Commit ·
b1a62c7
1
Parent(s): 420d87d
Training in progress, step 500
Browse files- .gitignore +1 -0
- config.json +70 -0
- merges.txt +0 -0
- preprocessor_config.json +27 -0
- pytorch_model.bin +3 -0
- runs/Sep04_10-36-10_bd6f5a0f7455/1693823786.7011194/events.out.tfevents.1693823786.bd6f5a0f7455.526.4 +3 -0
- runs/Sep04_10-36-10_bd6f5a0f7455/events.out.tfevents.1693823786.bd6f5a0f7455.526.3 +3 -0
- runs/Sep04_11-12-16_bd6f5a0f7455/1693825943.883744/events.out.tfevents.1693825943.bd6f5a0f7455.526.6 +3 -0
- runs/Sep04_11-12-16_bd6f5a0f7455/events.out.tfevents.1693825943.bd6f5a0f7455.526.5 +3 -0
- runs/Sep04_11-15-02_bd6f5a0f7455/1693826104.5744925/events.out.tfevents.1693826104.bd6f5a0f7455.30635.1 +3 -0
- runs/Sep04_11-15-02_bd6f5a0f7455/events.out.tfevents.1693826104.bd6f5a0f7455.30635.0 +3 -0
- runs/Sep04_11-16-47_bd6f5a0f7455/1693826216.957724/events.out.tfevents.1693826216.bd6f5a0f7455.30635.3 +3 -0
- runs/Sep04_11-16-47_bd6f5a0f7455/events.out.tfevents.1693826216.bd6f5a0f7455.30635.2 +3 -0
- runs/Sep04_11-40-02_bd6f5a0f7455/1693827616.2034547/events.out.tfevents.1693827616.bd6f5a0f7455.31753.1 +3 -0
- runs/Sep04_11-40-02_bd6f5a0f7455/events.out.tfevents.1693827616.bd6f5a0f7455.31753.0 +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +86 -0
- training_args.bin +3 -0
- vocab.json +0 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
config.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/layoutlmv3-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LayoutLMv3ForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"coordinate_size": 171,
|
| 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": 1024,
|
| 16 |
+
"id2label": {
|
| 17 |
+
"0": "O",
|
| 18 |
+
"1": "Ref",
|
| 19 |
+
"2": "NumFa",
|
| 20 |
+
"3": "Fourniss",
|
| 21 |
+
"4": "DateFa",
|
| 22 |
+
"5": "DateLim",
|
| 23 |
+
"6": "TotalHT",
|
| 24 |
+
"7": "TVA",
|
| 25 |
+
"8": "TotalTTc",
|
| 26 |
+
"9": "unitP",
|
| 27 |
+
"10": "Qt",
|
| 28 |
+
"11": "TVAP",
|
| 29 |
+
"12": "descp"
|
| 30 |
+
},
|
| 31 |
+
"initializer_range": 0.02,
|
| 32 |
+
"input_size": 224,
|
| 33 |
+
"intermediate_size": 4096,
|
| 34 |
+
"label2id": {
|
| 35 |
+
"DateFa": 4,
|
| 36 |
+
"DateLim": 5,
|
| 37 |
+
"Fourniss": 3,
|
| 38 |
+
"NumFa": 2,
|
| 39 |
+
"O": 0,
|
| 40 |
+
"Qt": 10,
|
| 41 |
+
"Ref": 1,
|
| 42 |
+
"TVA": 7,
|
| 43 |
+
"TVAP": 11,
|
| 44 |
+
"TotalHT": 6,
|
| 45 |
+
"TotalTTc": 8,
|
| 46 |
+
"descp": 12,
|
| 47 |
+
"unitP": 9
|
| 48 |
+
},
|
| 49 |
+
"layer_norm_eps": 1e-05,
|
| 50 |
+
"max_2d_position_embeddings": 1024,
|
| 51 |
+
"max_position_embeddings": 514,
|
| 52 |
+
"max_rel_2d_pos": 256,
|
| 53 |
+
"max_rel_pos": 128,
|
| 54 |
+
"model_type": "layoutlmv3",
|
| 55 |
+
"num_attention_heads": 16,
|
| 56 |
+
"num_channels": 3,
|
| 57 |
+
"num_hidden_layers": 24,
|
| 58 |
+
"pad_token_id": 1,
|
| 59 |
+
"patch_size": 16,
|
| 60 |
+
"rel_2d_pos_bins": 64,
|
| 61 |
+
"rel_pos_bins": 32,
|
| 62 |
+
"second_input_size": 112,
|
| 63 |
+
"shape_size": 170,
|
| 64 |
+
"text_embed": true,
|
| 65 |
+
"torch_dtype": "float32",
|
| 66 |
+
"transformers_version": "4.29.2",
|
| 67 |
+
"type_vocab_size": 1,
|
| 68 |
+
"visual_embed": true,
|
| 69 |
+
"vocab_size": 50265
|
| 70 |
+
}
|
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 sha256:6eb7317dc82d597d128447e1cc587fe086d111f75827d4619cce1751aa7dd1df
|
| 3 |
+
size 1428419377
|
runs/Sep04_10-36-10_bd6f5a0f7455/1693823786.7011194/events.out.tfevents.1693823786.bd6f5a0f7455.526.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:133890aebd441bda72ae7e5e79a5ce9a086b623dc03d13d0239d741c5d9c5809
|
| 3 |
+
size 5931
|
runs/Sep04_10-36-10_bd6f5a0f7455/events.out.tfevents.1693823786.bd6f5a0f7455.526.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:874b253554a2b327518ce8f10cc57dc86142eff41b4e824725b7387fff6034c9
|
| 3 |
+
size 6789
|
runs/Sep04_11-12-16_bd6f5a0f7455/1693825943.883744/events.out.tfevents.1693825943.bd6f5a0f7455.526.6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60a881b4ecd7e118a127a2963bbaff572be8c5a74881c583bd4c7b6606151a06
|
| 3 |
+
size 5931
|
runs/Sep04_11-12-16_bd6f5a0f7455/events.out.tfevents.1693825943.bd6f5a0f7455.526.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c26a681b57023d33fd48c044bb7fea24657d07579c92544817b787259ac37b8
|
| 3 |
+
size 4184
|
runs/Sep04_11-15-02_bd6f5a0f7455/1693826104.5744925/events.out.tfevents.1693826104.bd6f5a0f7455.30635.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:140bbc75c3e0bf9e849f946a998722f54fe8328988a681c64fd5b4bdd6ef75bf
|
| 3 |
+
size 5931
|
runs/Sep04_11-15-02_bd6f5a0f7455/events.out.tfevents.1693826104.bd6f5a0f7455.30635.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9f39edce783738b232a15ae4f4ad72f6882180c8b890a84491eb1d5cb010ad2
|
| 3 |
+
size 4845
|
runs/Sep04_11-16-47_bd6f5a0f7455/1693826216.957724/events.out.tfevents.1693826216.bd6f5a0f7455.30635.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:975423d92be39632d1f303900617fdf3ef2f6c54e509af00c7543bec793ec5a2
|
| 3 |
+
size 5931
|
runs/Sep04_11-16-47_bd6f5a0f7455/events.out.tfevents.1693826216.bd6f5a0f7455.30635.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b198621b6bd84455a64e25c820c348c3addd4d9f73b39bd59eb8dcd0caabeaa9
|
| 3 |
+
size 4184
|
runs/Sep04_11-40-02_bd6f5a0f7455/1693827616.2034547/events.out.tfevents.1693827616.bd6f5a0f7455.31753.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d10dc79274a08782f5c78e98ae58bf8bf3c28fa2562eba874f6f8bf2ee7499b
|
| 3 |
+
size 5931
|
runs/Sep04_11-40-02_bd6f5a0f7455/events.out.tfevents.1693827616.bd6f5a0f7455.31753.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:425da90bd64e318596172667d4e340149b6dcf6bca55a53fc327b80c5aa0e6d5
|
| 3 |
+
size 7354
|
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 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"clean_up_tokenization_spaces": true,
|
| 13 |
+
"cls_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"cls_token_box": [
|
| 22 |
+
0,
|
| 23 |
+
0,
|
| 24 |
+
0,
|
| 25 |
+
0
|
| 26 |
+
],
|
| 27 |
+
"eos_token": {
|
| 28 |
+
"__type": "AddedToken",
|
| 29 |
+
"content": "</s>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false
|
| 34 |
+
},
|
| 35 |
+
"errors": "replace",
|
| 36 |
+
"mask_token": {
|
| 37 |
+
"__type": "AddedToken",
|
| 38 |
+
"content": "<mask>",
|
| 39 |
+
"lstrip": true,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"model_max_length": 512,
|
| 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 |
+
"tokenizer_class": "LayoutLMv3Tokenizer",
|
| 77 |
+
"trim_offsets": true,
|
| 78 |
+
"unk_token": {
|
| 79 |
+
"__type": "AddedToken",
|
| 80 |
+
"content": "<unk>",
|
| 81 |
+
"lstrip": false,
|
| 82 |
+
"normalized": true,
|
| 83 |
+
"rstrip": false,
|
| 84 |
+
"single_word": false
|
| 85 |
+
}
|
| 86 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fa01ca6c3fd81e3431e0e8855a17191b67f403ff1ccf3534667a481979ef063
|
| 3 |
+
size 3963
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|