update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- precision
|
7 |
+
- recall
|
8 |
+
- f1
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: xlnet-large-cased-ner-food-combined-v2
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# xlnet-large-cased-ner-food-combined-v2
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [xlnet-large-cased](https://huggingface.co/xlnet-large-cased) on the None dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.1884
|
23 |
+
- Precision: 0.8153
|
24 |
+
- Recall: 0.8947
|
25 |
+
- F1: 0.8531
|
26 |
+
- Accuracy: 0.9729
|
27 |
+
|
28 |
+
## Model description
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Intended uses & limitations
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training and evaluation data
|
37 |
+
|
38 |
+
More information needed
|
39 |
+
|
40 |
+
## Training procedure
|
41 |
+
|
42 |
+
### Training hyperparameters
|
43 |
+
|
44 |
+
The following hyperparameters were used during training:
|
45 |
+
- learning_rate: 5e-06
|
46 |
+
- train_batch_size: 8
|
47 |
+
- eval_batch_size: 8
|
48 |
+
- seed: 42
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: linear
|
51 |
+
- num_epochs: 10
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
56 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
57 |
+
| No log | 0.45 | 400 | 0.1389 | 0.7251 | 0.8609 | 0.7872 | 0.9622 |
|
58 |
+
| 0.2073 | 0.9 | 800 | 0.1628 | 0.8309 | 0.8797 | 0.8546 | 0.9747 |
|
59 |
+
| 0.157 | 1.35 | 1200 | 0.1346 | 0.7899 | 0.8888 | 0.8364 | 0.9710 |
|
60 |
+
| 0.1362 | 1.8 | 1600 | 0.1191 | 0.7340 | 0.8880 | 0.8037 | 0.9633 |
|
61 |
+
| 0.1356 | 2.25 | 2000 | 0.1253 | 0.6966 | 0.8888 | 0.7810 | 0.9570 |
|
62 |
+
| 0.1356 | 2.7 | 2400 | 0.1194 | 0.7556 | 0.8855 | 0.8154 | 0.9659 |
|
63 |
+
| 0.1175 | 3.15 | 2800 | 0.1546 | 0.8378 | 0.8880 | 0.8622 | 0.9754 |
|
64 |
+
| 0.1064 | 3.6 | 3200 | 0.1342 | 0.7955 | 0.8909 | 0.8405 | 0.9711 |
|
65 |
+
| 0.1116 | 4.04 | 3600 | 0.1314 | 0.7981 | 0.8984 | 0.8453 | 0.9713 |
|
66 |
+
| 0.0981 | 4.49 | 4000 | 0.1433 | 0.8059 | 0.8834 | 0.8429 | 0.9717 |
|
67 |
+
| 0.0981 | 4.94 | 4400 | 0.1439 | 0.8051 | 0.9026 | 0.8510 | 0.9719 |
|
68 |
+
| 0.0936 | 5.39 | 4800 | 0.1661 | 0.8180 | 0.8943 | 0.8544 | 0.9735 |
|
69 |
+
| 0.082 | 5.84 | 5200 | 0.1558 | 0.8179 | 0.8843 | 0.8498 | 0.9727 |
|
70 |
+
| 0.084 | 6.29 | 5600 | 0.1553 | 0.7918 | 0.8930 | 0.8394 | 0.9699 |
|
71 |
+
| 0.0782 | 6.74 | 6000 | 0.1457 | 0.7817 | 0.8943 | 0.8342 | 0.9684 |
|
72 |
+
| 0.0782 | 7.19 | 6400 | 0.1793 | 0.8134 | 0.8913 | 0.8506 | 0.9726 |
|
73 |
+
| 0.0694 | 7.64 | 6800 | 0.1638 | 0.7974 | 0.8930 | 0.8425 | 0.9707 |
|
74 |
+
| 0.0757 | 8.09 | 7200 | 0.1690 | 0.8042 | 0.8976 | 0.8483 | 0.9714 |
|
75 |
+
| 0.0665 | 8.54 | 7600 | 0.1813 | 0.8110 | 0.8951 | 0.8510 | 0.9724 |
|
76 |
+
| 0.0607 | 8.99 | 8000 | 0.1907 | 0.8226 | 0.8938 | 0.8567 | 0.9738 |
|
77 |
+
| 0.0607 | 9.44 | 8400 | 0.1848 | 0.8062 | 0.8938 | 0.8478 | 0.9719 |
|
78 |
+
| 0.0649 | 9.89 | 8800 | 0.1884 | 0.8153 | 0.8947 | 0.8531 | 0.9729 |
|
79 |
+
|
80 |
+
|
81 |
+
### Framework versions
|
82 |
+
|
83 |
+
- Transformers 4.27.4
|
84 |
+
- Pytorch 2.0.0+cu118
|
85 |
+
- Datasets 2.11.0
|
86 |
+
- Tokenizers 0.13.3
|