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

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

This model is not accepting new access requests at the moment. Access is still available for our accepted users, which requires the gated access setting to stay active. For inquiries or custom project requests, please contact us at miklos[dot]sebok[at]poltextlab[dot]com.

Model description

An xlm-roberta-large model fine-tuned on UK survey questions (in English) labeled with the 16 major topic codes from the CLOSER codecook https://ucldata.atlassian.net/wiki/spaces/CLOS/pages/37323020/Topics.

Fine-tuning procedure

This model was fine-tuned with the following key hyperparameters:

Number of Training Epochs: 10
Batch Size: 8
Learning Rate: 5e-06

How to Use the Model

from transformers import AutoTokenizer, pipeline

tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-large")
pipe = pipeline(
    model="poltextlab/xlm-roberta-large_ontolisst_v1",
    task="text-classification",
    tokenizer=tokenizer,
    use_fast=False,
    token="<your_hf_read_only_token>"
)

text = "Apart from yourself (and your husband/wife/partner), does anyone else living in your household make a contribution towards the cost of the accommodation?
pipe(text)

Gated Access

This model requires 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 4,216 English examples.

Accuracy: 0.88

Precision: 0.88

Recall: 0.88

Weighted Average F1-score: 0.88

Unnamed: 0	precision	recall	f1-score	support
0	0.88	0.86	0.87	111.00
1	0.88	0.82	0.85	207.00
2	0.89	0.89	0.89	789.00
3	0.89	0.90	0.90	504.00
4	0.67	0.73	0.70	98.00
5	0.89	0.93	0.91	297.00
6	0.90	0.89	0.89	416.00
7	0.90	0.89	0.89	301.00
8	0.91	0.91	0.91	426.00
9	0.77	0.81	0.79	79.00
10	0.90	0.93	0.91	244.00
11	0.86	0.92	0.89	48.00
12	0.00	0.00	0.00	1.00
13	0.89	0.69	0.78	61.00
14	0.76	0.75	0.76	113.00
15	0.91	0.89	0.90	521.00
accuracy			0.88	4216.00
macro avg	0.81	0.80	0.80	4216.00
weighted avg	0.88	0.88	0.88	4216.00

Downloads last month: 28

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collection including poltextlab/xlm-roberta-large_ontolisst_v1

ML12

Collection

11 items • Updated Mar 25