Hugging Face's logo

pierreguillou
/
layout-xlm-base-finetuned-with-DocLayNet-base-at-linelevel-ml384

Token Classification
Transformers
PyTorch
TensorBoard
layoutlmv2
object-detection
vision
Generated from Trainer
DocLayNet
COCO
PDF
IBM
Financial-Reports
Finance
Manuals
Scientific-Articles
Science
Laws
Law
Regulations
Patents
Government-Tenders
image-segmentation
Eval Results
Model card Files Metrics Community
2
pierreguillou commited on
Commit
f8c1ed3
•
1 Parent(s): da7dcb6

Training in progress, step 300

Browse files
Files changed (19) hide show
  1. checkpoint-300/config.json +161 -0
  2. checkpoint-300/optimizer.pt +3 -0
  3. checkpoint-300/pytorch_model.bin +3 -0
  4. checkpoint-300/rng_state.pth +3 -0
  5. checkpoint-300/scaler.pt +3 -0
  6. checkpoint-300/scheduler.pt +3 -0
  7. checkpoint-300/sentencepiece.bpe.model +3 -0
  8. checkpoint-300/special_tokens_map.json +15 -0
  9. checkpoint-300/tokenizer_config.json +41 -0
  10. checkpoint-300/trainer_state.json +28 -0
  11. checkpoint-300/training_args.bin +3 -0
  12. config.json +161 -0
  13. pytorch_model.bin +3 -0
  14. runs/Mar02_12-49-31_6f61ebd14765/1677761390.9870367/events.out.tfevents.1677761390.6f61ebd14765.170.1 +3 -0
  15. runs/Mar02_12-49-31_6f61ebd14765/events.out.tfevents.1677761390.6f61ebd14765.170.0 +3 -0
  16. sentencepiece.bpe.model +3 -0
  17. special_tokens_map.json +15 -0
  18. tokenizer_config.json +41 -0
  19. training_args.bin +3 -0
checkpoint-300/config.json ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "Caption",
112
+ "1": "Footnote",
113
+ "2": "Formula",
114
+ "3": "List-item",
115
+ "4": "Page-footer",
116
+ "5": "Page-header",
117
+ "6": "Picture",
118
+ "7": "Section-header",
119
+ "8": "Table",
120
+ "9": "Text",
121
+ "10": "Title"
122
+ },
123
+ "image_feature_pool_shape": [
124
+ 7,
125
+ 7,
126
+ 256
127
+ ],
128
+ "initializer_range": 0.02,
129
+ "intermediate_size": 3072,
130
+ "label2id": {
131
+ "Caption": 0,
132
+ "Footnote": 1,
133
+ "Formula": 2,
134
+ "List-item": 3,
135
+ "Page-footer": 4,
136
+ "Page-header": 5,
137
+ "Picture": 6,
138
+ "Section-header": 7,
139
+ "Table": 8,
140
+ "Text": 9,
141
+ "Title": 10
142
+ },
143
+ "layer_norm_eps": 1e-05,
144
+ "max_2d_position_embeddings": 1024,
145
+ "max_position_embeddings": 514,
146
+ "max_rel_2d_pos": 256,
147
+ "max_rel_pos": 128,
148
+ "model_type": "layoutlmv2",
149
+ "num_attention_heads": 12,
150
+ "num_hidden_layers": 12,
151
+ "output_past": true,
152
+ "pad_token_id": 1,
153
+ "rel_2d_pos_bins": 64,
154
+ "rel_pos_bins": 32,
155
+ "shape_size": 128,
156
+ "tokenizer_class": "LayoutXLMTokenizer",
157
+ "torch_dtype": "float32",
158
+ "transformers_version": "4.26.1",
159
+ "type_vocab_size": 1,
160
+ "vocab_size": 250002
161
+ }
checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75b918c198ceff9e5d2a80a21c110fae31da84fd0c30ddafcf693f6e2a39540c
3
+ size 2927175013
checkpoint-300/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:879442aff39518855cc2fbacae334bc828d5d2a1df39a837a572ce36b96cccef
3
+ size 1476527911
checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a6cf308420067c671c2b8c5d4e233965abaa542f1bba82be2e53454be566d39
3
+ size 14503
checkpoint-300/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bcb42b3bf1898edd5501a8a54e2a8aeeb1c707564ea3cf64a8e4a8749b091df
3
+ size 559
checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e509ac3bb008e65ccc6440025c2eee9512a75e4b791a3878d7d9d2e1b31e20f9
3
+ size 623
checkpoint-300/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
checkpoint-300/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
+ }
checkpoint-300/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
+ }
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.13109765966400155,
3
+ "best_model_checkpoint": "DocLayNet/layout-xlm-base-finetuned-DocLayNet-base_lines_ml384-v1/checkpoint-300",
4
+ "epoch": 0.12463647694225176,
5
+ "global_step": 300,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.12,
12
+ "eval_accuracy": 0.8412902144953226,
13
+ "eval_f1": 0.13109765966400155,
14
+ "eval_loss": 0.5185014605522156,
15
+ "eval_precision": 0.14371894960965223,
16
+ "eval_recall": 0.1205141938939475,
17
+ "eval_runtime": 363.2958,
18
+ "eval_samples_per_second": 5.648,
19
+ "eval_steps_per_second": 0.355,
20
+ "step": 300
21
+ }
22
+ ],
23
+ "max_steps": 7221,
24
+ "num_train_epochs": 3,
25
+ "total_flos": 971764002201600.0,
26
+ "trial_name": null,
27
+ "trial_params": null
28
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1abf369559179e8514a3a374fd35ebd67ed07d23a983236704f1d39b8e94622b
3
+ size 3631
config.json ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "Caption",
112
+ "1": "Footnote",
113
+ "2": "Formula",
114
+ "3": "List-item",
115
+ "4": "Page-footer",
116
+ "5": "Page-header",
117
+ "6": "Picture",
118
+ "7": "Section-header",
119
+ "8": "Table",
120
+ "9": "Text",
121
+ "10": "Title"
122
+ },
123
+ "image_feature_pool_shape": [
124
+ 7,
125
+ 7,
126
+ 256
127
+ ],
128
+ "initializer_range": 0.02,
129
+ "intermediate_size": 3072,
130
+ "label2id": {
131
+ "Caption": 0,
132
+ "Footnote": 1,
133
+ "Formula": 2,
134
+ "List-item": 3,
135
+ "Page-footer": 4,
136
+ "Page-header": 5,
137
+ "Picture": 6,
138
+ "Section-header": 7,
139
+ "Table": 8,
140
+ "Text": 9,
141
+ "Title": 10
142
+ },
143
+ "layer_norm_eps": 1e-05,
144
+ "max_2d_position_embeddings": 1024,
145
+ "max_position_embeddings": 514,
146
+ "max_rel_2d_pos": 256,
147
+ "max_rel_pos": 128,
148
+ "model_type": "layoutlmv2",
149
+ "num_attention_heads": 12,
150
+ "num_hidden_layers": 12,
151
+ "output_past": true,
152
+ "pad_token_id": 1,
153
+ "rel_2d_pos_bins": 64,
154
+ "rel_pos_bins": 32,
155
+ "shape_size": 128,
156
+ "tokenizer_class": "LayoutXLMTokenizer",
157
+ "torch_dtype": "float32",
158
+ "transformers_version": "4.26.1",
159
+ "type_vocab_size": 1,
160
+ "vocab_size": 250002
161
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:879442aff39518855cc2fbacae334bc828d5d2a1df39a837a572ce36b96cccef
3
+ size 1476527911
runs/Mar02_12-49-31_6f61ebd14765/1677761390.9870367/events.out.tfevents.1677761390.6f61ebd14765.170.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1204a440edc97aad10d497b71d97367f0b8f6867cba75082ef4f5786e3d8494
3
+ size 5889
runs/Mar02_12-49-31_6f61ebd14765/events.out.tfevents.1677761390.6f61ebd14765.170.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:006f5fd941888d0aafd83a8faf27fcfac78bb60d0a459dadadad8f52985fea37
3
+ size 7079
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
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1abf369559179e8514a3a374fd35ebd67ed07d23a983236704f1d39b8e94622b
3
+ size 3631