Hugging Face's logo

xiaoyao9184
/
surya_text_recognition

Transformers
Safetensors
surya-multimodal-foundation
Model card Files
xet
 Community
xiaoyao9184 commited on
Commit
a7e5dd6
·
verified ·
1 Parent(s): c720d45

Add files using upload-large-folder tool

Browse files
Files changed (5) hide show
  1. config.json +8 -5
  2. manifest.json +1 -1
  3. model.safetensors +2 -2
  4. processor_config.json +2 -2
  5. training_args.bin +2 -2
config.json CHANGED
@@ -1,11 +1,12 @@
1
  {
 
2
  "architectures": [
3
  "SuryaModel"
4
  ],
5
  "bbox_embed_size": 1024,
6
  "bbox_size": 1024,
7
  "beacon_token_id": 66567,
8
- "beacon_token_interval": 512,
9
  "blank_bbox_token_id": 1025,
10
  "bos_token_id": {
11
  "block_without_boxes": 66561,
@@ -15,6 +16,7 @@
15
  "table_structure": 66569
16
  },
17
  "decoder": {
 
18
  "attention_dropout": 0.0,
19
  "hidden_act": "silu",
20
  "hidden_size": 1280,
@@ -34,6 +36,7 @@
34
  "use_sliding_window": false,
35
  "vocab_size": 218225
36
  },
 
37
  "embedding_dropout_prob": 0.05,
38
  "eoi_token_id": 66554,
39
  "eos_token_id": 66553,
@@ -47,7 +50,7 @@
47
  "model_type": "surya-multimodal-foundation",
48
  "multi_output_distance": 0,
49
  "num_attention_heads": 16,
50
- "num_beacon_tokens": 4,
51
  "num_hidden_layers": 12,
52
  "num_key_value_heads": 4,
53
  "num_register_tokens": 4,
@@ -60,7 +63,7 @@
60
  66565,
61
  66566
62
  ],
63
- "sliding_window": 2048,
64
  "special_ocr_tokens": {
65
  "all": [
66
  "</S>",
@@ -2157,10 +2160,10 @@
2157
  ]
2158
  }
2159
  },
2160
- "torch_dtype": "bfloat16",
2161
- "transformers_version": "4.53.0",
2162
  "use_cut_cross_entropy": false,
2163
  "vision_encoder": {
 
2164
  "depth": 12,
2165
  "fullatt_block_indexes": [
2166
  3,
 
1
  {
2
+ "_attn_implementation_autoset": true,
3
  "architectures": [
4
  "SuryaModel"
5
  ],
6
  "bbox_embed_size": 1024,
7
  "bbox_size": 1024,
8
  "beacon_token_id": 66567,
9
+ "beacon_token_interval": 64,
10
  "blank_bbox_token_id": 1025,
11
  "bos_token_id": {
12
  "block_without_boxes": 66561,
 
16
  "table_structure": 66569
17
  },
18
  "decoder": {
19
+ "_attn_implementation_autoset": true,
20
  "attention_dropout": 0.0,
21
  "hidden_act": "silu",
22
  "hidden_size": 1280,
 
36
  "use_sliding_window": false,
37
  "vocab_size": 218225
38
  },
39
+ "dtype": "bfloat16",
40
  "embedding_dropout_prob": 0.05,
41
  "eoi_token_id": 66554,
42
  "eos_token_id": 66553,
 
50
  "model_type": "surya-multimodal-foundation",
51
  "multi_output_distance": 0,
52
  "num_attention_heads": 16,
53
+ "num_beacon_tokens": 1,
54
  "num_hidden_layers": 12,
55
  "num_key_value_heads": 4,
56
  "num_register_tokens": 4,
 
63
  66565,
64
  66566
65
  ],
66
+ "sliding_window": 256,
67
  "special_ocr_tokens": {
68
  "all": [
69
  "</S>",
 
2160
  ]
2161
  }
2162
  },
2163
+ "transformers_version": "4.56.0",
 
2164
  "use_cut_cross_entropy": false,
2165
  "vision_encoder": {
2166
+ "_attn_implementation_autoset": true,
2167
  "depth": 12,
2168
  "fullatt_block_indexes": [
2169
  3,
manifest.json CHANGED
@@ -1 +1 @@
1
- {"files": [".gitattributes", "README.md", "specials_dict.json", "special_tokens_map.json", "vocab_math.json", "specials.json", "tokenizer_config.json", "preprocessor_config.json", "config.json", "processor_config.json", "model.safetensors"]}
 
1
+ {"files": [".gitattributes", "README.md", "specials_dict.json", "training_args.bin", "special_tokens_map.json", "vocab_math.json", "specials.json", "tokenizer_config.json", "preprocessor_config.json", "config.json", "processor_config.json", "model.safetensors"]}
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:911020f51fd087cadeaa73bdcd5ded69beee3b4ede068a72600699af4bc301ca
3
- size 1406572398
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd07d624d42dc1b2f6bd871777c088052635294e1637805304300bc4ec8fb3ed
3
+ size 1438882494
processor_config.json CHANGED
@@ -2,10 +2,10 @@
2
  "add_reasoning_tokens": false,
3
  "apply_augments": true,
4
  "bbox_size": 1024,
5
- "beacon_token_interval": 512,
6
  "blank_bbox_token_id": 1025,
7
  "merge_size": 2,
8
- "num_beacon_tokens": 4,
9
  "num_register_tokens": 4,
10
  "patch_size": 14,
11
  "processor_class": "MultimodalFoundationProcessor",
 
2
  "add_reasoning_tokens": false,
3
  "apply_augments": true,
4
  "bbox_size": 1024,
5
+ "beacon_token_interval": 64,
6
  "blank_bbox_token_id": 1025,
7
  "merge_size": 2,
8
+ "num_beacon_tokens": 1,
9
  "num_register_tokens": 4,
10
  "patch_size": 14,
11
  "processor_class": "MultimodalFoundationProcessor",
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:555cdb558a7ac24c8241387d5c72b1657c81cae1b0a50ad4be0d65eb65d3a487
3
- size 7505
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e72b841abbff68eaf26ecde03b48678ef52085b5d75542d3818391f8cac43be0
3
+ size 7633