nvm472001 commited on
Commit
989847a
1 Parent(s): ab67346

Upload 9 files

Browse files
config.json ADDED
@@ -0,0 +1,193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/layoutxlm-base",
3
+ "architectures": [
4
+ "LayoutLMv2ForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "convert_sync_batchnorm": true,
9
+ "coordinate_size": 128,
10
+ "detectron2_config_args": {
11
+ "MODEL.ANCHOR_GENERATOR.SIZES": [
12
+ [
13
+ 32
14
+ ],
15
+ [
16
+ 64
17
+ ],
18
+ [
19
+ 128
20
+ ],
21
+ [
22
+ 256
23
+ ],
24
+ [
25
+ 512
26
+ ]
27
+ ],
28
+ "MODEL.BACKBONE.NAME": "build_resnet_fpn_backbone",
29
+ "MODEL.FPN.IN_FEATURES": [
30
+ "res2",
31
+ "res3",
32
+ "res4",
33
+ "res5"
34
+ ],
35
+ "MODEL.MASK_ON": true,
36
+ "MODEL.PIXEL_STD": [
37
+ 57.375,
38
+ 57.12,
39
+ 58.395
40
+ ],
41
+ "MODEL.POST_NMS_TOPK_TEST": 1000,
42
+ "MODEL.RESNETS.ASPECT_RATIOS": [
43
+ [
44
+ 0.5,
45
+ 1.0,
46
+ 2.0
47
+ ]
48
+ ],
49
+ "MODEL.RESNETS.DEPTH": 101,
50
+ "MODEL.RESNETS.NUM_GROUPS": 32,
51
+ "MODEL.RESNETS.OUT_FEATURES": [
52
+ "res2",
53
+ "res3",
54
+ "res4",
55
+ "res5"
56
+ ],
57
+ "MODEL.RESNETS.SIZES": [
58
+ [
59
+ 32
60
+ ],
61
+ [
62
+ 64
63
+ ],
64
+ [
65
+ 128
66
+ ],
67
+ [
68
+ 256
69
+ ],
70
+ [
71
+ 512
72
+ ]
73
+ ],
74
+ "MODEL.RESNETS.STRIDE_IN_1X1": false,
75
+ "MODEL.RESNETS.WIDTH_PER_GROUP": 8,
76
+ "MODEL.ROI_BOX_HEAD.NAME": "FastRCNNConvFCHead",
77
+ "MODEL.ROI_BOX_HEAD.NUM_FC": 2,
78
+ "MODEL.ROI_BOX_HEAD.POOLER_RESOLUTION": 14,
79
+ "MODEL.ROI_HEADS.IN_FEATURES": [
80
+ "p2",
81
+ "p3",
82
+ "p4",
83
+ "p5"
84
+ ],
85
+ "MODEL.ROI_HEADS.NAME": "StandardROIHeads",
86
+ "MODEL.ROI_HEADS.NUM_CLASSES": 5,
87
+ "MODEL.ROI_MASK_HEAD.NAME": "MaskRCNNConvUpsampleHead",
88
+ "MODEL.ROI_MASK_HEAD.NUM_CONV": 4,
89
+ "MODEL.ROI_MASK_HEAD.POOLER_RESOLUTION": 7,
90
+ "MODEL.RPN.IN_FEATURES": [
91
+ "p2",
92
+ "p3",
93
+ "p4",
94
+ "p5",
95
+ "p6"
96
+ ],
97
+ "MODEL.RPN.POST_NMS_TOPK_TRAIN": 1000,
98
+ "MODEL.RPN.PRE_NMS_TOPK_TEST": 1000,
99
+ "MODEL.RPN.PRE_NMS_TOPK_TRAIN": 2000
100
+ },
101
+ "eos_token_id": 2,
102
+ "fast_qkv": false,
103
+ "gradient_checkpointing": false,
104
+ "has_relative_attention_bias": false,
105
+ "has_spatial_attention_bias": false,
106
+ "has_visual_segment_embedding": true,
107
+ "hidden_act": "gelu",
108
+ "hidden_dropout_prob": 0.1,
109
+ "hidden_size": 768,
110
+ "id2label": {
111
+ "0": "person_name",
112
+ "1": "dob_key",
113
+ "2": "dob_value",
114
+ "3": "gender_key",
115
+ "4": "gender_value",
116
+ "5": "phonenumber_key",
117
+ "6": "phonenumber_value",
118
+ "7": "email_key",
119
+ "8": "email_value",
120
+ "9": "address_key",
121
+ "10": "address_value",
122
+ "11": "socical_address_value",
123
+ "12": "education",
124
+ "13": "education_name",
125
+ "14": "education_time",
126
+ "15": "experience",
127
+ "16": "experience_name",
128
+ "17": "experience_time",
129
+ "18": "information",
130
+ "19": "Education_detail",
131
+ "20": "Experience_detail",
132
+ "21": "Personal_skill",
133
+ "22": "Personal_skill_detail",
134
+ "23": "Certificate",
135
+ "24": "Certificate_name",
136
+ "25": "Certificate_detail",
137
+ "26": "Certificate_time"
138
+ },
139
+ "image_feature_pool_shape": [
140
+ 7,
141
+ 7,
142
+ 256
143
+ ],
144
+ "initializer_range": 0.02,
145
+ "intermediate_size": 3072,
146
+ "label2id": {
147
+ "Certificate": 23,
148
+ "Certificate_detail": 25,
149
+ "Certificate_name": 24,
150
+ "Certificate_time": 26,
151
+ "Education_detail": 19,
152
+ "Experience_detail": 20,
153
+ "Personal_skill": 21,
154
+ "Personal_skill_detail": 22,
155
+ "address_key": 9,
156
+ "address_value": 10,
157
+ "dob_key": 1,
158
+ "dob_value": 2,
159
+ "education": 12,
160
+ "education_name": 13,
161
+ "education_time": 14,
162
+ "email_key": 7,
163
+ "email_value": 8,
164
+ "experience": 15,
165
+ "experience_name": 16,
166
+ "experience_time": 17,
167
+ "gender_key": 3,
168
+ "gender_value": 4,
169
+ "information": 18,
170
+ "person_name": 0,
171
+ "phonenumber_key": 5,
172
+ "phonenumber_value": 6,
173
+ "socical_address_value": 11
174
+ },
175
+ "layer_norm_eps": 1e-05,
176
+ "max_2d_position_embeddings": 1024,
177
+ "max_position_embeddings": 514,
178
+ "max_rel_2d_pos": 256,
179
+ "max_rel_pos": 128,
180
+ "model_type": "layoutlmv2",
181
+ "num_attention_heads": 12,
182
+ "num_hidden_layers": 12,
183
+ "output_past": true,
184
+ "pad_token_id": 1,
185
+ "rel_2d_pos_bins": 64,
186
+ "rel_pos_bins": 32,
187
+ "shape_size": 128,
188
+ "tokenizer_class": "LayoutXLMTokenizer",
189
+ "torch_dtype": "float32",
190
+ "transformers_version": "4.25.1",
191
+ "type_vocab_size": 1,
192
+ "vocab_size": 250002
193
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab36f9f6cf55adb97b1a4f05e93b94b6b9411d600dc2048a7d6be3841c13e6a9
3
+ size 1476582209
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51b57f02fbe76bafb218521f9642fd3fad5aeeb257ad48f54babe5945c3a1825
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf4ec6118d86e98b5c1e9216a55a555ad97f965447b129371d09ec14096da902
3
+ size 627
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "cls_token_box": [
5
+ 0,
6
+ 0,
7
+ 0,
8
+ 0
9
+ ],
10
+ "eos_token": "</s>",
11
+ "mask_token": {
12
+ "__type": "AddedToken",
13
+ "content": "<mask>",
14
+ "lstrip": true,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "model_max_length": 1000000000000000019884624838656,
20
+ "name_or_path": "microsoft/layoutxlm-base",
21
+ "only_label_first_subword": true,
22
+ "pad_token": "<pad>",
23
+ "pad_token_box": [
24
+ 0,
25
+ 0,
26
+ 0,
27
+ 0
28
+ ],
29
+ "pad_token_label": -100,
30
+ "sep_token": "</s>",
31
+ "sep_token_box": [
32
+ 1000,
33
+ 1000,
34
+ 1000,
35
+ 1000
36
+ ],
37
+ "sp_model_kwargs": {},
38
+ "special_tokens_map_file": null,
39
+ "tokenizer_class": "LayoutXLMTokenizer",
40
+ "unk_token": "<unk>"
41
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:942191a74f6577f52fb36a5ef1ccabc87aa261c95ac6c3329efb35ecf29ad0dd
3
+ size 3579