xlm-roberta-large-judiciary-cap-v3

Model description

An xlm-roberta-large model finetuned on multilingual training data containing texts of the judiciary domain labelled with major topic codes from the Comparative Agendas Project.

How to use the model

from transformers import AutoTokenizer, pipeline

tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-large")
pipe = pipeline(
    model="poltextlab/xlm-roberta-large-judiciary-cap-v3",
    task="text-classification",
    tokenizer=tokenizer,
    use_fast=False,
    token="<your_hf_read_only_token>"
)

text = "We will place an immediate 6-month halt on the finance driven closure of beds and wards, and set up an independent audit of needs and facilities."
pipe(text)

Gated access

Due to the gated access, you must pass the token parameter when loading the model. In earlier versions of the Transformers package, you may need to use the use_auth_token parameter instead.

Model performance

The model was evaluated on a test set of 1833 examples (10% of the available data).
Model accuracy is 0.77.

label	precision	recall	f1-score	support
0	0.39	0.26	0.32	34
1	0.82	0.8	0.81	296
2	0.64	0.62	0.63	34
3	0.7	0.78	0.74	9
4	0.87	0.77	0.82	171
5	0.48	0.41	0.44	29
6	0.68	0.65	0.67	20
7	0.79	0.88	0.83	56
8	0.65	0.79	0.71	33
9	0.68	0.8	0.74	81
10	0.88	0.83	0.85	489
11	0.71	0.79	0.75	28
12	0.67	0.89	0.76	9
13	0.77	0.82	0.79	251
14	0.59	0.86	0.7	37
15	0.65	0.62	0.64	24
16	0.64	0.43	0.51	21
17	1	0.14	0.25	7
18	0.59	0.67	0.63	139
19	0.77	0.86	0.81	63
20	0	0	0	2
21	0	0	0	0
macro avg	0.64	0.62	0.61	1833
weighted avg	0.78	0.77	0.77	1833

Inference platform

This model is used by the CAP Babel Machine, an open-source and free natural language processing tool, designed to simplify and speed up projects for comparative research.

Cooperation

Model performance can be significantly improved by extending our training sets. We appreciate every submission of CAP-coded corpora (of any domain and language) at poltextlab{at}poltextlab{dot}com or by using the CAP Babel Machine.

Debugging and issues

This architecture uses the sentencepiece tokenizer. In order to run the model before transformers==4.27 you need to install it manually.

If you encounter a RuntimeError when loading the model using the from_pretrained() method, adding ignore_mismatched_sizes=True should solve the issue.

poltextlab
/

xlm-roberta-large-judiciary-cap-v3

You need to agree to share your contact information to access this model