Spaces:
Running
Running
João Pedro
commited on
Commit
·
8756989
1
Parent(s):
23057e8
add model files
Browse files
model/layoutlmv3/config.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"attention_probs_dropout_prob": 0.1,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"classifier_dropout": null,
|
5 |
+
"coordinate_size": 128,
|
6 |
+
"eos_token_id": 2,
|
7 |
+
"has_relative_attention_bias": true,
|
8 |
+
"has_spatial_attention_bias": true,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 768,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"input_size": 224,
|
14 |
+
"intermediate_size": 3072,
|
15 |
+
"layer_norm_eps": 1e-5,
|
16 |
+
"max_2d_position_embeddings": 1024,
|
17 |
+
"max_position_embeddings": 514,
|
18 |
+
"max_rel_2d_pos": 256,
|
19 |
+
"max_rel_pos": 128,
|
20 |
+
"model_type": "layoutlmv3",
|
21 |
+
"num_attention_heads": 12,
|
22 |
+
"num_hidden_layers": 12,
|
23 |
+
"num_labels": 13,
|
24 |
+
"id2label": {
|
25 |
+
"0": "budget",
|
26 |
+
"1": "email",
|
27 |
+
"2": "form",
|
28 |
+
"3": "handwritten",
|
29 |
+
"4": "invoice",
|
30 |
+
"5": "language",
|
31 |
+
"6": "letter",
|
32 |
+
"7": "memo",
|
33 |
+
"8": "news article",
|
34 |
+
"9": "questionnaire",
|
35 |
+
"10": "resume",
|
36 |
+
"11": "scientific publication",
|
37 |
+
"12": "specification"
|
38 |
+
},
|
39 |
+
"label2id": {
|
40 |
+
"budget": 0,
|
41 |
+
"email": 1,
|
42 |
+
"form": 2,
|
43 |
+
"handwritten": 3,
|
44 |
+
"invoice": 4,
|
45 |
+
"language": 5,
|
46 |
+
"letter": 6,
|
47 |
+
"memo": 7,
|
48 |
+
"news article": 8,
|
49 |
+
"questionnaire": 9,
|
50 |
+
"resume": 10,
|
51 |
+
"scientific publication": 11,
|
52 |
+
"specification": 12
|
53 |
+
},
|
54 |
+
"pad_token_id": 1,
|
55 |
+
"rel_2d_pos_bins": 64,
|
56 |
+
"rel_pos_bins": 32,
|
57 |
+
"second_input_size": 112,
|
58 |
+
"shape_size": 128,
|
59 |
+
"torch_dtype": "float32",
|
60 |
+
"transformers_version": "4.12.5",
|
61 |
+
"type_vocab_size": 1,
|
62 |
+
"visual_embed": true,
|
63 |
+
"vocab_size": 50265
|
64 |
+
}
|
model/layoutlmv3/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model/layoutlmv3/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3c6bd09290e28867c32adf634a6b36574ff43a9e53539633f5ac540015ac2baf
|
3 |
+
size 501338056
|
model/layoutlmv3/preprocessor_config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"apply_ocr": true,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_resize": true,
|
5 |
+
"feature_extractor_type": "LayoutLMv3FeatureExtractor",
|
6 |
+
"image_mean": [
|
7 |
+
0.5,
|
8 |
+
0.5,
|
9 |
+
0.5
|
10 |
+
],
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"ocr_lang": null,
|
17 |
+
"resample": 2,
|
18 |
+
"size": 224
|
19 |
+
}
|
model/layoutlmv3/tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"errors": "replace", "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": true, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "roberta-base", "tokenizer_class": "LayoutLMv3Tokenizer"}
|
model/layoutlmv3/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|