update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
metrics:
|
5 |
+
- precision
|
6 |
+
- recall
|
7 |
+
- f1
|
8 |
+
- accuracy
|
9 |
+
model-index:
|
10 |
+
- name: layoutlmv1-er-ner
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# layoutlmv1-er-ner
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [renjithks/layoutlmv1-cord-ner](https://huggingface.co/renjithks/layoutlmv1-cord-ner) on an unknown dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.2936
|
22 |
+
- Precision: 0.6097
|
23 |
+
- Recall: 0.6192
|
24 |
+
- F1: 0.6144
|
25 |
+
- Accuracy: 0.9479
|
26 |
+
|
27 |
+
## Model description
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Intended uses & limitations
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training and evaluation data
|
36 |
+
|
37 |
+
More information needed
|
38 |
+
|
39 |
+
## Training procedure
|
40 |
+
|
41 |
+
### Training hyperparameters
|
42 |
+
|
43 |
+
The following hyperparameters were used during training:
|
44 |
+
- learning_rate: 5e-05
|
45 |
+
- train_batch_size: 8
|
46 |
+
- eval_batch_size: 8
|
47 |
+
- seed: 42
|
48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
+
- lr_scheduler_type: linear
|
50 |
+
- num_epochs: 20
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
56 |
+
| No log | 1.0 | 22 | 0.3856 | 0.3047 | 0.1453 | 0.1968 | 0.8885 |
|
57 |
+
| No log | 2.0 | 44 | 0.2637 | 0.3725 | 0.3625 | 0.3674 | 0.9197 |
|
58 |
+
| No log | 3.0 | 66 | 0.2184 | 0.5117 | 0.4612 | 0.4852 | 0.9361 |
|
59 |
+
| No log | 4.0 | 88 | 0.2321 | 0.4714 | 0.5585 | 0.5113 | 0.9361 |
|
60 |
+
| No log | 5.0 | 110 | 0.2183 | 0.5453 | 0.5853 | 0.5646 | 0.9440 |
|
61 |
+
| No log | 6.0 | 132 | 0.2243 | 0.5977 | 0.5867 | 0.5922 | 0.9459 |
|
62 |
+
| No log | 7.0 | 154 | 0.2451 | 0.5716 | 0.5910 | 0.5811 | 0.9410 |
|
63 |
+
| No log | 8.0 | 176 | 0.2387 | 0.5881 | 0.5839 | 0.5860 | 0.9474 |
|
64 |
+
| No log | 9.0 | 198 | 0.2702 | 0.5794 | 0.6023 | 0.5906 | 0.9430 |
|
65 |
+
| No log | 10.0 | 220 | 0.2450 | 0.5920 | 0.6079 | 0.5999 | 0.9480 |
|
66 |
+
| No log | 11.0 | 242 | 0.2697 | 0.6151 | 0.5994 | 0.6071 | 0.9467 |
|
67 |
+
| No log | 12.0 | 264 | 0.2607 | 0.6022 | 0.6234 | 0.6126 | 0.9497 |
|
68 |
+
| No log | 13.0 | 286 | 0.2737 | 0.6172 | 0.6276 | 0.6224 | 0.9488 |
|
69 |
+
| No log | 14.0 | 308 | 0.2840 | 0.6117 | 0.6333 | 0.6223 | 0.9474 |
|
70 |
+
| No log | 15.0 | 330 | 0.2833 | 0.6030 | 0.6192 | 0.6110 | 0.9476 |
|
71 |
+
| No log | 16.0 | 352 | 0.3009 | 0.6161 | 0.6135 | 0.6148 | 0.9449 |
|
72 |
+
| No log | 17.0 | 374 | 0.2920 | 0.6098 | 0.6150 | 0.6124 | 0.9473 |
|
73 |
+
| No log | 18.0 | 396 | 0.2931 | 0.6017 | 0.6135 | 0.6075 | 0.9471 |
|
74 |
+
| No log | 19.0 | 418 | 0.2935 | 0.6103 | 0.6206 | 0.6154 | 0.9476 |
|
75 |
+
| No log | 20.0 | 440 | 0.2936 | 0.6097 | 0.6192 | 0.6144 | 0.9479 |
|
76 |
+
|
77 |
+
|
78 |
+
### Framework versions
|
79 |
+
|
80 |
+
- Transformers 4.18.0
|
81 |
+
- Pytorch 1.11.0
|
82 |
+
- Datasets 2.1.0
|
83 |
+
- Tokenizers 0.12.1
|