Upload folder using huggingface_hub
Browse files- checkpoint-1378/config.json +50 -0
- checkpoint-1378/model.safetensors +3 -0
- checkpoint-1378/optimizer.pt +3 -0
- checkpoint-1378/preprocessor_config.json +24 -0
- checkpoint-1378/rng_state.pth +3 -0
- checkpoint-1378/scheduler.pt +3 -0
- checkpoint-1378/trainer_state.json +58 -0
- checkpoint-1378/training_args.bin +3 -0
- checkpoint-2756/config.json +50 -0
- checkpoint-2756/model.safetensors +3 -0
- checkpoint-2756/optimizer.pt +3 -0
- checkpoint-2756/preprocessor_config.json +24 -0
- checkpoint-2756/rng_state.pth +3 -0
- checkpoint-2756/scheduler.pt +3 -0
- checkpoint-2756/trainer_state.json +89 -0
- checkpoint-2756/training_args.bin +3 -0
- checkpoint-4134/config.json +50 -0
- checkpoint-4134/model.safetensors +3 -0
- checkpoint-4134/optimizer.pt +3 -0
- checkpoint-4134/preprocessor_config.json +24 -0
- checkpoint-4134/rng_state.pth +3 -0
- checkpoint-4134/scheduler.pt +3 -0
- checkpoint-4134/trainer_state.json +120 -0
- checkpoint-4134/training_args.bin +3 -0
- config.json +50 -0
- model.safetensors +3 -0
- preprocessor_config.json +24 -0
- training_args.bin +3 -0
checkpoint-1378/config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SiglipForImageClassification"
|
4 |
+
],
|
5 |
+
"id2label": {
|
6 |
+
"0": "Fall",
|
7 |
+
"1": "Spring",
|
8 |
+
"2": "Summer",
|
9 |
+
"3": "Winter"
|
10 |
+
},
|
11 |
+
"initializer_factor": 1.0,
|
12 |
+
"label2id": {
|
13 |
+
"Fall": 0,
|
14 |
+
"Spring": 1,
|
15 |
+
"Summer": 2,
|
16 |
+
"Winter": 3
|
17 |
+
},
|
18 |
+
"model_type": "siglip",
|
19 |
+
"problem_type": "single_label_classification",
|
20 |
+
"text_config": {
|
21 |
+
"attention_dropout": 0.0,
|
22 |
+
"hidden_act": "gelu_pytorch_tanh",
|
23 |
+
"hidden_size": 768,
|
24 |
+
"intermediate_size": 3072,
|
25 |
+
"layer_norm_eps": 1e-06,
|
26 |
+
"max_position_embeddings": 64,
|
27 |
+
"model_type": "siglip_text_model",
|
28 |
+
"num_attention_heads": 12,
|
29 |
+
"num_hidden_layers": 12,
|
30 |
+
"projection_size": 768,
|
31 |
+
"torch_dtype": "float32",
|
32 |
+
"vocab_size": 256000
|
33 |
+
},
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.50.3",
|
36 |
+
"vision_config": {
|
37 |
+
"attention_dropout": 0.0,
|
38 |
+
"hidden_act": "gelu_pytorch_tanh",
|
39 |
+
"hidden_size": 768,
|
40 |
+
"image_size": 224,
|
41 |
+
"intermediate_size": 3072,
|
42 |
+
"layer_norm_eps": 1e-06,
|
43 |
+
"model_type": "siglip_vision_model",
|
44 |
+
"num_attention_heads": 12,
|
45 |
+
"num_channels": 3,
|
46 |
+
"num_hidden_layers": 12,
|
47 |
+
"patch_size": 16,
|
48 |
+
"torch_dtype": "float32"
|
49 |
+
}
|
50 |
+
}
|
checkpoint-1378/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b1e487dc9e1fc1beb957782f9898f2cee610bbc4682548f267575bf582c4dfdc
|
3 |
+
size 371574144
|
checkpoint-1378/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:081292f13d33bec035d678e1c726c00ec1b843acacb1a40905a4fba8c2de804d
|
3 |
+
size 686568058
|
checkpoint-1378/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-1378/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:be3800323adfbabbfee073f7c32c3d86bc0c9fd7937c41a132afa461c6c74cbe
|
3 |
+
size 14244
|
checkpoint-1378/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c7ebc3161ea90f601447bf199c462c7c459338b91b880eb89a89451f474a0e64
|
3 |
+
size 1064
|
checkpoint-1378/trainer_state.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_global_step": 1378,
|
3 |
+
"best_metric": 0.7454163432121277,
|
4 |
+
"best_model_checkpoint": "siglip2-finetune-full/checkpoint-1378",
|
5 |
+
"epoch": 1.0,
|
6 |
+
"eval_steps": 500,
|
7 |
+
"global_step": 1378,
|
8 |
+
"is_hyper_param_search": false,
|
9 |
+
"is_local_process_zero": true,
|
10 |
+
"is_world_process_zero": true,
|
11 |
+
"log_history": [
|
12 |
+
{
|
13 |
+
"epoch": 0.36284470246734396,
|
14 |
+
"grad_norm": 22.187894821166992,
|
15 |
+
"learning_rate": 8.898139079333986e-07,
|
16 |
+
"loss": 0.9313,
|
17 |
+
"step": 500
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.7256894049346879,
|
21 |
+
"grad_norm": 24.758203506469727,
|
22 |
+
"learning_rate": 7.67384916748286e-07,
|
23 |
+
"loss": 0.7624,
|
24 |
+
"step": 1000
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 1.0,
|
28 |
+
"eval_accuracy": 0.6740333998910873,
|
29 |
+
"eval_loss": 0.7454163432121277,
|
30 |
+
"eval_model_preparation_time": 0.0023,
|
31 |
+
"eval_runtime": 535.389,
|
32 |
+
"eval_samples_per_second": 82.318,
|
33 |
+
"eval_steps_per_second": 10.29,
|
34 |
+
"step": 1378
|
35 |
+
}
|
36 |
+
],
|
37 |
+
"logging_steps": 500,
|
38 |
+
"max_steps": 4134,
|
39 |
+
"num_input_tokens_seen": 0,
|
40 |
+
"num_train_epochs": 3,
|
41 |
+
"save_steps": 500,
|
42 |
+
"stateful_callbacks": {
|
43 |
+
"TrainerControl": {
|
44 |
+
"args": {
|
45 |
+
"should_epoch_stop": false,
|
46 |
+
"should_evaluate": false,
|
47 |
+
"should_log": false,
|
48 |
+
"should_save": true,
|
49 |
+
"should_training_stop": false
|
50 |
+
},
|
51 |
+
"attributes": {}
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"total_flos": 3.69133343103497e+18,
|
55 |
+
"train_batch_size": 32,
|
56 |
+
"trial_name": null,
|
57 |
+
"trial_params": null
|
58 |
+
}
|
checkpoint-1378/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:57fdf6fbaad0fbc617e51d43e3e727a2cb5eff0f72c5fb35af4caa4e562ea3de
|
3 |
+
size 5304
|
checkpoint-2756/config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SiglipForImageClassification"
|
4 |
+
],
|
5 |
+
"id2label": {
|
6 |
+
"0": "Fall",
|
7 |
+
"1": "Spring",
|
8 |
+
"2": "Summer",
|
9 |
+
"3": "Winter"
|
10 |
+
},
|
11 |
+
"initializer_factor": 1.0,
|
12 |
+
"label2id": {
|
13 |
+
"Fall": 0,
|
14 |
+
"Spring": 1,
|
15 |
+
"Summer": 2,
|
16 |
+
"Winter": 3
|
17 |
+
},
|
18 |
+
"model_type": "siglip",
|
19 |
+
"problem_type": "single_label_classification",
|
20 |
+
"text_config": {
|
21 |
+
"attention_dropout": 0.0,
|
22 |
+
"hidden_act": "gelu_pytorch_tanh",
|
23 |
+
"hidden_size": 768,
|
24 |
+
"intermediate_size": 3072,
|
25 |
+
"layer_norm_eps": 1e-06,
|
26 |
+
"max_position_embeddings": 64,
|
27 |
+
"model_type": "siglip_text_model",
|
28 |
+
"num_attention_heads": 12,
|
29 |
+
"num_hidden_layers": 12,
|
30 |
+
"projection_size": 768,
|
31 |
+
"torch_dtype": "float32",
|
32 |
+
"vocab_size": 256000
|
33 |
+
},
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.50.3",
|
36 |
+
"vision_config": {
|
37 |
+
"attention_dropout": 0.0,
|
38 |
+
"hidden_act": "gelu_pytorch_tanh",
|
39 |
+
"hidden_size": 768,
|
40 |
+
"image_size": 224,
|
41 |
+
"intermediate_size": 3072,
|
42 |
+
"layer_norm_eps": 1e-06,
|
43 |
+
"model_type": "siglip_vision_model",
|
44 |
+
"num_attention_heads": 12,
|
45 |
+
"num_channels": 3,
|
46 |
+
"num_hidden_layers": 12,
|
47 |
+
"patch_size": 16,
|
48 |
+
"torch_dtype": "float32"
|
49 |
+
}
|
50 |
+
}
|
checkpoint-2756/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:04c0f19caa857fcc7208966ded57d4d707ea0aef7e0eaf5c77fa8eb5df9c4968
|
3 |
+
size 371574144
|
checkpoint-2756/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02012cede08c264bc7d9e1438555893b58cf9f3d84fd54da6acc8613f1c2f8d8
|
3 |
+
size 686568058
|
checkpoint-2756/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-2756/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f6895266ce7775b656647ddf97f9b07ff3ff78add76255d92106d9353ec64e09
|
3 |
+
size 14244
|
checkpoint-2756/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d3f70973b0be83f2cc696ea78084c8887bb441571a706de188d5249acfa60190
|
3 |
+
size 1064
|
checkpoint-2756/trainer_state.json
ADDED
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_global_step": 2756,
|
3 |
+
"best_metric": 0.6948277354240417,
|
4 |
+
"best_model_checkpoint": "siglip2-finetune-full/checkpoint-2756",
|
5 |
+
"epoch": 2.0,
|
6 |
+
"eval_steps": 500,
|
7 |
+
"global_step": 2756,
|
8 |
+
"is_hyper_param_search": false,
|
9 |
+
"is_local_process_zero": true,
|
10 |
+
"is_world_process_zero": true,
|
11 |
+
"log_history": [
|
12 |
+
{
|
13 |
+
"epoch": 0.36284470246734396,
|
14 |
+
"grad_norm": 22.187894821166992,
|
15 |
+
"learning_rate": 8.898139079333986e-07,
|
16 |
+
"loss": 0.9313,
|
17 |
+
"step": 500
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.7256894049346879,
|
21 |
+
"grad_norm": 24.758203506469727,
|
22 |
+
"learning_rate": 7.67384916748286e-07,
|
23 |
+
"loss": 0.7624,
|
24 |
+
"step": 1000
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 1.0,
|
28 |
+
"eval_accuracy": 0.6740333998910873,
|
29 |
+
"eval_loss": 0.7454163432121277,
|
30 |
+
"eval_model_preparation_time": 0.0023,
|
31 |
+
"eval_runtime": 535.389,
|
32 |
+
"eval_samples_per_second": 82.318,
|
33 |
+
"eval_steps_per_second": 10.29,
|
34 |
+
"step": 1378
|
35 |
+
},
|
36 |
+
{
|
37 |
+
"epoch": 1.0885341074020318,
|
38 |
+
"grad_norm": 25.052143096923828,
|
39 |
+
"learning_rate": 6.449559255631733e-07,
|
40 |
+
"loss": 0.7082,
|
41 |
+
"step": 1500
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 1.4513788098693758,
|
45 |
+
"grad_norm": 26.29668617248535,
|
46 |
+
"learning_rate": 5.225269343780607e-07,
|
47 |
+
"loss": 0.6842,
|
48 |
+
"step": 2000
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.8142235123367199,
|
52 |
+
"grad_norm": 33.645015716552734,
|
53 |
+
"learning_rate": 4.000979431929481e-07,
|
54 |
+
"loss": 0.6637,
|
55 |
+
"step": 2500
|
56 |
+
},
|
57 |
+
{
|
58 |
+
"epoch": 2.0,
|
59 |
+
"eval_accuracy": 0.7012842621165366,
|
60 |
+
"eval_loss": 0.6948277354240417,
|
61 |
+
"eval_model_preparation_time": 0.0023,
|
62 |
+
"eval_runtime": 531.6957,
|
63 |
+
"eval_samples_per_second": 82.89,
|
64 |
+
"eval_steps_per_second": 10.361,
|
65 |
+
"step": 2756
|
66 |
+
}
|
67 |
+
],
|
68 |
+
"logging_steps": 500,
|
69 |
+
"max_steps": 4134,
|
70 |
+
"num_input_tokens_seen": 0,
|
71 |
+
"num_train_epochs": 3,
|
72 |
+
"save_steps": 500,
|
73 |
+
"stateful_callbacks": {
|
74 |
+
"TrainerControl": {
|
75 |
+
"args": {
|
76 |
+
"should_epoch_stop": false,
|
77 |
+
"should_evaluate": false,
|
78 |
+
"should_log": false,
|
79 |
+
"should_save": true,
|
80 |
+
"should_training_stop": false
|
81 |
+
},
|
82 |
+
"attributes": {}
|
83 |
+
}
|
84 |
+
},
|
85 |
+
"total_flos": 7.38266686206994e+18,
|
86 |
+
"train_batch_size": 32,
|
87 |
+
"trial_name": null,
|
88 |
+
"trial_params": null
|
89 |
+
}
|
checkpoint-2756/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:57fdf6fbaad0fbc617e51d43e3e727a2cb5eff0f72c5fb35af4caa4e562ea3de
|
3 |
+
size 5304
|
checkpoint-4134/config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SiglipForImageClassification"
|
4 |
+
],
|
5 |
+
"id2label": {
|
6 |
+
"0": "Fall",
|
7 |
+
"1": "Spring",
|
8 |
+
"2": "Summer",
|
9 |
+
"3": "Winter"
|
10 |
+
},
|
11 |
+
"initializer_factor": 1.0,
|
12 |
+
"label2id": {
|
13 |
+
"Fall": 0,
|
14 |
+
"Spring": 1,
|
15 |
+
"Summer": 2,
|
16 |
+
"Winter": 3
|
17 |
+
},
|
18 |
+
"model_type": "siglip",
|
19 |
+
"problem_type": "single_label_classification",
|
20 |
+
"text_config": {
|
21 |
+
"attention_dropout": 0.0,
|
22 |
+
"hidden_act": "gelu_pytorch_tanh",
|
23 |
+
"hidden_size": 768,
|
24 |
+
"intermediate_size": 3072,
|
25 |
+
"layer_norm_eps": 1e-06,
|
26 |
+
"max_position_embeddings": 64,
|
27 |
+
"model_type": "siglip_text_model",
|
28 |
+
"num_attention_heads": 12,
|
29 |
+
"num_hidden_layers": 12,
|
30 |
+
"projection_size": 768,
|
31 |
+
"torch_dtype": "float32",
|
32 |
+
"vocab_size": 256000
|
33 |
+
},
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.50.3",
|
36 |
+
"vision_config": {
|
37 |
+
"attention_dropout": 0.0,
|
38 |
+
"hidden_act": "gelu_pytorch_tanh",
|
39 |
+
"hidden_size": 768,
|
40 |
+
"image_size": 224,
|
41 |
+
"intermediate_size": 3072,
|
42 |
+
"layer_norm_eps": 1e-06,
|
43 |
+
"model_type": "siglip_vision_model",
|
44 |
+
"num_attention_heads": 12,
|
45 |
+
"num_channels": 3,
|
46 |
+
"num_hidden_layers": 12,
|
47 |
+
"patch_size": 16,
|
48 |
+
"torch_dtype": "float32"
|
49 |
+
}
|
50 |
+
}
|
checkpoint-4134/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b9d36e8e7b1fa970a6d3d42382bd392470b39f99bea801ed1c0f9ba7259ce8da
|
3 |
+
size 371574144
|
checkpoint-4134/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:defdb96d9779cb6b8962dbff2fceedc97cd09ee58bb4e2d14c32bb1269a47489
|
3 |
+
size 686568058
|
checkpoint-4134/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-4134/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5478a919be85a579b7e7075ad64e2b2fdb812bb5f7dc5b68704fe8bb15677b4e
|
3 |
+
size 14244
|
checkpoint-4134/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e88d9ae696b399ca28ed144a2a4bf78436721fd3b11372f182950dd0afa86753
|
3 |
+
size 1064
|
checkpoint-4134/trainer_state.json
ADDED
@@ -0,0 +1,120 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_global_step": 4134,
|
3 |
+
"best_metric": 0.6796342730522156,
|
4 |
+
"best_model_checkpoint": "siglip2-finetune-full/checkpoint-4134",
|
5 |
+
"epoch": 3.0,
|
6 |
+
"eval_steps": 500,
|
7 |
+
"global_step": 4134,
|
8 |
+
"is_hyper_param_search": false,
|
9 |
+
"is_local_process_zero": true,
|
10 |
+
"is_world_process_zero": true,
|
11 |
+
"log_history": [
|
12 |
+
{
|
13 |
+
"epoch": 0.36284470246734396,
|
14 |
+
"grad_norm": 22.187894821166992,
|
15 |
+
"learning_rate": 8.898139079333986e-07,
|
16 |
+
"loss": 0.9313,
|
17 |
+
"step": 500
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.7256894049346879,
|
21 |
+
"grad_norm": 24.758203506469727,
|
22 |
+
"learning_rate": 7.67384916748286e-07,
|
23 |
+
"loss": 0.7624,
|
24 |
+
"step": 1000
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 1.0,
|
28 |
+
"eval_accuracy": 0.6740333998910873,
|
29 |
+
"eval_loss": 0.7454163432121277,
|
30 |
+
"eval_model_preparation_time": 0.0023,
|
31 |
+
"eval_runtime": 535.389,
|
32 |
+
"eval_samples_per_second": 82.318,
|
33 |
+
"eval_steps_per_second": 10.29,
|
34 |
+
"step": 1378
|
35 |
+
},
|
36 |
+
{
|
37 |
+
"epoch": 1.0885341074020318,
|
38 |
+
"grad_norm": 25.052143096923828,
|
39 |
+
"learning_rate": 6.449559255631733e-07,
|
40 |
+
"loss": 0.7082,
|
41 |
+
"step": 1500
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 1.4513788098693758,
|
45 |
+
"grad_norm": 26.29668617248535,
|
46 |
+
"learning_rate": 5.225269343780607e-07,
|
47 |
+
"loss": 0.6842,
|
48 |
+
"step": 2000
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.8142235123367199,
|
52 |
+
"grad_norm": 33.645015716552734,
|
53 |
+
"learning_rate": 4.000979431929481e-07,
|
54 |
+
"loss": 0.6637,
|
55 |
+
"step": 2500
|
56 |
+
},
|
57 |
+
{
|
58 |
+
"epoch": 2.0,
|
59 |
+
"eval_accuracy": 0.7012842621165366,
|
60 |
+
"eval_loss": 0.6948277354240417,
|
61 |
+
"eval_model_preparation_time": 0.0023,
|
62 |
+
"eval_runtime": 531.6957,
|
63 |
+
"eval_samples_per_second": 82.89,
|
64 |
+
"eval_steps_per_second": 10.361,
|
65 |
+
"step": 2756
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 2.1770682148040637,
|
69 |
+
"grad_norm": 32.486167907714844,
|
70 |
+
"learning_rate": 2.7766895200783546e-07,
|
71 |
+
"loss": 0.6562,
|
72 |
+
"step": 3000
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 2.539912917271408,
|
76 |
+
"grad_norm": 23.1379337310791,
|
77 |
+
"learning_rate": 1.552399608227228e-07,
|
78 |
+
"loss": 0.6325,
|
79 |
+
"step": 3500
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 2.9027576197387517,
|
83 |
+
"grad_norm": 29.30535125732422,
|
84 |
+
"learning_rate": 3.281096963761019e-08,
|
85 |
+
"loss": 0.6381,
|
86 |
+
"step": 4000
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 3.0,
|
90 |
+
"eval_accuracy": 0.7121301506625521,
|
91 |
+
"eval_loss": 0.6796342730522156,
|
92 |
+
"eval_model_preparation_time": 0.0023,
|
93 |
+
"eval_runtime": 530.9342,
|
94 |
+
"eval_samples_per_second": 83.008,
|
95 |
+
"eval_steps_per_second": 10.376,
|
96 |
+
"step": 4134
|
97 |
+
}
|
98 |
+
],
|
99 |
+
"logging_steps": 500,
|
100 |
+
"max_steps": 4134,
|
101 |
+
"num_input_tokens_seen": 0,
|
102 |
+
"num_train_epochs": 3,
|
103 |
+
"save_steps": 500,
|
104 |
+
"stateful_callbacks": {
|
105 |
+
"TrainerControl": {
|
106 |
+
"args": {
|
107 |
+
"should_epoch_stop": false,
|
108 |
+
"should_evaluate": false,
|
109 |
+
"should_log": false,
|
110 |
+
"should_save": true,
|
111 |
+
"should_training_stop": true
|
112 |
+
},
|
113 |
+
"attributes": {}
|
114 |
+
}
|
115 |
+
},
|
116 |
+
"total_flos": 1.107400029310491e+19,
|
117 |
+
"train_batch_size": 32,
|
118 |
+
"trial_name": null,
|
119 |
+
"trial_params": null
|
120 |
+
}
|
checkpoint-4134/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:57fdf6fbaad0fbc617e51d43e3e727a2cb5eff0f72c5fb35af4caa4e562ea3de
|
3 |
+
size 5304
|
config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SiglipForImageClassification"
|
4 |
+
],
|
5 |
+
"id2label": {
|
6 |
+
"0": "Fall",
|
7 |
+
"1": "Spring",
|
8 |
+
"2": "Summer",
|
9 |
+
"3": "Winter"
|
10 |
+
},
|
11 |
+
"initializer_factor": 1.0,
|
12 |
+
"label2id": {
|
13 |
+
"Fall": 0,
|
14 |
+
"Spring": 1,
|
15 |
+
"Summer": 2,
|
16 |
+
"Winter": 3
|
17 |
+
},
|
18 |
+
"model_type": "siglip",
|
19 |
+
"problem_type": "single_label_classification",
|
20 |
+
"text_config": {
|
21 |
+
"attention_dropout": 0.0,
|
22 |
+
"hidden_act": "gelu_pytorch_tanh",
|
23 |
+
"hidden_size": 768,
|
24 |
+
"intermediate_size": 3072,
|
25 |
+
"layer_norm_eps": 1e-06,
|
26 |
+
"max_position_embeddings": 64,
|
27 |
+
"model_type": "siglip_text_model",
|
28 |
+
"num_attention_heads": 12,
|
29 |
+
"num_hidden_layers": 12,
|
30 |
+
"projection_size": 768,
|
31 |
+
"torch_dtype": "float32",
|
32 |
+
"vocab_size": 256000
|
33 |
+
},
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.50.3",
|
36 |
+
"vision_config": {
|
37 |
+
"attention_dropout": 0.0,
|
38 |
+
"hidden_act": "gelu_pytorch_tanh",
|
39 |
+
"hidden_size": 768,
|
40 |
+
"image_size": 224,
|
41 |
+
"intermediate_size": 3072,
|
42 |
+
"layer_norm_eps": 1e-06,
|
43 |
+
"model_type": "siglip_vision_model",
|
44 |
+
"num_attention_heads": 12,
|
45 |
+
"num_channels": 3,
|
46 |
+
"num_hidden_layers": 12,
|
47 |
+
"patch_size": 16,
|
48 |
+
"torch_dtype": "float32"
|
49 |
+
}
|
50 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b9d36e8e7b1fa970a6d3d42382bd392470b39f99bea801ed1c0f9ba7259ce8da
|
3 |
+
size 371574144
|
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 sha256:57fdf6fbaad0fbc617e51d43e3e727a2cb5eff0f72c5fb35af4caa4e562ea3de
|
3 |
+
size 5304
|