prithivMLmods commited on
Commit
dfac7d6
·
verified ·
1 Parent(s): 8a7b2f5

Upload folder using huggingface_hub

Browse files
checkpoint-1070/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "basophil",
7
+ "1": "eosinophil",
8
+ "2": "erythroblast",
9
+ "3": "ig",
10
+ "4": "lymphocyte",
11
+ "5": "monocyte",
12
+ "6": "neutrophil",
13
+ "7": "platelet"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "basophil": 0,
18
+ "eosinophil": 1,
19
+ "erythroblast": 2,
20
+ "ig": 3,
21
+ "lymphocyte": 4,
22
+ "monocyte": 5,
23
+ "neutrophil": 6,
24
+ "platelet": 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": "4.51.0.dev0",
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-1070/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa9b9a9c8fb1917b5cda250e73ab1533177fdd6b58f2ec4db22b4011a3e61403
3
+ size 371586448
checkpoint-1070/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8704b773a8ac68324f0f3e07217a82f0c4cd14b15b7694ea9c525595a5ccdee
3
+ size 686592634
checkpoint-1070/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-1070/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81bcddb93fe6018ac3646487566c2ad45be4acd34d68b869ce631df20af5af07
3
+ size 14244
checkpoint-1070/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c61e1e62d815f95433f214394ccaa7815b6cc77e9008f7d41d497eb9ca08a458
3
+ size 1064
checkpoint-1070/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1070,
3
+ "best_metric": 0.04388454183936119,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1070",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1070,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.9345794392523364,
14
+ "grad_norm": 31.528762817382812,
15
+ "learning_rate": 1.569377990430622e-06,
16
+ "loss": 0.4718,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.9598642639831501,
22
+ "eval_loss": 0.11998829245567322,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 230.2816,
25
+ "eval_samples_per_second": 74.222,
26
+ "eval_steps_per_second": 9.28,
27
+ "step": 535
28
+ },
29
+ {
30
+ "epoch": 1.8691588785046729,
31
+ "grad_norm": 64.2021484375,
32
+ "learning_rate": 1.0909090909090908e-06,
33
+ "loss": 0.0791,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.9866604259302598,
39
+ "eval_loss": 0.04388454183936119,
40
+ "eval_model_preparation_time": 0.0023,
41
+ "eval_runtime": 230.6112,
42
+ "eval_samples_per_second": 74.116,
43
+ "eval_steps_per_second": 9.267,
44
+ "step": 1070
45
+ }
46
+ ],
47
+ "logging_steps": 500,
48
+ "max_steps": 2140,
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": 2.8632403213068534e+18,
65
+ "train_batch_size": 32,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
checkpoint-1070/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9ff2e89b955b9e4f83ca055f98105617928027ddb830251a26d76b910092a18
3
+ size 5304
checkpoint-1605/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "basophil",
7
+ "1": "eosinophil",
8
+ "2": "erythroblast",
9
+ "3": "ig",
10
+ "4": "lymphocyte",
11
+ "5": "monocyte",
12
+ "6": "neutrophil",
13
+ "7": "platelet"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "basophil": 0,
18
+ "eosinophil": 1,
19
+ "erythroblast": 2,
20
+ "ig": 3,
21
+ "lymphocyte": 4,
22
+ "monocyte": 5,
23
+ "neutrophil": 6,
24
+ "platelet": 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": "4.51.0.dev0",
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-1605/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33c37debdb92a1526fa0ff9281ef0ccd5abbad4a998c9296fbe790d11eb3155d
3
+ size 371586448
checkpoint-1605/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:924537f2c7e748546dd44f16d6962b5a4fedfe7afffa0671a33d3ea5e483e4f8
3
+ size 686592634
checkpoint-1605/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-1605/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d94df35f1d70785a743980724b84704b63c58e5985fc3ed4275777048cda1254
3
+ size 14244
checkpoint-1605/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f7439f18692f7a05cf6fe09d1ed825198bec89a15c27c70652b41b6fcaea030
3
+ size 1064
checkpoint-1605/trainer_state.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1605,
3
+ "best_metric": 0.037551190704107285,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1605",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1605,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.9345794392523364,
14
+ "grad_norm": 31.528762817382812,
15
+ "learning_rate": 1.569377990430622e-06,
16
+ "loss": 0.4718,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.9598642639831501,
22
+ "eval_loss": 0.11998829245567322,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 230.2816,
25
+ "eval_samples_per_second": 74.222,
26
+ "eval_steps_per_second": 9.28,
27
+ "step": 535
28
+ },
29
+ {
30
+ "epoch": 1.8691588785046729,
31
+ "grad_norm": 64.2021484375,
32
+ "learning_rate": 1.0909090909090908e-06,
33
+ "loss": 0.0791,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.9866604259302598,
39
+ "eval_loss": 0.04388454183936119,
40
+ "eval_model_preparation_time": 0.0023,
41
+ "eval_runtime": 230.6112,
42
+ "eval_samples_per_second": 74.116,
43
+ "eval_steps_per_second": 9.267,
44
+ "step": 1070
45
+ },
46
+ {
47
+ "epoch": 2.803738317757009,
48
+ "grad_norm": 25.20854377746582,
49
+ "learning_rate": 6.124401913875598e-07,
50
+ "loss": 0.0556,
51
+ "step": 1500
52
+ },
53
+ {
54
+ "epoch": 3.0,
55
+ "eval_accuracy": 0.9891762227942897,
56
+ "eval_loss": 0.037551190704107285,
57
+ "eval_model_preparation_time": 0.0023,
58
+ "eval_runtime": 229.0505,
59
+ "eval_samples_per_second": 74.621,
60
+ "eval_steps_per_second": 9.33,
61
+ "step": 1605
62
+ }
63
+ ],
64
+ "logging_steps": 500,
65
+ "max_steps": 2140,
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": 4.29486048196028e+18,
82
+ "train_batch_size": 32,
83
+ "trial_name": null,
84
+ "trial_params": null
85
+ }
checkpoint-1605/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9ff2e89b955b9e4f83ca055f98105617928027ddb830251a26d76b910092a18
3
+ size 5304
checkpoint-2140/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "basophil",
7
+ "1": "eosinophil",
8
+ "2": "erythroblast",
9
+ "3": "ig",
10
+ "4": "lymphocyte",
11
+ "5": "monocyte",
12
+ "6": "neutrophil",
13
+ "7": "platelet"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "basophil": 0,
18
+ "eosinophil": 1,
19
+ "erythroblast": 2,
20
+ "ig": 3,
21
+ "lymphocyte": 4,
22
+ "monocyte": 5,
23
+ "neutrophil": 6,
24
+ "platelet": 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": "4.51.0.dev0",
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-2140/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e71c95f170829815538bcb829f09cd9630caf7a0c66878baedd19d755bdf0fd
3
+ size 371586448
checkpoint-2140/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8257911556d2660ce0325e3421ae8701103075f35ebc2fab66ea6a2bf7ec6099
3
+ size 686592634
checkpoint-2140/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-2140/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e51a7a213df4c5d68312c2f6b444d67cc2d3c77333574691afd2fec386114483
3
+ size 14244
checkpoint-2140/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a60a6379ba53a957b110e691d3d01177a148545809d1de3f0fe47742d058a45a
3
+ size 1064
checkpoint-2140/trainer_state.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2140,
3
+ "best_metric": 0.036939047276973724,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-2140",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2140,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.9345794392523364,
14
+ "grad_norm": 31.528762817382812,
15
+ "learning_rate": 1.569377990430622e-06,
16
+ "loss": 0.4718,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.9598642639831501,
22
+ "eval_loss": 0.11998829245567322,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 230.2816,
25
+ "eval_samples_per_second": 74.222,
26
+ "eval_steps_per_second": 9.28,
27
+ "step": 535
28
+ },
29
+ {
30
+ "epoch": 1.8691588785046729,
31
+ "grad_norm": 64.2021484375,
32
+ "learning_rate": 1.0909090909090908e-06,
33
+ "loss": 0.0791,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_accuracy": 0.9866604259302598,
39
+ "eval_loss": 0.04388454183936119,
40
+ "eval_model_preparation_time": 0.0023,
41
+ "eval_runtime": 230.6112,
42
+ "eval_samples_per_second": 74.116,
43
+ "eval_steps_per_second": 9.267,
44
+ "step": 1070
45
+ },
46
+ {
47
+ "epoch": 2.803738317757009,
48
+ "grad_norm": 25.20854377746582,
49
+ "learning_rate": 6.124401913875598e-07,
50
+ "loss": 0.0556,
51
+ "step": 1500
52
+ },
53
+ {
54
+ "epoch": 3.0,
55
+ "eval_accuracy": 0.9891762227942897,
56
+ "eval_loss": 0.037551190704107285,
57
+ "eval_model_preparation_time": 0.0023,
58
+ "eval_runtime": 229.0505,
59
+ "eval_samples_per_second": 74.621,
60
+ "eval_steps_per_second": 9.33,
61
+ "step": 1605
62
+ },
63
+ {
64
+ "epoch": 3.7383177570093458,
65
+ "grad_norm": 4.499149799346924,
66
+ "learning_rate": 1.3397129186602872e-07,
67
+ "loss": 0.0462,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 4.0,
72
+ "eval_accuracy": 0.989117715890475,
73
+ "eval_loss": 0.036939047276973724,
74
+ "eval_model_preparation_time": 0.0023,
75
+ "eval_runtime": 228.5424,
76
+ "eval_samples_per_second": 74.787,
77
+ "eval_steps_per_second": 9.351,
78
+ "step": 2140
79
+ }
80
+ ],
81
+ "logging_steps": 500,
82
+ "max_steps": 2140,
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": 5.726480642613707e+18,
99
+ "train_batch_size": 32,
100
+ "trial_name": null,
101
+ "trial_params": null
102
+ }
checkpoint-2140/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9ff2e89b955b9e4f83ca055f98105617928027ddb830251a26d76b910092a18
3
+ size 5304
checkpoint-535/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "basophil",
7
+ "1": "eosinophil",
8
+ "2": "erythroblast",
9
+ "3": "ig",
10
+ "4": "lymphocyte",
11
+ "5": "monocyte",
12
+ "6": "neutrophil",
13
+ "7": "platelet"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "basophil": 0,
18
+ "eosinophil": 1,
19
+ "erythroblast": 2,
20
+ "ig": 3,
21
+ "lymphocyte": 4,
22
+ "monocyte": 5,
23
+ "neutrophil": 6,
24
+ "platelet": 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": "4.51.0.dev0",
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-535/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2ad74f8a336f1e58eff8c17d6d0ce4d2e45eb9514b888f5f50b4c92e333cc8c
3
+ size 371586448
checkpoint-535/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0daebcb0cbadb4cf1b5f2626e0a68778059e02bab767f1fe84ba0a3c9ce64016
3
+ size 686592634
checkpoint-535/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-535/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1b8fdedee5961b621965ce1c343807b0b29f952efbb2a869746b2c56d434332
3
+ size 14244
checkpoint-535/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2bf46e5db91dad6e575166b600f8e203aa97eed5abca817b4f77057899ee987
3
+ size 1064
checkpoint-535/trainer_state.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 535,
3
+ "best_metric": 0.11998829245567322,
4
+ "best_model_checkpoint": "siglip2-finetune-full/checkpoint-535",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 535,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.9345794392523364,
14
+ "grad_norm": 31.528762817382812,
15
+ "learning_rate": 1.569377990430622e-06,
16
+ "loss": 0.4718,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.9598642639831501,
22
+ "eval_loss": 0.11998829245567322,
23
+ "eval_model_preparation_time": 0.0023,
24
+ "eval_runtime": 230.2816,
25
+ "eval_samples_per_second": 74.222,
26
+ "eval_steps_per_second": 9.28,
27
+ "step": 535
28
+ }
29
+ ],
30
+ "logging_steps": 500,
31
+ "max_steps": 2140,
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": 1.4316201606534267e+18,
48
+ "train_batch_size": 32,
49
+ "trial_name": null,
50
+ "trial_params": null
51
+ }
checkpoint-535/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9ff2e89b955b9e4f83ca055f98105617928027ddb830251a26d76b910092a18
3
+ size 5304
config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SiglipForImageClassification"
4
+ ],
5
+ "id2label": {
6
+ "0": "basophil",
7
+ "1": "eosinophil",
8
+ "2": "erythroblast",
9
+ "3": "ig",
10
+ "4": "lymphocyte",
11
+ "5": "monocyte",
12
+ "6": "neutrophil",
13
+ "7": "platelet"
14
+ },
15
+ "initializer_factor": 1.0,
16
+ "label2id": {
17
+ "basophil": 0,
18
+ "eosinophil": 1,
19
+ "erythroblast": 2,
20
+ "ig": 3,
21
+ "lymphocyte": 4,
22
+ "monocyte": 5,
23
+ "neutrophil": 6,
24
+ "platelet": 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": "4.51.0.dev0",
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 sha256:6e71c95f170829815538bcb829f09cd9630caf7a0c66878baedd19d755bdf0fd
3
+ size 371586448
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:c9ff2e89b955b9e4f83ca055f98105617928027ddb830251a26d76b910092a18
3
+ size 5304