Training in progress, step 500
Browse files- config.json +186 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- preprocessor_config.json +26 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +80 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
@@ -0,0 +1,186 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/layoutlmv3-large",
|
3 |
+
"architectures": [
|
4 |
+
"LayoutLMv3ForTokenClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"coordinate_size": 171,
|
10 |
+
"eos_token_id": 2,
|
11 |
+
"has_relative_attention_bias": true,
|
12 |
+
"has_spatial_attention_bias": true,
|
13 |
+
"hidden_act": "gelu",
|
14 |
+
"hidden_dropout_prob": 0.1,
|
15 |
+
"hidden_size": 1024,
|
16 |
+
"id2label": {
|
17 |
+
"0": "O",
|
18 |
+
"1": "B-ENERGY_KJ_100G",
|
19 |
+
"2": "I-ENERGY_KJ_100G",
|
20 |
+
"3": "B-VITAMIN_D_SERVING",
|
21 |
+
"4": "I-VITAMIN_D_SERVING",
|
22 |
+
"5": "B-SODIUM_SERVING",
|
23 |
+
"6": "I-SODIUM_SERVING",
|
24 |
+
"7": "B-PROTEINS_SERVING",
|
25 |
+
"8": "I-PROTEINS_SERVING",
|
26 |
+
"9": "B-ADDED_SUGARS_SERVING",
|
27 |
+
"10": "I-ADDED_SUGARS_SERVING",
|
28 |
+
"11": "B-CALCIUM_SERVING",
|
29 |
+
"12": "I-CALCIUM_SERVING",
|
30 |
+
"13": "B-FAT_SERVING",
|
31 |
+
"14": "I-FAT_SERVING",
|
32 |
+
"15": "B-ENERGY_KJ_SERVING",
|
33 |
+
"16": "I-ENERGY_KJ_SERVING",
|
34 |
+
"17": "B-SUGARS_100G",
|
35 |
+
"18": "I-SUGARS_100G",
|
36 |
+
"19": "B-SATURATED_FAT_SERVING",
|
37 |
+
"20": "I-SATURATED_FAT_SERVING",
|
38 |
+
"21": "B-SERVING_SIZE",
|
39 |
+
"22": "I-SERVING_SIZE",
|
40 |
+
"23": "B-SALT_SERVING",
|
41 |
+
"24": "I-SALT_SERVING",
|
42 |
+
"25": "B-ENERGY_KCAL_SERVING",
|
43 |
+
"26": "I-ENERGY_KCAL_SERVING",
|
44 |
+
"27": "B-FAT_100G",
|
45 |
+
"28": "I-FAT_100G",
|
46 |
+
"29": "B-SUGARS_SERVING",
|
47 |
+
"30": "I-SUGARS_SERVING",
|
48 |
+
"31": "B-FIBER_SERVING",
|
49 |
+
"32": "I-FIBER_SERVING",
|
50 |
+
"33": "B-TRANS_FAT_SERVING",
|
51 |
+
"34": "I-TRANS_FAT_SERVING",
|
52 |
+
"35": "B-POTASSIUM_SERVING",
|
53 |
+
"36": "I-POTASSIUM_SERVING",
|
54 |
+
"37": "B-CARBOHYDRATES_100G",
|
55 |
+
"38": "I-CARBOHYDRATES_100G",
|
56 |
+
"39": "B-POTASSIUM_100G",
|
57 |
+
"40": "I-POTASSIUM_100G",
|
58 |
+
"41": "B-IRON_SERVING",
|
59 |
+
"42": "I-IRON_SERVING",
|
60 |
+
"43": "B-CHOLESTEROL_100G",
|
61 |
+
"44": "I-CHOLESTEROL_100G",
|
62 |
+
"45": "B-TRANS_FAT_100G",
|
63 |
+
"46": "I-TRANS_FAT_100G",
|
64 |
+
"47": "B-ADDED_SUGARS_100G",
|
65 |
+
"48": "I-ADDED_SUGARS_100G",
|
66 |
+
"49": "B-FIBER_100G",
|
67 |
+
"50": "I-FIBER_100G",
|
68 |
+
"51": "B-CALCIUM_100G",
|
69 |
+
"52": "I-CALCIUM_100G",
|
70 |
+
"53": "B-SODIUM_100G",
|
71 |
+
"54": "I-SODIUM_100G",
|
72 |
+
"55": "B-ENERGY_KCAL_100G",
|
73 |
+
"56": "I-ENERGY_KCAL_100G",
|
74 |
+
"57": "B-CHOLESTEROL_SERVING",
|
75 |
+
"58": "I-CHOLESTEROL_SERVING",
|
76 |
+
"59": "B-CARBOHYDRATES_SERVING",
|
77 |
+
"60": "I-CARBOHYDRATES_SERVING",
|
78 |
+
"61": "B-SALT_100G",
|
79 |
+
"62": "I-SALT_100G",
|
80 |
+
"63": "B-VITAMIN_D_100G",
|
81 |
+
"64": "I-VITAMIN_D_100G",
|
82 |
+
"65": "B-SATURATED_FAT_100G",
|
83 |
+
"66": "I-SATURATED_FAT_100G",
|
84 |
+
"67": "B-PROTEINS_100G",
|
85 |
+
"68": "I-PROTEINS_100G",
|
86 |
+
"69": "B-IRON_100G",
|
87 |
+
"70": "I-IRON_100G"
|
88 |
+
},
|
89 |
+
"initializer_range": 0.02,
|
90 |
+
"input_size": 224,
|
91 |
+
"intermediate_size": 4096,
|
92 |
+
"label2id": {
|
93 |
+
"B-ADDED_SUGARS_100G": 47,
|
94 |
+
"B-ADDED_SUGARS_SERVING": 9,
|
95 |
+
"B-CALCIUM_100G": 51,
|
96 |
+
"B-CALCIUM_SERVING": 11,
|
97 |
+
"B-CARBOHYDRATES_100G": 37,
|
98 |
+
"B-CARBOHYDRATES_SERVING": 59,
|
99 |
+
"B-CHOLESTEROL_100G": 43,
|
100 |
+
"B-CHOLESTEROL_SERVING": 57,
|
101 |
+
"B-ENERGY_KCAL_100G": 55,
|
102 |
+
"B-ENERGY_KCAL_SERVING": 25,
|
103 |
+
"B-ENERGY_KJ_100G": 1,
|
104 |
+
"B-ENERGY_KJ_SERVING": 15,
|
105 |
+
"B-FAT_100G": 27,
|
106 |
+
"B-FAT_SERVING": 13,
|
107 |
+
"B-FIBER_100G": 49,
|
108 |
+
"B-FIBER_SERVING": 31,
|
109 |
+
"B-IRON_100G": 69,
|
110 |
+
"B-IRON_SERVING": 41,
|
111 |
+
"B-POTASSIUM_100G": 39,
|
112 |
+
"B-POTASSIUM_SERVING": 35,
|
113 |
+
"B-PROTEINS_100G": 67,
|
114 |
+
"B-PROTEINS_SERVING": 7,
|
115 |
+
"B-SALT_100G": 61,
|
116 |
+
"B-SALT_SERVING": 23,
|
117 |
+
"B-SATURATED_FAT_100G": 65,
|
118 |
+
"B-SATURATED_FAT_SERVING": 19,
|
119 |
+
"B-SERVING_SIZE": 21,
|
120 |
+
"B-SODIUM_100G": 53,
|
121 |
+
"B-SODIUM_SERVING": 5,
|
122 |
+
"B-SUGARS_100G": 17,
|
123 |
+
"B-SUGARS_SERVING": 29,
|
124 |
+
"B-TRANS_FAT_100G": 45,
|
125 |
+
"B-TRANS_FAT_SERVING": 33,
|
126 |
+
"B-VITAMIN_D_100G": 63,
|
127 |
+
"B-VITAMIN_D_SERVING": 3,
|
128 |
+
"I-ADDED_SUGARS_100G": 48,
|
129 |
+
"I-ADDED_SUGARS_SERVING": 10,
|
130 |
+
"I-CALCIUM_100G": 52,
|
131 |
+
"I-CALCIUM_SERVING": 12,
|
132 |
+
"I-CARBOHYDRATES_100G": 38,
|
133 |
+
"I-CARBOHYDRATES_SERVING": 60,
|
134 |
+
"I-CHOLESTEROL_100G": 44,
|
135 |
+
"I-CHOLESTEROL_SERVING": 58,
|
136 |
+
"I-ENERGY_KCAL_100G": 56,
|
137 |
+
"I-ENERGY_KCAL_SERVING": 26,
|
138 |
+
"I-ENERGY_KJ_100G": 2,
|
139 |
+
"I-ENERGY_KJ_SERVING": 16,
|
140 |
+
"I-FAT_100G": 28,
|
141 |
+
"I-FAT_SERVING": 14,
|
142 |
+
"I-FIBER_100G": 50,
|
143 |
+
"I-FIBER_SERVING": 32,
|
144 |
+
"I-IRON_100G": 70,
|
145 |
+
"I-IRON_SERVING": 42,
|
146 |
+
"I-POTASSIUM_100G": 40,
|
147 |
+
"I-POTASSIUM_SERVING": 36,
|
148 |
+
"I-PROTEINS_100G": 68,
|
149 |
+
"I-PROTEINS_SERVING": 8,
|
150 |
+
"I-SALT_100G": 62,
|
151 |
+
"I-SALT_SERVING": 24,
|
152 |
+
"I-SATURATED_FAT_100G": 66,
|
153 |
+
"I-SATURATED_FAT_SERVING": 20,
|
154 |
+
"I-SERVING_SIZE": 22,
|
155 |
+
"I-SODIUM_100G": 54,
|
156 |
+
"I-SODIUM_SERVING": 6,
|
157 |
+
"I-SUGARS_100G": 18,
|
158 |
+
"I-SUGARS_SERVING": 30,
|
159 |
+
"I-TRANS_FAT_100G": 46,
|
160 |
+
"I-TRANS_FAT_SERVING": 34,
|
161 |
+
"I-VITAMIN_D_100G": 64,
|
162 |
+
"I-VITAMIN_D_SERVING": 4,
|
163 |
+
"O": 0
|
164 |
+
},
|
165 |
+
"layer_norm_eps": 1e-05,
|
166 |
+
"max_2d_position_embeddings": 1024,
|
167 |
+
"max_position_embeddings": 514,
|
168 |
+
"max_rel_2d_pos": 256,
|
169 |
+
"max_rel_pos": 128,
|
170 |
+
"model_type": "layoutlmv3",
|
171 |
+
"num_attention_heads": 16,
|
172 |
+
"num_channels": 3,
|
173 |
+
"num_hidden_layers": 24,
|
174 |
+
"pad_token_id": 1,
|
175 |
+
"patch_size": 16,
|
176 |
+
"rel_2d_pos_bins": 64,
|
177 |
+
"rel_pos_bins": 32,
|
178 |
+
"second_input_size": 112,
|
179 |
+
"shape_size": 170,
|
180 |
+
"text_embed": true,
|
181 |
+
"torch_dtype": "float32",
|
182 |
+
"transformers_version": "4.48.3",
|
183 |
+
"type_vocab_size": 1,
|
184 |
+
"visual_embed": true,
|
185 |
+
"vocab_size": 50265
|
186 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bd1fddc7a4976427529736af9bf2992e1bd9d999b49fe509aadefbc27401263b
|
3 |
+
size 1428562028
|
preprocessor_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"apply_ocr": false,
|
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": "LayoutLMv3ImageProcessor",
|
12 |
+
"image_std": [
|
13 |
+
0.5,
|
14 |
+
0.5,
|
15 |
+
0.5
|
16 |
+
],
|
17 |
+
"ocr_lang": null,
|
18 |
+
"processor_class": "LayoutLMv3Processor",
|
19 |
+
"resample": 2,
|
20 |
+
"rescale_factor": 0.00392156862745098,
|
21 |
+
"size": {
|
22 |
+
"height": 224,
|
23 |
+
"width": 224
|
24 |
+
},
|
25 |
+
"tesseract_config": ""
|
26 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": true,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"cls_token": {
|
10 |
+
"content": "<s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": true,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"eos_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": true,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"mask_token": {
|
24 |
+
"content": "<mask>",
|
25 |
+
"lstrip": true,
|
26 |
+
"normalized": true,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"pad_token": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": true,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
},
|
37 |
+
"sep_token": {
|
38 |
+
"content": "</s>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": true,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false
|
43 |
+
},
|
44 |
+
"unk_token": {
|
45 |
+
"content": "<unk>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": true,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false
|
50 |
+
}
|
51 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": true,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"0": {
|
5 |
+
"content": "<s>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
},
|
12 |
+
"1": {
|
13 |
+
"content": "<pad>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": true,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": true
|
19 |
+
},
|
20 |
+
"2": {
|
21 |
+
"content": "</s>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": true,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false,
|
26 |
+
"special": true
|
27 |
+
},
|
28 |
+
"3": {
|
29 |
+
"content": "<unk>",
|
30 |
+
"lstrip": false,
|
31 |
+
"normalized": true,
|
32 |
+
"rstrip": false,
|
33 |
+
"single_word": false,
|
34 |
+
"special": true
|
35 |
+
},
|
36 |
+
"50264": {
|
37 |
+
"content": "<mask>",
|
38 |
+
"lstrip": true,
|
39 |
+
"normalized": true,
|
40 |
+
"rstrip": false,
|
41 |
+
"single_word": false,
|
42 |
+
"special": true
|
43 |
+
}
|
44 |
+
},
|
45 |
+
"apply_ocr": false,
|
46 |
+
"bos_token": "<s>",
|
47 |
+
"clean_up_tokenization_spaces": false,
|
48 |
+
"cls_token": "<s>",
|
49 |
+
"cls_token_box": [
|
50 |
+
0,
|
51 |
+
0,
|
52 |
+
0,
|
53 |
+
0
|
54 |
+
],
|
55 |
+
"eos_token": "</s>",
|
56 |
+
"errors": "replace",
|
57 |
+
"extra_special_tokens": {},
|
58 |
+
"mask_token": "<mask>",
|
59 |
+
"model_max_length": 512,
|
60 |
+
"only_label_first_subword": true,
|
61 |
+
"pad_token": "<pad>",
|
62 |
+
"pad_token_box": [
|
63 |
+
0,
|
64 |
+
0,
|
65 |
+
0,
|
66 |
+
0
|
67 |
+
],
|
68 |
+
"pad_token_label": -100,
|
69 |
+
"processor_class": "LayoutLMv3Processor",
|
70 |
+
"sep_token": "</s>",
|
71 |
+
"sep_token_box": [
|
72 |
+
0,
|
73 |
+
0,
|
74 |
+
0,
|
75 |
+
0
|
76 |
+
],
|
77 |
+
"tokenizer_class": "LayoutLMv3Tokenizer",
|
78 |
+
"trim_offsets": true,
|
79 |
+
"unk_token": "<unk>"
|
80 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0abf3b068bef9045677f907446719fea97d2f826bd2a286df65ac4047b6f7ed2
|
3 |
+
size 5368
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|