Update readme.py
Browse files
readme.py
CHANGED
@@ -1,10 +1,26 @@
|
|
1 |
import os
|
2 |
from typing import Dict
|
3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
|
5 |
def get_readme(model_name: str,
|
6 |
metric: Dict,
|
7 |
-
metric_span: Dict,
|
8 |
config: Dict):
|
9 |
language_model = config['model']
|
10 |
dataset = None
|
@@ -23,100 +39,51 @@ def get_readme(model_name: str,
|
|
23 |
dataset_link = ','.join([f"[{d}](https://huggingface.co/datasets/{d})" for d in dataset])
|
24 |
return f"""---
|
25 |
datasets:
|
26 |
-
-
|
27 |
metrics:
|
28 |
- f1
|
29 |
-
-
|
30 |
-
- recall
|
31 |
model-index:
|
32 |
- name: {model_name}
|
33 |
results:
|
34 |
- task:
|
35 |
-
|
36 |
-
|
37 |
dataset:
|
38 |
-
name:
|
39 |
-
type:
|
40 |
-
args:
|
|
|
41 |
metrics:
|
42 |
- name: F1
|
43 |
type: f1
|
44 |
-
value: {metric['
|
45 |
-
- name: Precision
|
46 |
-
type: precision
|
47 |
-
value: {metric['micro/precision']}
|
48 |
-
- name: Recall
|
49 |
-
type: recall
|
50 |
-
value: {metric['micro/recall']}
|
51 |
- name: F1 (macro)
|
52 |
type: f1_macro
|
53 |
-
value: {metric['
|
54 |
-
- name:
|
55 |
-
type:
|
56 |
-
value: {metric['
|
57 |
-
|
58 |
-
type: recall_macro
|
59 |
-
value: {metric['macro/recall']}
|
60 |
-
- name: F1 (entity span)
|
61 |
-
type: f1_entity_span
|
62 |
-
value: {metric_span['micro/f1']}
|
63 |
-
- name: Precision (entity span)
|
64 |
-
type: precision_entity_span
|
65 |
-
value: {metric_span['micro/precision']}
|
66 |
-
- name: Recall (entity span)
|
67 |
-
type: recall_entity_span
|
68 |
-
value: {metric_span['micro/recall']}
|
69 |
-
|
70 |
-
pipeline_tag: token-classification
|
71 |
widget:
|
72 |
-
- text: "
|
73 |
-
example_title: "
|
|
|
|
|
74 |
---
|
75 |
# {model_name}
|
76 |
|
77 |
-
This model is a fine-tuned version of [{language_model}](https://huggingface.co/{language_model}) on the
|
78 |
-
{dataset_link} dataset.
|
79 |
-
Model fine-tuning is done via [T-NER](https://github.com/asahi417/tner)'s hyper-parameter search (see the repository
|
80 |
-
for more detail). It achieves the following results on the test set:
|
81 |
-
- F1 (micro): {metric['micro/f1']}
|
82 |
-
- Precision (micro): {metric['micro/precision']}
|
83 |
-
- Recall (micro): {metric['micro/recall']}
|
84 |
-
- F1 (macro): {metric['macro/f1']}
|
85 |
-
- Precision (macro): {metric['macro/precision']}
|
86 |
-
- Recall (macro): {metric['macro/recall']}
|
87 |
|
88 |
-
|
89 |
-
{
|
|
|
90 |
|
91 |
-
For F1 scores, the confidence interval is obtained by bootstrap as below:
|
92 |
-
- F1 (micro):
|
93 |
-
{ci_micro}
|
94 |
-
- F1 (macro):
|
95 |
-
{ci_macro}
|
96 |
-
|
97 |
-
Full evaluation can be found at [metric file of NER](https://huggingface.co/{model_name}/raw/main/eval/metric.json)
|
98 |
-
and [metric file of entity span](https://huggingface.co/{model_name}/raw/main/eval/metric_span.json).
|
99 |
|
100 |
### Usage
|
101 |
-
This model can be used through the [tner library](https://github.com/asahi417/tner). Install the library via pip
|
102 |
-
```shell
|
103 |
-
pip install tner
|
104 |
-
```
|
105 |
-
and activate model as below.
|
106 |
```python
|
107 |
-
|
108 |
-
model = TransformersNER("{model_name}")
|
109 |
-
model.predict(["Jacob Collier is a Grammy awarded English artist from London"])
|
110 |
```
|
111 |
-
It can be used via transformers library but it is not recommended as CRF layer is not supported at the moment.
|
112 |
-
|
113 |
-
### Training hyperparameters
|
114 |
-
|
115 |
-
The following hyperparameters were used during training:
|
116 |
-
{config_text}
|
117 |
-
|
118 |
-
The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/{model_name}/raw/main/trainer_config.json).
|
119 |
-
|
120 |
### Reference
|
121 |
If you use any resource from T-NER, please consider to cite our [paper](https://aclanthology.org/2021.eacl-demos.7/).
|
122 |
|
|
|
1 |
import os
|
2 |
from typing import Dict
|
3 |
|
4 |
+
bib = """
|
5 |
+
@inproceedings{dimosthenis-etal-2022-twitter,
|
6 |
+
title = "{T}witter {T}opic {C}lassification",
|
7 |
+
author = "Antypas, Dimosthenis and
|
8 |
+
Ushio, Asahi and
|
9 |
+
Camacho-Collados, Jose and
|
10 |
+
Neves, Leonardo and
|
11 |
+
Silva, Vitor and
|
12 |
+
Barbieri, Francesco",
|
13 |
+
booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
|
14 |
+
month = oct,
|
15 |
+
year = "2022",
|
16 |
+
address = "Gyeongju, Republic of Korea",
|
17 |
+
publisher = "International Committee on Computational Linguistics"
|
18 |
+
}
|
19 |
+
"""
|
20 |
+
|
21 |
|
22 |
def get_readme(model_name: str,
|
23 |
metric: Dict,
|
|
|
24 |
config: Dict):
|
25 |
language_model = config['model']
|
26 |
dataset = None
|
|
|
39 |
dataset_link = ','.join([f"[{d}](https://huggingface.co/datasets/{d})" for d in dataset])
|
40 |
return f"""---
|
41 |
datasets:
|
42 |
+
- cardiffnlp/tweet_topic_multi
|
43 |
metrics:
|
44 |
- f1
|
45 |
+
- accuracy
|
|
|
46 |
model-index:
|
47 |
- name: {model_name}
|
48 |
results:
|
49 |
- task:
|
50 |
+
type: text-classification
|
51 |
+
name: Text Classification
|
52 |
dataset:
|
53 |
+
name: cardiffnlp/tweet_topic_multi
|
54 |
+
type: cardiffnlp/tweet_topic_multi
|
55 |
+
args: cardiffnlp/tweet_topic_multi
|
56 |
+
split: test_2021
|
57 |
metrics:
|
58 |
- name: F1
|
59 |
type: f1
|
60 |
+
value: {metric['test/eval_f1']}
|
|
|
|
|
|
|
|
|
|
|
|
|
61 |
- name: F1 (macro)
|
62 |
type: f1_macro
|
63 |
+
value: {metric['test/eval_f1_macro']}
|
64 |
+
- name: Accuracy
|
65 |
+
type: accuracy
|
66 |
+
value: {metric['test/eval_accuracy']}
|
67 |
+
pipeline_tag: text-classification
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
widget:
|
69 |
+
- text: "I'm sure the {"{@Tampa Bay Lightning@}"} would’ve rather faced the Flyers but man does their experience versus the Blue Jackets this year and last help them a lot versus this Islanders team. Another meat grinder upcoming for the good guys"
|
70 |
+
example_title: "Example 1"
|
71 |
+
- text: "Love to take night time bike rides at the jersey shore. Seaside Heights boardwalk. Beautiful weather. Wishing everyone a safe Labor Day weekend in the US."
|
72 |
+
example_title: "Example 2"
|
73 |
---
|
74 |
# {model_name}
|
75 |
|
76 |
+
This model is a fine-tuned version of [{language_model}](https://huggingface.co/{language_model}) on the [tweet_topic_multi](https://huggingface.co/datasets/cardiffnlp/tweet_topic_multi). Fine-tuning script can be found [here](https://huggingface.co/datasets/cardiffnlp/tweet_topic_multi/blob/main/lm_finetuning.py). It achieves the following results on the test_2021 set:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
77 |
|
78 |
+
- F1 (micro): {metric['test/eval_f1']}
|
79 |
+
- F1 (macro): {metric['test/eval_f1_macro']}
|
80 |
+
- Accuracy: {metric['test/eval_accuracy']}
|
81 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
82 |
|
83 |
### Usage
|
|
|
|
|
|
|
|
|
|
|
84 |
```python
|
85 |
+
pipe = pipeline("text-classification", "cardiffnlp/twitter-roberta-base-dec2020-tweet-topic-multi-2020", problem_type="multi_label_classification")
|
|
|
|
|
86 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
87 |
### Reference
|
88 |
If you use any resource from T-NER, please consider to cite our [paper](https://aclanthology.org/2021.eacl-demos.7/).
|
89 |
|