pierreguillou
commited on
Commit
•
ed8608e
1
Parent(s):
a0eaeee
Training in progress, step 1600
Browse files- checkpoint-1600/config.json +161 -0
- checkpoint-1600/optimizer.pt +3 -0
- checkpoint-1600/pytorch_model.bin +3 -0
- checkpoint-1600/rng_state.pth +3 -0
- checkpoint-1600/scaler.pt +3 -0
- checkpoint-1600/scheduler.pt +3 -0
- checkpoint-1600/sentencepiece.bpe.model +3 -0
- checkpoint-1600/special_tokens_map.json +15 -0
- checkpoint-1600/tokenizer_config.json +40 -0
- checkpoint-1600/trainer_state.json +130 -0
- checkpoint-1600/training_args.bin +3 -0
- pytorch_model.bin +1 -1
- runs/Mar25_06-46-06_ec50091eea4b/events.out.tfevents.1679726789.ec50091eea4b.176.0 +2 -2
checkpoint-1600/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": "4.27.3",
|
159 |
+
"type_vocab_size": 1,
|
160 |
+
"vocab_size": 250002
|
161 |
+
}
|
checkpoint-1600/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e35bda540483272c539c271ae9404b13f3a24eb50f11f5e0e1aa80e1146ea898
|
3 |
+
size 2927175013
|
checkpoint-1600/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4f3bdf8398cba50172f6dc4ec825723a3898bdffa68b76cc5210c569d2a34e0
|
3 |
+
size 1476527911
|
checkpoint-1600/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c7a2c6ad54089ea16ba38c24c1e7b5bf252870ddf951c2955072e8c25e833a48
|
3 |
+
size 14503
|
checkpoint-1600/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5e5567fe28edb1ab30fec226c30ecaf80c8543edb4d3918f998be856615d396c
|
3 |
+
size 559
|
checkpoint-1600/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:566e77325629a9740eb99438e24b74ad8fa13e91d945f5746119b41639d7ce59
|
3 |
+
size 623
|
checkpoint-1600/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-1600/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-1600/tokenizer_config.json
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"only_label_first_subword": true,
|
21 |
+
"pad_token": "<pad>",
|
22 |
+
"pad_token_box": [
|
23 |
+
0,
|
24 |
+
0,
|
25 |
+
0,
|
26 |
+
0
|
27 |
+
],
|
28 |
+
"pad_token_label": -100,
|
29 |
+
"sep_token": "</s>",
|
30 |
+
"sep_token_box": [
|
31 |
+
1000,
|
32 |
+
1000,
|
33 |
+
1000,
|
34 |
+
1000
|
35 |
+
],
|
36 |
+
"sp_model_kwargs": {},
|
37 |
+
"special_tokens_map_file": null,
|
38 |
+
"tokenizer_class": "LayoutXLMTokenizer",
|
39 |
+
"unk_token": "<unk>"
|
40 |
+
}
|
checkpoint-1600/trainer_state.json
ADDED
@@ -0,0 +1,130 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.6604133950171894,
|
3 |
+
"best_model_checkpoint": "DocLayNet/layout-xlm-base-finetuned-DocLayNet-base_paragraphs_ml512-v6/checkpoint-1200",
|
4 |
+
"epoch": 0.8528784648187633,
|
5 |
+
"global_step": 1600,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.11,
|
12 |
+
"eval_accuracy": 0.8841533429944805,
|
13 |
+
"eval_f1": 0.10664094791953704,
|
14 |
+
"eval_loss": 0.4427741765975952,
|
15 |
+
"eval_precision": 0.11536168521462639,
|
16 |
+
"eval_recall": 0.09914602903501281,
|
17 |
+
"eval_runtime": 348.7843,
|
18 |
+
"eval_samples_per_second": 4.607,
|
19 |
+
"eval_steps_per_second": 0.29,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.21,
|
24 |
+
"eval_accuracy": 0.9242655812394248,
|
25 |
+
"eval_f1": 0.44395873404441333,
|
26 |
+
"eval_loss": 0.30403900146484375,
|
27 |
+
"eval_precision": 0.4547734193086154,
|
28 |
+
"eval_recall": 0.4336464560204953,
|
29 |
+
"eval_runtime": 350.1211,
|
30 |
+
"eval_samples_per_second": 4.59,
|
31 |
+
"eval_steps_per_second": 0.288,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.27,
|
36 |
+
"learning_rate": 1.3209054593874834e-05,
|
37 |
+
"loss": 0.7241,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.32,
|
42 |
+
"eval_accuracy": 0.9359434268564053,
|
43 |
+
"eval_f1": 0.5543897040741045,
|
44 |
+
"eval_loss": 0.22647875547409058,
|
45 |
+
"eval_precision": 0.5330233291298865,
|
46 |
+
"eval_recall": 0.5775405636208369,
|
47 |
+
"eval_runtime": 342.3688,
|
48 |
+
"eval_samples_per_second": 4.694,
|
49 |
+
"eval_steps_per_second": 0.295,
|
50 |
+
"step": 600
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.43,
|
54 |
+
"eval_accuracy": 0.947861481323102,
|
55 |
+
"eval_f1": 0.6015281512784394,
|
56 |
+
"eval_loss": 0.2140066921710968,
|
57 |
+
"eval_precision": 0.6013484680378937,
|
58 |
+
"eval_recall": 0.6017079419299743,
|
59 |
+
"eval_runtime": 342.7133,
|
60 |
+
"eval_samples_per_second": 4.689,
|
61 |
+
"eval_steps_per_second": 0.295,
|
62 |
+
"step": 800
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.53,
|
66 |
+
"learning_rate": 1.927439656448986e-05,
|
67 |
+
"loss": 0.2343,
|
68 |
+
"step": 1000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.53,
|
72 |
+
"eval_accuracy": 0.9402091136821377,
|
73 |
+
"eval_f1": 0.6132131856006622,
|
74 |
+
"eval_loss": 0.285193532705307,
|
75 |
+
"eval_precision": 0.6641768748132656,
|
76 |
+
"eval_recall": 0.5695132365499573,
|
77 |
+
"eval_runtime": 341.4863,
|
78 |
+
"eval_samples_per_second": 4.706,
|
79 |
+
"eval_steps_per_second": 0.296,
|
80 |
+
"step": 1000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.64,
|
84 |
+
"eval_accuracy": 0.9539986854504707,
|
85 |
+
"eval_f1": 0.6604133950171894,
|
86 |
+
"eval_loss": 0.16944007575511932,
|
87 |
+
"eval_precision": 0.6564846848367226,
|
88 |
+
"eval_recall": 0.6643894107600341,
|
89 |
+
"eval_runtime": 342.4627,
|
90 |
+
"eval_samples_per_second": 4.692,
|
91 |
+
"eval_steps_per_second": 0.295,
|
92 |
+
"step": 1200
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.75,
|
96 |
+
"eval_accuracy": 0.9354234142916347,
|
97 |
+
"eval_f1": 0.6198065842234727,
|
98 |
+
"eval_loss": 0.230770543217659,
|
99 |
+
"eval_precision": 0.5118828964212166,
|
100 |
+
"eval_recall": 0.785397096498719,
|
101 |
+
"eval_runtime": 332.1227,
|
102 |
+
"eval_samples_per_second": 4.839,
|
103 |
+
"eval_steps_per_second": 0.304,
|
104 |
+
"step": 1400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.8,
|
108 |
+
"learning_rate": 1.7793573226714055e-05,
|
109 |
+
"loss": 0.1913,
|
110 |
+
"step": 1500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.85,
|
114 |
+
"eval_accuracy": 0.9593994250825553,
|
115 |
+
"eval_f1": 0.6589563286454478,
|
116 |
+
"eval_loss": 0.1601293832063675,
|
117 |
+
"eval_precision": 0.7189582071471835,
|
118 |
+
"eval_recall": 0.6081981212638771,
|
119 |
+
"eval_runtime": 327.08,
|
120 |
+
"eval_samples_per_second": 4.913,
|
121 |
+
"eval_steps_per_second": 0.309,
|
122 |
+
"step": 1600
|
123 |
+
}
|
124 |
+
],
|
125 |
+
"max_steps": 7504,
|
126 |
+
"num_train_epochs": 4,
|
127 |
+
"total_flos": 6910321793433600.0,
|
128 |
+
"trial_name": null,
|
129 |
+
"trial_params": null
|
130 |
+
}
|
checkpoint-1600/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e73f2238132d120ef0436e98ad1ae838499d80731c759af8178e8ff167146b05
|
3 |
+
size 3759
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1476527911
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4f3bdf8398cba50172f6dc4ec825723a3898bdffa68b76cc5210c569d2a34e0
|
3 |
size 1476527911
|
runs/Mar25_06-46-06_ec50091eea4b/events.out.tfevents.1679726789.ec50091eea4b.176.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef3de2ee40a33c21a45fb4d623dcc5ceef87a1709d730cf94c8411a8d065563f
|
3 |
+
size 10978
|