ahlem-phantom commited on
Commit
60783bc
·
1 Parent(s): f7eca3b

End of training

Browse files
README.md ADDED
@@ -0,0 +1,109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-sa-4.0
3
+ base_model: microsoft/layoutlmv3-base
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - ls-generated4
8
+ metrics:
9
+ - precision
10
+ - recall
11
+ - f1
12
+ - accuracy
13
+ model-index:
14
+ - name: layoutlmv3-invoice-model
15
+ results:
16
+ - task:
17
+ name: Token Classification
18
+ type: token-classification
19
+ dataset:
20
+ name: ls-generated4
21
+ type: ls-generated4
22
+ config: invoice
23
+ split: test
24
+ args: invoice
25
+ metrics:
26
+ - name: Precision
27
+ type: precision
28
+ value: 0.9185733512786003
29
+ - name: Recall
30
+ type: recall
31
+ value: 0.9375
32
+ - name: F1
33
+ type: f1
34
+ value: 0.9279401767505099
35
+ - name: Accuracy
36
+ type: accuracy
37
+ value: 0.9536870503597122
38
+ ---
39
+
40
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
41
+ should probably proofread and complete it, then remove this comment. -->
42
+
43
+ # layoutlmv3-invoice-model
44
+
45
+ This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the ls-generated4 dataset.
46
+ It achieves the following results on the evaluation set:
47
+ - Loss: 0.3694
48
+ - Precision: 0.9186
49
+ - Recall: 0.9375
50
+ - F1: 0.9279
51
+ - Accuracy: 0.9537
52
+
53
+ ## Model description
54
+
55
+ More information needed
56
+
57
+ ## Intended uses & limitations
58
+
59
+ More information needed
60
+
61
+ ## Training and evaluation data
62
+
63
+ More information needed
64
+
65
+ ## Training procedure
66
+
67
+ ### Training hyperparameters
68
+
69
+ The following hyperparameters were used during training:
70
+ - learning_rate: 5e-05
71
+ - train_batch_size: 2
72
+ - eval_batch_size: 2
73
+ - seed: 42
74
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
75
+ - lr_scheduler_type: linear
76
+ - training_steps: 2000
77
+
78
+ ### Training results
79
+
80
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
81
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
82
+ | No log | 0.85 | 100 | 0.7836 | 0.5238 | 0.5982 | 0.5585 | 0.7680 |
83
+ | No log | 1.69 | 200 | 0.4954 | 0.6888 | 0.7479 | 0.7172 | 0.8422 |
84
+ | No log | 2.54 | 300 | 0.3483 | 0.7807 | 0.8462 | 0.8121 | 0.9040 |
85
+ | No log | 3.39 | 400 | 0.3200 | 0.8113 | 0.8654 | 0.8375 | 0.9125 |
86
+ | 0.5923 | 4.24 | 500 | 0.2775 | 0.8593 | 0.8853 | 0.8721 | 0.9319 |
87
+ | 0.5923 | 5.08 | 600 | 0.2674 | 0.8700 | 0.9052 | 0.8872 | 0.9377 |
88
+ | 0.5923 | 5.93 | 700 | 0.2766 | 0.8739 | 0.9135 | 0.8932 | 0.9386 |
89
+ | 0.5923 | 6.78 | 800 | 0.2641 | 0.8879 | 0.9190 | 0.9031 | 0.9472 |
90
+ | 0.5923 | 7.63 | 900 | 0.2893 | 0.9094 | 0.9238 | 0.9165 | 0.9447 |
91
+ | 0.0802 | 8.47 | 1000 | 0.3369 | 0.9145 | 0.9258 | 0.9201 | 0.9465 |
92
+ | 0.0802 | 9.32 | 1100 | 0.3037 | 0.9043 | 0.9341 | 0.9189 | 0.9505 |
93
+ | 0.0802 | 10.17 | 1200 | 0.3510 | 0.9032 | 0.9231 | 0.9130 | 0.9472 |
94
+ | 0.0802 | 11.02 | 1300 | 0.3224 | 0.9138 | 0.9251 | 0.9195 | 0.9501 |
95
+ | 0.0802 | 11.86 | 1400 | 0.3873 | 0.9133 | 0.9265 | 0.9199 | 0.9456 |
96
+ | 0.0198 | 12.71 | 1500 | 0.3786 | 0.9120 | 0.9327 | 0.9222 | 0.9492 |
97
+ | 0.0198 | 13.56 | 1600 | 0.3807 | 0.9050 | 0.9293 | 0.9170 | 0.9469 |
98
+ | 0.0198 | 14.41 | 1700 | 0.3664 | 0.9088 | 0.9313 | 0.9199 | 0.9510 |
99
+ | 0.0198 | 15.25 | 1800 | 0.3582 | 0.9152 | 0.9341 | 0.9245 | 0.9521 |
100
+ | 0.0198 | 16.1 | 1900 | 0.3736 | 0.9198 | 0.9368 | 0.9282 | 0.9528 |
101
+ | 0.007 | 16.95 | 2000 | 0.3694 | 0.9186 | 0.9375 | 0.9279 | 0.9537 |
102
+
103
+
104
+ ### Framework versions
105
+
106
+ - Transformers 4.35.0.dev0
107
+ - Pytorch 2.0.1+cu118
108
+ - Datasets 2.14.5
109
+ - Tokenizers 0.14.1
added_tokens.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<mask>": 50264,
4
+ "<pad>": 1,
5
+ "<s>": 0,
6
+ "<unk>": 3
7
+ }
config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "B-INVOICENUMBER",
18
+ "1": "B-INVOICEDATE",
19
+ "2": "I-INVOICEDATE",
20
+ "3": "B-DUEDATE",
21
+ "4": "I-DUEDATE",
22
+ "5": "B-VENDORNAME",
23
+ "6": "I-VENDORNAME",
24
+ "7": "B-VENDORADDRESS",
25
+ "8": "I-VENDORADDRESS",
26
+ "9": "B-VENDORPHONE",
27
+ "10": "I-VENDORPHONE",
28
+ "11": "B-VENDOREMAIL",
29
+ "12": "B-CUSTOMERNAME",
30
+ "13": "I-CUSTOMERNAME",
31
+ "14": "B-CUSTOMERADDRESS",
32
+ "15": "I-CUSTOMERADDRESS",
33
+ "16": "B-CUSTOMERPHONE",
34
+ "17": "I-CUSTOMERPHONE",
35
+ "18": "B-CUSTOMEREMAIL",
36
+ "19": "B-PRODUCTNAME",
37
+ "20": "I-PRODUCTNAME",
38
+ "21": "B-PRODUCTPRICE",
39
+ "22": "I-PRODUCTPRICE",
40
+ "23": "B-PRODUCTAMOUNT",
41
+ "24": "I-PRODUCTAMOUNT",
42
+ "25": "B-PRODUCTQTY",
43
+ "26": "B-SUBTOTAL",
44
+ "27": "I-SUBTOTAL",
45
+ "28": "B-TAX",
46
+ "29": "I-TAX",
47
+ "30": "B-TOTAL",
48
+ "31": "I-TOTAL",
49
+ "32": "B-DISCOUNT",
50
+ "33": "O"
51
+ },
52
+ "initializer_range": 0.02,
53
+ "input_size": 224,
54
+ "intermediate_size": 3072,
55
+ "label2id": {
56
+ "B-CUSTOMERADDRESS": 14,
57
+ "B-CUSTOMEREMAIL": 18,
58
+ "B-CUSTOMERNAME": 12,
59
+ "B-CUSTOMERPHONE": 16,
60
+ "B-DISCOUNT": 32,
61
+ "B-DUEDATE": 3,
62
+ "B-INVOICEDATE": 1,
63
+ "B-INVOICENUMBER": 0,
64
+ "B-PRODUCTAMOUNT": 23,
65
+ "B-PRODUCTNAME": 19,
66
+ "B-PRODUCTPRICE": 21,
67
+ "B-PRODUCTQTY": 25,
68
+ "B-SUBTOTAL": 26,
69
+ "B-TAX": 28,
70
+ "B-TOTAL": 30,
71
+ "B-VENDORADDRESS": 7,
72
+ "B-VENDOREMAIL": 11,
73
+ "B-VENDORNAME": 5,
74
+ "B-VENDORPHONE": 9,
75
+ "I-CUSTOMERADDRESS": 15,
76
+ "I-CUSTOMERNAME": 13,
77
+ "I-CUSTOMERPHONE": 17,
78
+ "I-DUEDATE": 4,
79
+ "I-INVOICEDATE": 2,
80
+ "I-PRODUCTAMOUNT": 24,
81
+ "I-PRODUCTNAME": 20,
82
+ "I-PRODUCTPRICE": 22,
83
+ "I-SUBTOTAL": 27,
84
+ "I-TAX": 29,
85
+ "I-TOTAL": 31,
86
+ "I-VENDORADDRESS": 8,
87
+ "I-VENDORNAME": 6,
88
+ "I-VENDORPHONE": 10,
89
+ "O": 33
90
+ },
91
+ "layer_norm_eps": 1e-05,
92
+ "max_2d_position_embeddings": 1024,
93
+ "max_position_embeddings": 514,
94
+ "max_rel_2d_pos": 256,
95
+ "max_rel_pos": 128,
96
+ "model_type": "layoutlmv3",
97
+ "num_attention_heads": 12,
98
+ "num_channels": 3,
99
+ "num_hidden_layers": 12,
100
+ "pad_token_id": 1,
101
+ "patch_size": 16,
102
+ "rel_2d_pos_bins": 64,
103
+ "rel_pos_bins": 32,
104
+ "second_input_size": 112,
105
+ "shape_size": 128,
106
+ "text_embed": true,
107
+ "torch_dtype": "float32",
108
+ "transformers_version": "4.35.0.dev0",
109
+ "type_vocab_size": 1,
110
+ "visual_embed": true,
111
+ "vocab_size": 50265
112
+ }
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:ddd01aef5b6c15e649dad0e216ae75eed3bec657173272bd2f03140a97413bd6
3
+ size 503849389
special_tokens_map.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": "<mask>",
6
+ "pad_token": "<pad>",
7
+ "sep_token": "</s>",
8
+ "unk_token": "<unk>"
9
+ }
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
+ "additional_special_tokens": [],
46
+ "apply_ocr": false,
47
+ "bos_token": "<s>",
48
+ "clean_up_tokenization_spaces": true,
49
+ "cls_token": "<s>",
50
+ "cls_token_box": [
51
+ 0,
52
+ 0,
53
+ 0,
54
+ 0
55
+ ],
56
+ "eos_token": "</s>",
57
+ "errors": "replace",
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 sha256:9c97334c502f7a0b3fa1c843761f2445f13bdc15fbd529f06264cd0866cc5950
3
+ size 4027
vocab.json ADDED
The diff for this file is too large to render. See raw diff