update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- wikiann
|
7 |
+
metrics:
|
8 |
+
- precision
|
9 |
+
- recall
|
10 |
+
- f1
|
11 |
+
- accuracy
|
12 |
+
model-index:
|
13 |
+
- name: roberta-large_ner_wikiann
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Token Classification
|
17 |
+
type: token-classification
|
18 |
+
dataset:
|
19 |
+
name: wikiann
|
20 |
+
type: wikiann
|
21 |
+
args: en
|
22 |
+
metrics:
|
23 |
+
- name: Precision
|
24 |
+
type: precision
|
25 |
+
value: 0.8462551098177787
|
26 |
+
- name: Recall
|
27 |
+
type: recall
|
28 |
+
value: 0.8634242895518167
|
29 |
+
- name: F1
|
30 |
+
type: f1
|
31 |
+
value: 0.8547534903250638
|
32 |
+
- name: Accuracy
|
33 |
+
type: accuracy
|
34 |
+
value: 0.9382388000397338
|
35 |
+
---
|
36 |
+
|
37 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
38 |
+
should probably proofread and complete it, then remove this comment. -->
|
39 |
+
|
40 |
+
# roberta-large_ner_wikiann
|
41 |
+
|
42 |
+
This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the wikiann dataset.
|
43 |
+
It achieves the following results on the evaluation set:
|
44 |
+
- Loss: 0.2783
|
45 |
+
- Precision: 0.8463
|
46 |
+
- Recall: 0.8634
|
47 |
+
- F1: 0.8548
|
48 |
+
- Accuracy: 0.9382
|
49 |
+
|
50 |
+
## Model description
|
51 |
+
|
52 |
+
More information needed
|
53 |
+
|
54 |
+
## Intended uses & limitations
|
55 |
+
|
56 |
+
More information needed
|
57 |
+
|
58 |
+
## Training and evaluation data
|
59 |
+
|
60 |
+
More information needed
|
61 |
+
|
62 |
+
## Training procedure
|
63 |
+
|
64 |
+
### Training hyperparameters
|
65 |
+
|
66 |
+
The following hyperparameters were used during training:
|
67 |
+
- learning_rate: 2e-05
|
68 |
+
- train_batch_size: 16
|
69 |
+
- eval_batch_size: 16
|
70 |
+
- seed: 42
|
71 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
72 |
+
- lr_scheduler_type: cosine
|
73 |
+
- num_epochs: 5
|
74 |
+
|
75 |
+
### Training results
|
76 |
+
|
77 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
78 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
79 |
+
| 0.3395 | 1.0 | 1250 | 0.2652 | 0.8039 | 0.8308 | 0.8171 | 0.9242 |
|
80 |
+
| 0.2343 | 2.0 | 2500 | 0.2431 | 0.8354 | 0.8503 | 0.8428 | 0.9329 |
|
81 |
+
| 0.1721 | 3.0 | 3750 | 0.2315 | 0.8330 | 0.8503 | 0.8416 | 0.9352 |
|
82 |
+
| 0.1156 | 4.0 | 5000 | 0.2709 | 0.8477 | 0.8634 | 0.8554 | 0.9385 |
|
83 |
+
| 0.1026 | 5.0 | 6250 | 0.2783 | 0.8463 | 0.8634 | 0.8548 | 0.9382 |
|
84 |
+
|
85 |
+
|
86 |
+
### Framework versions
|
87 |
+
|
88 |
+
- Transformers 4.20.1
|
89 |
+
- Pytorch 1.11.0
|
90 |
+
- Datasets 2.1.0
|
91 |
+
- Tokenizers 0.12.1
|