Update README.md
Browse files
README.md
CHANGED
@@ -11,22 +11,57 @@ tags:
|
|
11 |
|
12 |
## National Climate Targets Classifier - Climate Policy Radar
|
13 |
|
14 |
-
A
|
15 |
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
## Getting started
|
18 |
|
19 |
|
|
|
|
|
20 |
|
21 |
-
## Citation information
|
22 |
|
23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
## Links
|
26 |
- __Repository__: [coming soon]
|
27 |
- __Paper__: [coming soon]
|
28 |
|
29 |
|
|
|
|
|
|
|
|
|
30 |
## Authors & Contact
|
31 |
Climate Policy Radar team: Matyas Juhasz, Tina Marchand, Roshan Melwani, Kalyan Dutia, Sarah Goodenough, Harrison Pim, and Henry Franks.
|
32 |
|
|
|
11 |
|
12 |
## National Climate Targets Classifier - Climate Policy Radar
|
13 |
|
14 |
+
A multi-label text-classifier trained on the National Climate Targets dataset by Climate Policy Radar.
|
15 |
|
16 |
+
Using the [climatebert/distilroberta-base-climate-f](https://huggingface.co/climatebert/distilroberta-base-climate-f) model as a starting point, this classifier is trained on the [ClimatePolicyRadar/national-climate-targets](https://huggingface.co/datasets/ClimatePolicyRadar/national-climate-targets) dataset to predict Net Zero ("NZT")
|
17 |
+
, "Reduction" and "Other" targets in a multi-label setting. The training data is an expert annotated subset of national laws, policies and UNFCCC submissions.
|
18 |
+
|
19 |
+
|
20 |
+
For more information on the annotation methodology and classifier training see __our paper TBA__.
|
21 |
|
22 |
## Getting started
|
23 |
|
24 |
|
25 |
+
```python
|
26 |
+
from transformers import AutoModelForSequenceClassification, AutoTokenizer
|
27 |
|
|
|
28 |
|
29 |
+
model_name = "ClimatePolicyRadar/national-climate-targets"
|
30 |
+
example = "The Net Zero Strategy, published in October 2021, was the first "\
|
31 |
+
"document of its kind for a major economy. It set out the government’s "\
|
32 |
+
"vision for a market-led, technology-driven transition to decarbonise "\
|
33 |
+
"the UK economy and reach net zero by 2050."
|
34 |
+
|
35 |
+
|
36 |
+
model = AutoModelForSequenceClassification.from_pretrained(model_name)
|
37 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
38 |
+
|
39 |
+
# using sigmoid because the model is multi-label
|
40 |
+
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer, function_to_apply="sigmoid")
|
41 |
+
|
42 |
+
|
43 |
+
pipe(example)
|
44 |
+
|
45 |
+
>>> [{'label': 'NZT', 'score': 0.9142044186592102}]
|
46 |
+
```
|
47 |
+
|
48 |
+
|
49 |
+
## Licence
|
50 |
+
|
51 |
+
Our classifier is licensed as [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0).
|
52 |
+
|
53 |
+
Please read our Terms of Use, including any specific terms relevant to commercial use. Contact [email protected] with any questions.
|
54 |
+
|
55 |
|
56 |
## Links
|
57 |
- __Repository__: [coming soon]
|
58 |
- __Paper__: [coming soon]
|
59 |
|
60 |
|
61 |
+
## Citation
|
62 |
+
|
63 |
+
[coming soon]
|
64 |
+
|
65 |
## Authors & Contact
|
66 |
Climate Policy Radar team: Matyas Juhasz, Tina Marchand, Roshan Melwani, Kalyan Dutia, Sarah Goodenough, Harrison Pim, and Henry Franks.
|
67 |
|