.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 595,
3
+ "best_metric": 0.44112637639045715,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-595",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 595,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.8403361344537815,
14
+ "grad_norm": 37.12611770629883,
15
+ "learning_rate": 4.034334763948498e-06,
16
+ "loss": 0.757,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8265671013883046,
22
+ "eval_loss": 0.44112637639045715,
23
+ "eval_model_preparation_time": 0.0026,
24
+ "eval_runtime": 308.0267,
25
+ "eval_samples_per_second": 61.735,
26
+ "eval_steps_per_second": 7.717,
27
+ "step": 595
28
+ }
29
+ ],
30
+ "logging_steps": 500,
31
+ "max_steps": 2380,
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": 1.5927341304387502e+18,
48
+ "train_batch_size": 32,
49
+ "trial_name": null,
50
+ "trial_params": null
51
+ }
checkpoint-595/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8922fc74519c0ff839d6685cc5cbfe75096514003e98226b40754849ee63ae91
3
+ size 5304
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "Child 0-12",
7
+ "1": "Teenager 13-20",
8
+ "2": "Adult 21-44",
9
+ "3": "Middle Age 45-64",
10
+ "4": "Aged 65+"
11
+ },
12
+ "initializer_factor": 1.0,
13
+ "label2id": {
14
+ "Adult 21-44": 2,
15
+ "Aged 65+": 4,
16
+ "Child 0-12": 0,
17
+ "Middle Age 45-64": 3,
18
+ "Teenager 13-20": 1
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 sha256:7e0205ae229b38817996d56b5106c6843cfa6c93eb460b43e85f8a8d2fabf90a
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 sha256:8922fc74519c0ff839d6685cc5cbfe75096514003e98226b40754849ee63ae91
3
+ size 5304