Hugging Face's logo

tnnanh1005
/
PhysCLIP

Safetensors
Model card Files
xet
Community
tnnanh1005 commited on
Commit
4145f76
·
verified ·
1 Parent(s): 66b561a

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. physclip_physbert/checkpoint-epoch-100/config.json +53 -0
  2. physclip_physbert/checkpoint-epoch-100/preprocessor_config.json +27 -0
  3. physclip_physbert/checkpoint-epoch-100/special_tokens_map.json +37 -0
  4. physclip_physbert/checkpoint-epoch-100/tokenizer.json +0 -0
  5. physclip_physbert/checkpoint-epoch-100/tokenizer_config.json +63 -0
  6. physclip_physbert/checkpoint-epoch-100/vocab.txt +0 -0
  7. physclip_physbert/checkpoint-epoch-102/config.json +53 -0
  8. physclip_physbert/checkpoint-epoch-102/preprocessor_config.json +27 -0
  9. physclip_physbert/checkpoint-epoch-102/special_tokens_map.json +37 -0
  10. physclip_physbert/checkpoint-epoch-102/tokenizer.json +0 -0
  11. physclip_physbert/checkpoint-epoch-102/tokenizer_config.json +63 -0
  12. physclip_physbert/checkpoint-epoch-102/vocab.txt +0 -0
  13. physclip_physbert/checkpoint-epoch-104/config.json +53 -0
  14. physclip_physbert/checkpoint-epoch-104/preprocessor_config.json +27 -0
  15. physclip_physbert/checkpoint-epoch-104/special_tokens_map.json +37 -0
  16. physclip_physbert/checkpoint-epoch-104/tokenizer.json +0 -0
  17. physclip_physbert/checkpoint-epoch-104/tokenizer_config.json +63 -0
  18. physclip_physbert/checkpoint-epoch-104/vocab.txt +0 -0
  19. physclip_physbert/checkpoint-epoch-106/config.json +53 -0
  20. physclip_physbert/checkpoint-epoch-106/preprocessor_config.json +27 -0
  21. physclip_physbert/checkpoint-epoch-106/special_tokens_map.json +37 -0
  22. physclip_physbert/checkpoint-epoch-106/tokenizer.json +0 -0
  23. physclip_physbert/checkpoint-epoch-106/tokenizer_config.json +63 -0
  24. physclip_physbert/checkpoint-epoch-106/vocab.txt +0 -0
  25. physclip_physbert/checkpoint-epoch-108/config.json +53 -0
  26. physclip_physbert/checkpoint-epoch-108/preprocessor_config.json +27 -0
  27. physclip_physbert/checkpoint-epoch-108/special_tokens_map.json +37 -0
  28. physclip_physbert/checkpoint-epoch-108/tokenizer.json +0 -0
  29. physclip_physbert/checkpoint-epoch-108/tokenizer_config.json +63 -0
  30. physclip_physbert/checkpoint-epoch-108/vocab.txt +0 -0
  31. physclip_physbert/checkpoint-epoch-110/config.json +53 -0
  32. physclip_physbert/checkpoint-epoch-110/preprocessor_config.json +27 -0
  33. physclip_physbert/checkpoint-epoch-110/special_tokens_map.json +37 -0
  34. physclip_physbert/checkpoint-epoch-110/tokenizer.json +0 -0
  35. physclip_physbert/checkpoint-epoch-110/tokenizer_config.json +63 -0
  36. physclip_physbert/checkpoint-epoch-110/vocab.txt +0 -0
  37. physclip_physbert/checkpoint-epoch-112/config.json +53 -0
  38. physclip_physbert/checkpoint-epoch-112/preprocessor_config.json +27 -0
  39. physclip_physbert/checkpoint-epoch-112/special_tokens_map.json +37 -0
  40. physclip_physbert/checkpoint-epoch-112/tokenizer.json +0 -0
  41. physclip_physbert/checkpoint-epoch-112/tokenizer_config.json +63 -0
  42. physclip_physbert/checkpoint-epoch-112/vocab.txt +0 -0
  43. physclip_physbert/checkpoint-epoch-114/config.json +53 -0
  44. physclip_physbert/checkpoint-epoch-114/preprocessor_config.json +27 -0
  45. physclip_physbert/checkpoint-epoch-114/special_tokens_map.json +37 -0
  46. physclip_physbert/checkpoint-epoch-114/tokenizer.json +0 -0
  47. physclip_physbert/checkpoint-epoch-114/tokenizer_config.json +63 -0
  48. physclip_physbert/checkpoint-epoch-114/vocab.txt +0 -0
  49. physclip_physbert/checkpoint-epoch-116/config.json +53 -0
  50. physclip_physbert/checkpoint-epoch-116/preprocessor_config.json +27 -0
physclip_physbert/checkpoint-epoch-100/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-100/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-100/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-100/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-100/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-100/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-102/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-102/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-102/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-102/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-102/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-102/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-104/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-104/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-104/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-104/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-104/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-104/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-106/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-106/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-106/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-106/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-106/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-106/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-108/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-108/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-108/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-108/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-108/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-108/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-110/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-110/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-110/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-110/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-110/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-110/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-112/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-112/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-112/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-112/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-112/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-112/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-114/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-114/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
physclip_physbert/checkpoint-epoch-114/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
physclip_physbert/checkpoint-epoch-114/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-114/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
physclip_physbert/checkpoint-epoch-114/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
physclip_physbert/checkpoint-epoch-116/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionTextDualEncoderModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "logit_scale_init_value": 2.6592,
7
+ "model_type": "vision-text-dual-encoder",
8
+ "projection_dim": 512,
9
+ "text_config": {
10
+ "_name_or_path": "thellert/physbert_uncased",
11
+ "architectures": [
12
+ "BertModel"
13
+ ],
14
+ "attention_probs_dropout_prob": 0.1,
15
+ "classifier_dropout": null,
16
+ "dtype": "float32",
17
+ "gradient_checkpointing": false,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "position_embedding_type": "absolute",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ },
33
+ "transformers_version": "4.57.3",
34
+ "vision_config": {
35
+ "_name_or_path": "openai/clip-vit-base-patch32",
36
+ "attention_dropout": 0.0,
37
+ "dropout": 0.0,
38
+ "dtype": "float32",
39
+ "hidden_act": "quick_gelu",
40
+ "hidden_size": 768,
41
+ "image_size": 224,
42
+ "initializer_factor": 1.0,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 3072,
45
+ "layer_norm_eps": 1e-05,
46
+ "model_type": "clip_vision_model",
47
+ "num_attention_heads": 12,
48
+ "num_channels": 3,
49
+ "num_hidden_layers": 12,
50
+ "patch_size": 32,
51
+ "projection_dim": 512
52
+ }
53
+ }
physclip_physbert/checkpoint-epoch-116/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }