Hugging Face's logo

prithivMLmods
/
facial-age-detection

Image Classification
Transformers
Safetensors
English
siglip
age-detection
SigLIP2
biology
Model card Files
xet
Community
prithivMLmods commited on
Commit
87c252f
·
verified ·
1 Parent(s): d971608

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-1148/config.json +58 -0
  2. checkpoint-1148/model.safetensors +3 -0
  3. checkpoint-1148/optimizer.pt +3 -0
  4. checkpoint-1148/preprocessor_config.json +24 -0
  5. checkpoint-1148/rng_state.pth +3 -0
  6. checkpoint-1148/scheduler.pt +3 -0
  7. checkpoint-1148/trainer_state.json +88 -0
  8. checkpoint-1148/training_args.bin +3 -0
  9. checkpoint-1435/config.json +58 -0
  10. checkpoint-1435/model.safetensors +3 -0
  11. checkpoint-1435/optimizer.pt +3 -0
  12. checkpoint-1435/preprocessor_config.json +24 -0
  13. checkpoint-1435/rng_state.pth +3 -0
  14. checkpoint-1435/scheduler.pt +3 -0
  15. checkpoint-1435/trainer_state.json +98 -0
  16. checkpoint-1435/training_args.bin +3 -0
  17. checkpoint-1722/config.json +58 -0
  18. checkpoint-1722/model.safetensors +3 -0
  19. checkpoint-1722/optimizer.pt +3 -0
  20. checkpoint-1722/preprocessor_config.json +24 -0
  21. checkpoint-1722/rng_state.pth +3 -0
  22. checkpoint-1722/scheduler.pt +3 -0
  23. checkpoint-1722/trainer_state.json +115 -0
  24. checkpoint-1722/training_args.bin +3 -0
  25. checkpoint-287/config.json +58 -0
  26. checkpoint-287/model.safetensors +3 -0
  27. checkpoint-287/optimizer.pt +3 -0
  28. checkpoint-287/preprocessor_config.json +24 -0
  29. checkpoint-287/rng_state.pth +3 -0
  30. checkpoint-287/scheduler.pt +3 -0
  31. checkpoint-287/trainer_state.json +44 -0
  32. checkpoint-287/training_args.bin +3 -0
  33. checkpoint-574/config.json +58 -0
  34. checkpoint-574/model.safetensors +3 -0
  35. checkpoint-574/optimizer.pt +3 -0
  36. checkpoint-574/preprocessor_config.json +24 -0
  37. checkpoint-574/rng_state.pth +3 -0
  38. checkpoint-574/scheduler.pt +3 -0
  39. checkpoint-574/trainer_state.json +61 -0
  40. checkpoint-574/training_args.bin +3 -0
  41. checkpoint-861/config.json +58 -0
  42. checkpoint-861/model.safetensors +3 -0
  43. checkpoint-861/optimizer.pt +3 -0
  44. checkpoint-861/preprocessor_config.json +24 -0
  45. checkpoint-861/rng_state.pth +3 -0
  46. checkpoint-861/scheduler.pt +3 -0
  47. checkpoint-861/trainer_state.json +71 -0
  48. checkpoint-861/training_args.bin +3 -0
  49. config.json +58 -0
  50. model.safetensors +3 -0
checkpoint-1148/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
checkpoint-1148/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c91bad396704fe24816ee281e79890103b88ce05912af2f156526b6c09180c34
3
+ size 371586448
checkpoint-1148/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2509c41bb16bc61731b63da74c2a8e0bc8e15cea341c661146ec8fc610af4c53
3
+ size 686592634
checkpoint-1148/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-1148/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1e2ce22cd85e09a233fa80c476a2af0a599030daa001cacb597feaec8f25b19
3
+ size 14244
checkpoint-1148/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0f883ead62c8b6e814f3609b8fd803dce1d1f0b759c8956ef06ac0fbefc1f6a
3
+ size 1064
checkpoint-1148/trainer_state.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1148,
3
+ "best_metric": 0.552396297454834,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1148",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1148,
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": 0.646481178396072,
15
+ "eval_loss": 0.8569742441177368,
16
+ "eval_model_preparation_time": 0.0022,
17
+ "eval_runtime": 120.9539,
18
+ "eval_samples_per_second": 75.773,
19
+ "eval_steps_per_second": 9.475,
20
+ "step": 287
21
+ },
22
+ {
23
+ "epoch": 1.7421602787456445,
24
+ "grad_norm": 29.628196716308594,
25
+ "learning_rate": 1.4617224880382776e-05,
26
+ "loss": 1.0903,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.6763775231860338,
32
+ "eval_loss": 0.7689628005027771,
33
+ "eval_model_preparation_time": 0.0022,
34
+ "eval_runtime": 120.8745,
35
+ "eval_samples_per_second": 75.822,
36
+ "eval_steps_per_second": 9.481,
37
+ "step": 574
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "eval_accuracy": 0.7290780141843972,
42
+ "eval_loss": 0.6629450917243958,
43
+ "eval_model_preparation_time": 0.0022,
44
+ "eval_runtime": 121.8989,
45
+ "eval_samples_per_second": 75.185,
46
+ "eval_steps_per_second": 9.401,
47
+ "step": 861
48
+ },
49
+ {
50
+ "epoch": 3.484320557491289,
51
+ "grad_norm": 19.66444969177246,
52
+ "learning_rate": 8.636363636363637e-06,
53
+ "loss": 0.7613,
54
+ "step": 1000
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "eval_accuracy": 0.7742498636115658,
59
+ "eval_loss": 0.552396297454834,
60
+ "eval_model_preparation_time": 0.0022,
61
+ "eval_runtime": 121.0546,
62
+ "eval_samples_per_second": 75.71,
63
+ "eval_steps_per_second": 9.467,
64
+ "step": 1148
65
+ }
66
+ ],
67
+ "logging_steps": 500,
68
+ "max_steps": 1722,
69
+ "num_input_tokens_seen": 0,
70
+ "num_train_epochs": 6,
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": false
80
+ },
81
+ "attributes": {}
82
+ }
83
+ },
84
+ "total_flos": 3.07062924699009e+18,
85
+ "train_batch_size": 32,
86
+ "trial_name": null,
87
+ "trial_params": null
88
+ }
checkpoint-1148/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a497a9eb5e81ae5cb20a2ceefdd1321aa32c6b27b5c4bd8f2816c6cf571b8aed
3
+ size 5304
checkpoint-1435/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
checkpoint-1435/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60815197d8350ca8988443263b8dd4fd1e0639d78bcf974a34640a0d31338072
3
+ size 371586448
checkpoint-1435/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf242e6a4fbd6c9ed7bcf994732bdaa86cf58f5c1c794f17d57afa61bf35d859
3
+ size 686592634
checkpoint-1435/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-1435/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ff1c8e5adc3fe87431248921d26da697ec61f8556617d529523e88556b17fd8
3
+ size 14244
checkpoint-1435/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da2f0a1afe1cb2a189595ba315a20c9a2151481fdc442ba05e79e70f619b0625
3
+ size 1064
checkpoint-1435/trainer_state.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1435,
3
+ "best_metric": 0.4922109544277191,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1435",
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1435,
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": 0.646481178396072,
15
+ "eval_loss": 0.8569742441177368,
16
+ "eval_model_preparation_time": 0.0022,
17
+ "eval_runtime": 120.9539,
18
+ "eval_samples_per_second": 75.773,
19
+ "eval_steps_per_second": 9.475,
20
+ "step": 287
21
+ },
22
+ {
23
+ "epoch": 1.7421602787456445,
24
+ "grad_norm": 29.628196716308594,
25
+ "learning_rate": 1.4617224880382776e-05,
26
+ "loss": 1.0903,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.6763775231860338,
32
+ "eval_loss": 0.7689628005027771,
33
+ "eval_model_preparation_time": 0.0022,
34
+ "eval_runtime": 120.8745,
35
+ "eval_samples_per_second": 75.822,
36
+ "eval_steps_per_second": 9.481,
37
+ "step": 574
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "eval_accuracy": 0.7290780141843972,
42
+ "eval_loss": 0.6629450917243958,
43
+ "eval_model_preparation_time": 0.0022,
44
+ "eval_runtime": 121.8989,
45
+ "eval_samples_per_second": 75.185,
46
+ "eval_steps_per_second": 9.401,
47
+ "step": 861
48
+ },
49
+ {
50
+ "epoch": 3.484320557491289,
51
+ "grad_norm": 19.66444969177246,
52
+ "learning_rate": 8.636363636363637e-06,
53
+ "loss": 0.7613,
54
+ "step": 1000
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "eval_accuracy": 0.7742498636115658,
59
+ "eval_loss": 0.552396297454834,
60
+ "eval_model_preparation_time": 0.0022,
61
+ "eval_runtime": 121.0546,
62
+ "eval_samples_per_second": 75.71,
63
+ "eval_steps_per_second": 9.467,
64
+ "step": 1148
65
+ },
66
+ {
67
+ "epoch": 5.0,
68
+ "eval_accuracy": 0.8007637752318604,
69
+ "eval_loss": 0.4922109544277191,
70
+ "eval_model_preparation_time": 0.0022,
71
+ "eval_runtime": 121.4614,
72
+ "eval_samples_per_second": 75.456,
73
+ "eval_steps_per_second": 9.435,
74
+ "step": 1435
75
+ }
76
+ ],
77
+ "logging_steps": 500,
78
+ "max_steps": 1722,
79
+ "num_input_tokens_seen": 0,
80
+ "num_train_epochs": 6,
81
+ "save_steps": 500,
82
+ "stateful_callbacks": {
83
+ "TrainerControl": {
84
+ "args": {
85
+ "should_epoch_stop": false,
86
+ "should_evaluate": false,
87
+ "should_log": false,
88
+ "should_save": true,
89
+ "should_training_stop": false
90
+ },
91
+ "attributes": {}
92
+ }
93
+ },
94
+ "total_flos": 3.838286558737613e+18,
95
+ "train_batch_size": 32,
96
+ "trial_name": null,
97
+ "trial_params": null
98
+ }
checkpoint-1435/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a497a9eb5e81ae5cb20a2ceefdd1321aa32c6b27b5c4bd8f2816c6cf571b8aed
3
+ size 5304
checkpoint-1722/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
checkpoint-1722/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbe2101f827826eba65938ed4e574788e3bf3542a179b1d352c2f8b255452c68
3
+ size 371586448
checkpoint-1722/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b3164c6b6dc3361834336c5e8a99a84eee6b53526535444cf0c2972cb80621d
3
+ size 686592634
checkpoint-1722/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-1722/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f0d1da9544d8d5ff8def792568fe988e57addc62b4849048df7840c76f70476
3
+ size 14244
checkpoint-1722/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea0234fc2a4b91785bb19aa8cb52ba790ab2917af4d2023a186cdee4611c5571
3
+ size 1064
checkpoint-1722/trainer_state.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1722,
3
+ "best_metric": 0.44515112042427063,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1722",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1722,
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": 0.646481178396072,
15
+ "eval_loss": 0.8569742441177368,
16
+ "eval_model_preparation_time": 0.0022,
17
+ "eval_runtime": 120.9539,
18
+ "eval_samples_per_second": 75.773,
19
+ "eval_steps_per_second": 9.475,
20
+ "step": 287
21
+ },
22
+ {
23
+ "epoch": 1.7421602787456445,
24
+ "grad_norm": 29.628196716308594,
25
+ "learning_rate": 1.4617224880382776e-05,
26
+ "loss": 1.0903,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.6763775231860338,
32
+ "eval_loss": 0.7689628005027771,
33
+ "eval_model_preparation_time": 0.0022,
34
+ "eval_runtime": 120.8745,
35
+ "eval_samples_per_second": 75.822,
36
+ "eval_steps_per_second": 9.481,
37
+ "step": 574
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "eval_accuracy": 0.7290780141843972,
42
+ "eval_loss": 0.6629450917243958,
43
+ "eval_model_preparation_time": 0.0022,
44
+ "eval_runtime": 121.8989,
45
+ "eval_samples_per_second": 75.185,
46
+ "eval_steps_per_second": 9.401,
47
+ "step": 861
48
+ },
49
+ {
50
+ "epoch": 3.484320557491289,
51
+ "grad_norm": 19.66444969177246,
52
+ "learning_rate": 8.636363636363637e-06,
53
+ "loss": 0.7613,
54
+ "step": 1000
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "eval_accuracy": 0.7742498636115658,
59
+ "eval_loss": 0.552396297454834,
60
+ "eval_model_preparation_time": 0.0022,
61
+ "eval_runtime": 121.0546,
62
+ "eval_samples_per_second": 75.71,
63
+ "eval_steps_per_second": 9.467,
64
+ "step": 1148
65
+ },
66
+ {
67
+ "epoch": 5.0,
68
+ "eval_accuracy": 0.8007637752318604,
69
+ "eval_loss": 0.4922109544277191,
70
+ "eval_model_preparation_time": 0.0022,
71
+ "eval_runtime": 121.4614,
72
+ "eval_samples_per_second": 75.456,
73
+ "eval_steps_per_second": 9.435,
74
+ "step": 1435
75
+ },
76
+ {
77
+ "epoch": 5.2264808362369335,
78
+ "grad_norm": 19.047819137573242,
79
+ "learning_rate": 2.6555023923444976e-06,
80
+ "loss": 0.5947,
81
+ "step": 1500
82
+ },
83
+ {
84
+ "epoch": 6.0,
85
+ "eval_accuracy": 0.8224768139661757,
86
+ "eval_loss": 0.44515112042427063,
87
+ "eval_model_preparation_time": 0.0022,
88
+ "eval_runtime": 121.6668,
89
+ "eval_samples_per_second": 75.329,
90
+ "eval_steps_per_second": 9.419,
91
+ "step": 1722
92
+ }
93
+ ],
94
+ "logging_steps": 500,
95
+ "max_steps": 1722,
96
+ "num_input_tokens_seen": 0,
97
+ "num_train_epochs": 6,
98
+ "save_steps": 500,
99
+ "stateful_callbacks": {
100
+ "TrainerControl": {
101
+ "args": {
102
+ "should_epoch_stop": false,
103
+ "should_evaluate": false,
104
+ "should_log": false,
105
+ "should_save": true,
106
+ "should_training_stop": true
107
+ },
108
+ "attributes": {}
109
+ }
110
+ },
111
+ "total_flos": 4.6059438704851354e+18,
112
+ "train_batch_size": 32,
113
+ "trial_name": null,
114
+ "trial_params": null
115
+ }
checkpoint-1722/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a497a9eb5e81ae5cb20a2ceefdd1321aa32c6b27b5c4bd8f2816c6cf571b8aed
3
+ size 5304
checkpoint-287/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
checkpoint-287/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a0b35c5f7124fb8546a6c7ae1c7b6f6083f2d7e1e9c16d3a329713224394019
3
+ size 371586448
checkpoint-287/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ae3443c600a9856e1f0b8786275ac05174ed1e7ada730b00ea5fe1318ebd52f
3
+ size 686592634
checkpoint-287/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-287/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94d4167eb9695efadf5fae0f6733f49c4d2e0f8b027357dff8baf53ea791bcd3
3
+ size 14244
checkpoint-287/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a0f22126888db25140d8ca37d26bccded9119e4cfd672c9f9f6201bb8f6f1c5
3
+ size 1064
checkpoint-287/trainer_state.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 287,
3
+ "best_metric": 0.8569742441177368,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-287",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 287,
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": 0.646481178396072,
15
+ "eval_loss": 0.8569742441177368,
16
+ "eval_model_preparation_time": 0.0022,
17
+ "eval_runtime": 120.9539,
18
+ "eval_samples_per_second": 75.773,
19
+ "eval_steps_per_second": 9.475,
20
+ "step": 287
21
+ }
22
+ ],
23
+ "logging_steps": 500,
24
+ "max_steps": 1722,
25
+ "num_input_tokens_seen": 0,
26
+ "num_train_epochs": 6,
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": 7.676573117475226e+17,
41
+ "train_batch_size": 32,
42
+ "trial_name": null,
43
+ "trial_params": null
44
+ }
checkpoint-287/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a497a9eb5e81ae5cb20a2ceefdd1321aa32c6b27b5c4bd8f2816c6cf571b8aed
3
+ size 5304
checkpoint-574/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
checkpoint-574/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc0953884c77b7a37881be4dd541352f2cad1945464a43364873cc076e0474ee
3
+ size 371586448
checkpoint-574/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f62b0dfe1f1e8bcd7b731e9369c877643cac03453ab9ce228856819b7e3ada1
3
+ size 686592634
checkpoint-574/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-574/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf6fe4a70f9d3cd69383e5df6382774b34cffad85ee54931a76fb679e1b60fb1
3
+ size 14244
checkpoint-574/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8341dbc20e8d09dcd9180a96b9ad1abbf0b1b531833e452432c4e8c332dd8f7
3
+ size 1064
checkpoint-574/trainer_state.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 574,
3
+ "best_metric": 0.7689628005027771,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-574",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 574,
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": 0.646481178396072,
15
+ "eval_loss": 0.8569742441177368,
16
+ "eval_model_preparation_time": 0.0022,
17
+ "eval_runtime": 120.9539,
18
+ "eval_samples_per_second": 75.773,
19
+ "eval_steps_per_second": 9.475,
20
+ "step": 287
21
+ },
22
+ {
23
+ "epoch": 1.7421602787456445,
24
+ "grad_norm": 29.628196716308594,
25
+ "learning_rate": 1.4617224880382776e-05,
26
+ "loss": 1.0903,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.6763775231860338,
32
+ "eval_loss": 0.7689628005027771,
33
+ "eval_model_preparation_time": 0.0022,
34
+ "eval_runtime": 120.8745,
35
+ "eval_samples_per_second": 75.822,
36
+ "eval_steps_per_second": 9.481,
37
+ "step": 574
38
+ }
39
+ ],
40
+ "logging_steps": 500,
41
+ "max_steps": 1722,
42
+ "num_input_tokens_seen": 0,
43
+ "num_train_epochs": 6,
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": 1.535314623495045e+18,
58
+ "train_batch_size": 32,
59
+ "trial_name": null,
60
+ "trial_params": null
61
+ }
checkpoint-574/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a497a9eb5e81ae5cb20a2ceefdd1321aa32c6b27b5c4bd8f2816c6cf571b8aed
3
+ size 5304
checkpoint-861/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
checkpoint-861/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bed26da0481a45b099c675ad3c774564a30c28e5299a5e1d5102d95f29b8ca4f
3
+ size 371586448
checkpoint-861/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e397966ce338e6ad068b1dc8ce09b7e326a0238b9fb608004e68168720fd89b7
3
+ size 686592634
checkpoint-861/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-861/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a303624aee1ee47e399d7a0dc5ab60456d7cbea78de10f3b58c1e0fc1ac8f39b
3
+ size 14244
checkpoint-861/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0aa40d74352b0a1a5efcc83cd3091bb039d649e19ab7a1d651fd00c984213270
3
+ size 1064
checkpoint-861/trainer_state.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 861,
3
+ "best_metric": 0.6629450917243958,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-861",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 861,
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": 0.646481178396072,
15
+ "eval_loss": 0.8569742441177368,
16
+ "eval_model_preparation_time": 0.0022,
17
+ "eval_runtime": 120.9539,
18
+ "eval_samples_per_second": 75.773,
19
+ "eval_steps_per_second": 9.475,
20
+ "step": 287
21
+ },
22
+ {
23
+ "epoch": 1.7421602787456445,
24
+ "grad_norm": 29.628196716308594,
25
+ "learning_rate": 1.4617224880382776e-05,
26
+ "loss": 1.0903,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "eval_accuracy": 0.6763775231860338,
32
+ "eval_loss": 0.7689628005027771,
33
+ "eval_model_preparation_time": 0.0022,
34
+ "eval_runtime": 120.8745,
35
+ "eval_samples_per_second": 75.822,
36
+ "eval_steps_per_second": 9.481,
37
+ "step": 574
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "eval_accuracy": 0.7290780141843972,
42
+ "eval_loss": 0.6629450917243958,
43
+ "eval_model_preparation_time": 0.0022,
44
+ "eval_runtime": 121.8989,
45
+ "eval_samples_per_second": 75.185,
46
+ "eval_steps_per_second": 9.401,
47
+ "step": 861
48
+ }
49
+ ],
50
+ "logging_steps": 500,
51
+ "max_steps": 1722,
52
+ "num_input_tokens_seen": 0,
53
+ "num_train_epochs": 6,
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": 2.3029719352425677e+18,
68
+ "train_batch_size": 32,
69
+ "trial_name": null,
70
+ "trial_params": null
71
+ }
checkpoint-861/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a497a9eb5e81ae5cb20a2ceefdd1321aa32c6b27b5c4bd8f2816c6cf571b8aed
3
+ size 5304
config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "age 01-10",
7
+ "1": "age 11-20",
8
+ "2": "age 21-30",
9
+ "3": "age 31-40",
10
+ "4": "age 41-55",
11
+ "5": "age 56-65",
12
+ "6": "age 66-80",
13
+ "7": "age 80 +"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "age 01-10": 0,
18
+ "age 11-20": 1,
19
+ "age 21-30": 2,
20
+ "age 31-40": 3,
21
+ "age 41-55": 4,
22
+ "age 56-65": 5,
23
+ "age 66-80": 6,
24
+ "age 80 +": 7
25
+ },
26
+ "model_type": "siglip",
27
+ "problem_type": "single_label_classification",
28
+ "text_config": {
29
+ "attention_dropout": 0.0,
30
+ "hidden_act": "gelu_pytorch_tanh",
31
+ "hidden_size": 768,
32
+ "intermediate_size": 3072,
33
+ "layer_norm_eps": 1e-06,
34
+ "max_position_embeddings": 64,
35
+ "model_type": "siglip_text_model",
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 12,
38
+ "projection_size": 768,
39
+ "torch_dtype": "float32",
40
+ "vocab_size": 256000
41
+ },
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.50.0",
44
+ "vision_config": {
45
+ "attention_dropout": 0.0,
46
+ "hidden_act": "gelu_pytorch_tanh",
47
+ "hidden_size": 768,
48
+ "image_size": 224,
49
+ "intermediate_size": 3072,
50
+ "layer_norm_eps": 1e-06,
51
+ "model_type": "siglip_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "patch_size": 16,
56
+ "torch_dtype": "float32"
57
+ }
58
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbe2101f827826eba65938ed4e574788e3bf3542a179b1d352c2f8b255452c68
3
+ size 371586448