Hugging Face's logo

JoheunKang
/
temp_paddle_model

PaddlePaddle
Model card Files
xet
 Community
JoheunKang commited on
Commit
f2f1526
·
verified ·
1 Parent(s): 990f35c

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. .gitattributes +6 -0
  2. best_accuracy.pdopt +3 -0
  3. best_accuracy/metric.states +0 -0
  4. best_accuracy/model_config.json +45 -0
  5. best_accuracy/model_state.pdparams +3 -0
  6. best_model/model.pdopt +3 -0
  7. best_model/model_config.json +45 -0
  8. best_model/model_state.pdparams +3 -0
  9. config.yml +161 -0
  10. latest.pdopt +3 -0
  11. latest/metric.states +0 -0
  12. latest/model_config.json +45 -0
  13. latest/model_state.pdparams +3 -0
  14. train.log +0 -0
.gitattributes CHANGED
@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ best_accuracy/model_state.pdparams filter=lfs diff=lfs merge=lfs -text
37
+ best_accuracy.pdopt filter=lfs diff=lfs merge=lfs -text
38
+ best_model/model.pdopt filter=lfs diff=lfs merge=lfs -text
39
+ best_model/model_state.pdparams filter=lfs diff=lfs merge=lfs -text
40
+ latest/model_state.pdparams filter=lfs diff=lfs merge=lfs -text
41
+ latest.pdopt filter=lfs diff=lfs merge=lfs -text
best_accuracy.pdopt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e0d10ed58c0f46f4e5b76ff2648146636ce057ba838e95bd18a23d011da2ea0
3
+ size 2224406688
best_accuracy/metric.states ADDED
Binary file (384 Bytes). View file
 
best_accuracy/model_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "num_classes": 79,
3
+ "dropout": null,
4
+ "init_args": [
5
+ {
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "coordinate_size": 128,
9
+ "eos_token_id": 2,
10
+ "fast_qkv": false,
11
+ "gradient_checkpointing": false,
12
+ "has_relative_attention_bias": false,
13
+ "has_spatial_attention_bias": false,
14
+ "has_visual_segment_embedding": true,
15
+ "use_visual_backbone": false,
16
+ "hidden_act": "gelu",
17
+ "hidden_dropout_prob": 0.1,
18
+ "hidden_size": 768,
19
+ "image_feature_pool_shape": [
20
+ 7,
21
+ 7,
22
+ 256
23
+ ],
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 3072,
26
+ "layer_norm_eps": 1e-05,
27
+ "max_2d_position_embeddings": 1024,
28
+ "max_position_embeddings": 514,
29
+ "max_rel_2d_pos": 256,
30
+ "max_rel_pos": 128,
31
+ "model_type": "layoutlmv2",
32
+ "num_attention_heads": 12,
33
+ "num_hidden_layers": 12,
34
+ "output_past": true,
35
+ "pad_token_id": 1,
36
+ "shape_size": 128,
37
+ "rel_2d_pos_bins": 64,
38
+ "rel_pos_bins": 32,
39
+ "type_vocab_size": 1,
40
+ "vocab_size": 250002,
41
+ "init_class": "LayoutXLMModel"
42
+ }
43
+ ],
44
+ "init_class": "LayoutXLMForTokenClassification"
45
+ }
best_accuracy/model_state.pdparams ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc2883904bf8874460c2771be1fd97a9d45950b770216d8ebf18a21a79da1832
3
+ size 1114551884
best_model/model.pdopt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e0d10ed58c0f46f4e5b76ff2648146636ce057ba838e95bd18a23d011da2ea0
3
+ size 2224406688
best_model/model_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "num_classes": 79,
3
+ "dropout": null,
4
+ "init_args": [
5
+ {
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "coordinate_size": 128,
9
+ "eos_token_id": 2,
10
+ "fast_qkv": false,
11
+ "gradient_checkpointing": false,
12
+ "has_relative_attention_bias": false,
13
+ "has_spatial_attention_bias": false,
14
+ "has_visual_segment_embedding": true,
15
+ "use_visual_backbone": false,
16
+ "hidden_act": "gelu",
17
+ "hidden_dropout_prob": 0.1,
18
+ "hidden_size": 768,
19
+ "image_feature_pool_shape": [
20
+ 7,
21
+ 7,
22
+ 256
23
+ ],
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 3072,
26
+ "layer_norm_eps": 1e-05,
27
+ "max_2d_position_embeddings": 1024,
28
+ "max_position_embeddings": 514,
29
+ "max_rel_2d_pos": 256,
30
+ "max_rel_pos": 128,
31
+ "model_type": "layoutlmv2",
32
+ "num_attention_heads": 12,
33
+ "num_hidden_layers": 12,
34
+ "output_past": true,
35
+ "pad_token_id": 1,
36
+ "shape_size": 128,
37
+ "rel_2d_pos_bins": 64,
38
+ "rel_pos_bins": 32,
39
+ "type_vocab_size": 1,
40
+ "vocab_size": 250002,
41
+ "init_class": "LayoutXLMModel"
42
+ }
43
+ ],
44
+ "init_class": "LayoutXLMForTokenClassification"
45
+ }
best_model/model_state.pdparams ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc2883904bf8874460c2771be1fd97a9d45950b770216d8ebf18a21a79da1832
3
+ size 1114551884
config.yml ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Global:
2
+ use_gpu: true
3
+ epoch_num: 200
4
+ log_smooth_window: 10
5
+ print_batch_step: 10
6
+ save_model_dir: ./output/ser_vi_layoutxlm_xfund_zh
7
+ save_epoch_step: 2000
8
+ eval_batch_step:
9
+ - 0
10
+ - 19
11
+ cal_metric_during_train: false
12
+ save_inference_dir: null
13
+ use_visualdl: false
14
+ seed: 2022
15
+ infer_img: ppstructure/docs/kie/input/zh_val_42.jpg
16
+ d2s_train_image_shape:
17
+ - 3
18
+ - 224
19
+ - 224
20
+ save_res_path: /content/PaddleOCR/output/ser_layoutxlm_xfund_zh/res
21
+ kie_rec_model_dir: null
22
+ kie_det_model_dir: null
23
+ amp_custom_white_list:
24
+ - scale
25
+ - concat
26
+ - elementwise_add
27
+ Architecture:
28
+ model_type: kie
29
+ algorithm: LayoutXLM
30
+ Transform: null
31
+ Backbone:
32
+ name: LayoutXLMForSer
33
+ pretrained: true
34
+ checkpoints: null
35
+ mode: vi
36
+ num_classes: 79
37
+ Loss:
38
+ name: VQASerTokenLayoutLMLoss
39
+ num_classes: 79
40
+ key: backbone_out
41
+ Optimizer:
42
+ name: AdamW
43
+ beta1: 0.9
44
+ beta2: 0.999
45
+ lr:
46
+ name: Linear
47
+ learning_rate: 5.0e-05
48
+ epochs: 200
49
+ warmup_epoch: 2
50
+ regularizer:
51
+ name: L2
52
+ factor: 0.0
53
+ PostProcess:
54
+ name: VQASerTokenLayoutLMPostProcess
55
+ class_path: /content/PaddleOCR/dataset/class_list_jmfunsd.txt
56
+ Metric:
57
+ name: VQASerTokenMetric
58
+ main_indicator: hmean
59
+ Train:
60
+ dataset:
61
+ name: SimpleDataSet
62
+ data_dir: /content/PaddleOCR/dataset/training_data/images
63
+ label_file_list:
64
+ - /content/PaddleOCR/dataset/train.json
65
+ ratio_list:
66
+ - 1.0
67
+ transforms:
68
+ - DecodeImage:
69
+ img_mode: RGB
70
+ channel_first: false
71
+ - VQATokenLabelEncode:
72
+ contains_re: false
73
+ algorithm: LayoutXLM
74
+ class_path: /content/PaddleOCR/dataset/class_list_jmfunsd.txt
75
+ use_textline_bbox_info: true
76
+ order_method: tb-yx
77
+ - VQATokenPad:
78
+ max_seq_len: 512
79
+ return_attention_mask: true
80
+ - VQASerTokenChunk:
81
+ max_seq_len: 512
82
+ - Resize:
83
+ size:
84
+ - 224
85
+ - 224
86
+ - NormalizeImage:
87
+ scale: 1
88
+ mean:
89
+ - 123.675
90
+ - 116.28
91
+ - 103.53
92
+ std:
93
+ - 58.395
94
+ - 57.12
95
+ - 57.375
96
+ order: hwc
97
+ - ToCHWImage: null
98
+ - KeepKeys:
99
+ keep_keys:
100
+ - input_ids
101
+ - bbox
102
+ - attention_mask
103
+ - token_type_ids
104
+ - image
105
+ - labels
106
+ loader:
107
+ shuffle: true
108
+ drop_last: false
109
+ batch_size_per_card: 8
110
+ num_workers: 4
111
+ Eval:
112
+ dataset:
113
+ name: SimpleDataSet
114
+ data_dir: /content/PaddleOCR/dataset/testing_data/images
115
+ label_file_list:
116
+ - /content/PaddleOCR/dataset/test.json
117
+ transforms:
118
+ - DecodeImage:
119
+ img_mode: RGB
120
+ channel_first: false
121
+ - VQATokenLabelEncode:
122
+ contains_re: false
123
+ algorithm: LayoutXLM
124
+ class_path: /content/PaddleOCR/dataset/class_list_jmfunsd.txt
125
+ use_textline_bbox_info: true
126
+ order_method: tb-yx
127
+ - VQATokenPad:
128
+ max_seq_len: 512
129
+ return_attention_mask: true
130
+ - VQASerTokenChunk:
131
+ max_seq_len: 512
132
+ - Resize:
133
+ size:
134
+ - 224
135
+ - 224
136
+ - NormalizeImage:
137
+ scale: 1
138
+ mean:
139
+ - 123.675
140
+ - 116.28
141
+ - 103.53
142
+ std:
143
+ - 58.395
144
+ - 57.12
145
+ - 57.375
146
+ order: hwc
147
+ - ToCHWImage: null
148
+ - KeepKeys:
149
+ keep_keys:
150
+ - input_ids
151
+ - bbox
152
+ - attention_mask
153
+ - token_type_ids
154
+ - image
155
+ - labels
156
+ loader:
157
+ shuffle: false
158
+ drop_last: false
159
+ batch_size_per_card: 8
160
+ num_workers: 4
161
+ profiler_options: null
latest.pdopt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:521df5a5ebcf90aa3402b2370452cce35023c6381d7e92fed9b99e1942462e6a
3
+ size 2224406688
latest/metric.states ADDED
Binary file (384 Bytes). View file
 
latest/model_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "num_classes": 79,
3
+ "dropout": null,
4
+ "init_args": [
5
+ {
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "coordinate_size": 128,
9
+ "eos_token_id": 2,
10
+ "fast_qkv": false,
11
+ "gradient_checkpointing": false,
12
+ "has_relative_attention_bias": false,
13
+ "has_spatial_attention_bias": false,
14
+ "has_visual_segment_embedding": true,
15
+ "use_visual_backbone": false,
16
+ "hidden_act": "gelu",
17
+ "hidden_dropout_prob": 0.1,
18
+ "hidden_size": 768,
19
+ "image_feature_pool_shape": [
20
+ 7,
21
+ 7,
22
+ 256
23
+ ],
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 3072,
26
+ "layer_norm_eps": 1e-05,
27
+ "max_2d_position_embeddings": 1024,
28
+ "max_position_embeddings": 514,
29
+ "max_rel_2d_pos": 256,
30
+ "max_rel_pos": 128,
31
+ "model_type": "layoutlmv2",
32
+ "num_attention_heads": 12,
33
+ "num_hidden_layers": 12,
34
+ "output_past": true,
35
+ "pad_token_id": 1,
36
+ "shape_size": 128,
37
+ "rel_2d_pos_bins": 64,
38
+ "rel_pos_bins": 32,
39
+ "type_vocab_size": 1,
40
+ "vocab_size": 250002,
41
+ "init_class": "LayoutXLMModel"
42
+ }
43
+ ],
44
+ "init_class": "LayoutXLMForTokenClassification"
45
+ }
latest/model_state.pdparams ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25a3c91294ea276505641f8f223f66e6006e8860ee50715b1d16d977bedb799c
3
+ size 1114551884
train.log ADDED
The diff for this file is too large to render. See raw diff