Jyotiyadav
commited on
Commit
•
edcbc47
1
Parent(s):
9067370
End of training
Browse files- .gitignore +1 -0
- config.json +136 -0
- merges.txt +0 -0
- preprocessor_config.json +27 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +87 -0
- training_args.bin +3 -0
- vocab.json +0 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,136 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/layoutlmv3-base",
|
3 |
+
"architectures": [
|
4 |
+
"LayoutLMv3ForTokenClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"coordinate_size": 128,
|
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": 768,
|
16 |
+
"id2label": {
|
17 |
+
"0": "article_key",
|
18 |
+
"1": "article_value",
|
19 |
+
"2": "as_key",
|
20 |
+
"3": "as_value",
|
21 |
+
"4": "codicefiscale_key",
|
22 |
+
"5": "codicefiscale_value",
|
23 |
+
"6": "colli_key",
|
24 |
+
"7": "colli_value",
|
25 |
+
"8": "descrizione_key",
|
26 |
+
"9": "descrizione_value",
|
27 |
+
"10": "doctype_key",
|
28 |
+
"11": "doctype_value",
|
29 |
+
"12": "header",
|
30 |
+
"13": "iso_key",
|
31 |
+
"14": "iso_value",
|
32 |
+
"15": "iva_key",
|
33 |
+
"16": "iva_value",
|
34 |
+
"17": "importo_key",
|
35 |
+
"18": "importo_value",
|
36 |
+
"19": "lordo_key",
|
37 |
+
"20": "lordo_value",
|
38 |
+
"21": "lotto_key",
|
39 |
+
"22": "lotto_value",
|
40 |
+
"23": "netto_key",
|
41 |
+
"24": "netto_value",
|
42 |
+
"25": "pag._key",
|
43 |
+
"26": "pag._value",
|
44 |
+
"27": "pag1_key",
|
45 |
+
"28": "pag1_value",
|
46 |
+
"29": "partita iva_key",
|
47 |
+
"30": "partita iva_value",
|
48 |
+
"31": "prezzo_key",
|
49 |
+
"32": "prezzo_value",
|
50 |
+
"33": "reciever",
|
51 |
+
"34": "rif_key",
|
52 |
+
"35": "rif_value",
|
53 |
+
"36": "senderaddress",
|
54 |
+
"37": "spett.le_key",
|
55 |
+
"38": "spett.le_value",
|
56 |
+
"39": "tara_key",
|
57 |
+
"40": "tara_value",
|
58 |
+
"41": "transporto_key",
|
59 |
+
"42": "transporto_value",
|
60 |
+
"43": "um_key",
|
61 |
+
"44": "um_value",
|
62 |
+
"45": "others"
|
63 |
+
},
|
64 |
+
"initializer_range": 0.02,
|
65 |
+
"input_size": 224,
|
66 |
+
"intermediate_size": 3072,
|
67 |
+
"label2id": {
|
68 |
+
"article_key": 0,
|
69 |
+
"article_value": 1,
|
70 |
+
"as_key": 2,
|
71 |
+
"as_value": 3,
|
72 |
+
"codicefiscale_key": 4,
|
73 |
+
"codicefiscale_value": 5,
|
74 |
+
"colli_key": 6,
|
75 |
+
"colli_value": 7,
|
76 |
+
"descrizione_key": 8,
|
77 |
+
"descrizione_value": 9,
|
78 |
+
"doctype_key": 10,
|
79 |
+
"doctype_value": 11,
|
80 |
+
"header": 12,
|
81 |
+
"importo_key": 17,
|
82 |
+
"importo_value": 18,
|
83 |
+
"iso_key": 13,
|
84 |
+
"iso_value": 14,
|
85 |
+
"iva_key": 15,
|
86 |
+
"iva_value": 16,
|
87 |
+
"lordo_key": 19,
|
88 |
+
"lordo_value": 20,
|
89 |
+
"lotto_key": 21,
|
90 |
+
"lotto_value": 22,
|
91 |
+
"netto_key": 23,
|
92 |
+
"netto_value": 24,
|
93 |
+
"others": 45,
|
94 |
+
"pag._key": 25,
|
95 |
+
"pag._value": 26,
|
96 |
+
"pag1_key": 27,
|
97 |
+
"pag1_value": 28,
|
98 |
+
"partita iva_key": 29,
|
99 |
+
"partita iva_value": 30,
|
100 |
+
"prezzo_key": 31,
|
101 |
+
"prezzo_value": 32,
|
102 |
+
"reciever": 33,
|
103 |
+
"rif_key": 34,
|
104 |
+
"rif_value": 35,
|
105 |
+
"senderaddress": 36,
|
106 |
+
"spett.le_key": 37,
|
107 |
+
"spett.le_value": 38,
|
108 |
+
"tara_key": 39,
|
109 |
+
"tara_value": 40,
|
110 |
+
"transporto_key": 41,
|
111 |
+
"transporto_value": 42,
|
112 |
+
"um_key": 43,
|
113 |
+
"um_value": 44
|
114 |
+
},
|
115 |
+
"layer_norm_eps": 1e-05,
|
116 |
+
"max_2d_position_embeddings": 1024,
|
117 |
+
"max_position_embeddings": 514,
|
118 |
+
"max_rel_2d_pos": 256,
|
119 |
+
"max_rel_pos": 128,
|
120 |
+
"model_type": "layoutlmv3",
|
121 |
+
"num_attention_heads": 12,
|
122 |
+
"num_channels": 3,
|
123 |
+
"num_hidden_layers": 12,
|
124 |
+
"pad_token_id": 1,
|
125 |
+
"patch_size": 16,
|
126 |
+
"rel_2d_pos_bins": 64,
|
127 |
+
"rel_pos_bins": 32,
|
128 |
+
"second_input_size": 112,
|
129 |
+
"shape_size": 128,
|
130 |
+
"text_embed": true,
|
131 |
+
"torch_dtype": "float32",
|
132 |
+
"transformers_version": "4.26.0.dev0",
|
133 |
+
"type_vocab_size": 1,
|
134 |
+
"visual_embed": true,
|
135 |
+
"vocab_size": 50265
|
136 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
preprocessor_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"apply_ocr": false,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_rescale": true,
|
5 |
+
"do_resize": true,
|
6 |
+
"feature_extractor_type": "LayoutLMv3FeatureExtractor",
|
7 |
+
"image_mean": [
|
8 |
+
0.5,
|
9 |
+
0.5,
|
10 |
+
0.5
|
11 |
+
],
|
12 |
+
"image_processor_type": "LayoutLMv3ImageProcessor",
|
13 |
+
"image_std": [
|
14 |
+
0.5,
|
15 |
+
0.5,
|
16 |
+
0.5
|
17 |
+
],
|
18 |
+
"ocr_lang": null,
|
19 |
+
"processor_class": "LayoutLMv3Processor",
|
20 |
+
"resample": 2,
|
21 |
+
"rescale_factor": 0.00392156862745098,
|
22 |
+
"size": {
|
23 |
+
"height": 224,
|
24 |
+
"width": 224
|
25 |
+
},
|
26 |
+
"tesseract_config": ""
|
27 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:869228b02f818e0e389d8c3e21206041f333e85c92d2bf6c5f9f9092c1366452
|
3 |
+
size 503890737
|
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 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": true,
|
3 |
+
"apply_ocr": false,
|
4 |
+
"bos_token": {
|
5 |
+
"__type": "AddedToken",
|
6 |
+
"content": "<s>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": true,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false
|
11 |
+
},
|
12 |
+
"cls_token": {
|
13 |
+
"__type": "AddedToken",
|
14 |
+
"content": "<s>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": true,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"cls_token_box": [
|
21 |
+
0,
|
22 |
+
0,
|
23 |
+
0,
|
24 |
+
0
|
25 |
+
],
|
26 |
+
"eos_token": {
|
27 |
+
"__type": "AddedToken",
|
28 |
+
"content": "</s>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": true,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
},
|
34 |
+
"errors": "replace",
|
35 |
+
"mask_token": {
|
36 |
+
"__type": "AddedToken",
|
37 |
+
"content": "<mask>",
|
38 |
+
"lstrip": true,
|
39 |
+
"normalized": true,
|
40 |
+
"rstrip": false,
|
41 |
+
"single_word": false
|
42 |
+
},
|
43 |
+
"model_max_length": 512,
|
44 |
+
"name_or_path": "microsoft/layoutlmv3-base",
|
45 |
+
"only_label_first_subword": true,
|
46 |
+
"pad_token": {
|
47 |
+
"__type": "AddedToken",
|
48 |
+
"content": "<pad>",
|
49 |
+
"lstrip": false,
|
50 |
+
"normalized": true,
|
51 |
+
"rstrip": false,
|
52 |
+
"single_word": false
|
53 |
+
},
|
54 |
+
"pad_token_box": [
|
55 |
+
0,
|
56 |
+
0,
|
57 |
+
0,
|
58 |
+
0
|
59 |
+
],
|
60 |
+
"pad_token_label": -100,
|
61 |
+
"processor_class": "LayoutLMv3Processor",
|
62 |
+
"sep_token": {
|
63 |
+
"__type": "AddedToken",
|
64 |
+
"content": "</s>",
|
65 |
+
"lstrip": false,
|
66 |
+
"normalized": true,
|
67 |
+
"rstrip": false,
|
68 |
+
"single_word": false
|
69 |
+
},
|
70 |
+
"sep_token_box": [
|
71 |
+
0,
|
72 |
+
0,
|
73 |
+
0,
|
74 |
+
0
|
75 |
+
],
|
76 |
+
"special_tokens_map_file": null,
|
77 |
+
"tokenizer_class": "LayoutLMv3Tokenizer",
|
78 |
+
"trim_offsets": true,
|
79 |
+
"unk_token": {
|
80 |
+
"__type": "AddedToken",
|
81 |
+
"content": "<unk>",
|
82 |
+
"lstrip": false,
|
83 |
+
"normalized": true,
|
84 |
+
"rstrip": false,
|
85 |
+
"single_word": false
|
86 |
+
}
|
87 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b0d2c917d4ab65683bcca4d1c1f24b67bc7b4d7413f09f8fdf23f539581b5a35
|
3 |
+
size 3451
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|