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.

Our models are intended for academic projects and academic research only.If you are not affiliated with an academic institution, please reach out to us at huggingface [at] poltextlab [dot] com for further inquiry.If we cannot clearly determine your academic affiliation and use case based on your form data, your request may be rejected. Please allow us a few business days to manually review subscriptions.

xlm-roberta-large-ontolisst-minor-v2

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-minor-v2",
    task="text-classification",
    tokenizer=tokenizer,
    use_fast=False,
    token="<your_hf_read_only_token>"
)

text = "<text_to_classify>"
pipe(text)

Classification Report

Overall Performance:

Accuracy: 75%
Macro Avg: Precision: 0.77, Recall: 0.73, F1-score: 0.73
Weighted Avg: Precision: 0.75, Recall: 0.75, F1-score: 0.74

Per-Class Metrics:

Label	Precision	Recall	F1-score	Support
(201) Housing and accommodation	0.7	0.66	0.68	50
(202) Residential environment and infrastructure	0.85	0.88	0.86	50
(203) Environmental issues and energy consumption	0.73	0.8	0.76	30
(301) Health condition and medical history	0.68	0.78	0.73	50
(302) Health behaviour	0.88	0.9	0.89	50
(303) Health care and social services, infrastructure and policy	0.78	0.78	0.78	49
(304) General wellbeing	0.83	0.83	0.83	12
(305) Informal care work	1	0.2	0.33	5
(401) Work organisation and conditions: organizational level	0.54	0.74	0.63	50
(402) Employment status and work history: individual level	0.49	0.58	0.53	50
(403) Domestic work and work-life balance	1	0.23	0.38	26
(404) Civil and voluntary work	0.85	0.82	0.84	28
(405) Unemployment and job search	0	0	0	13
(501) Educational structure and organizations	0.67	0.72	0.69	50
(502) Educational attainment, history and plan	0.67	0.84	0.74	50
(503) Educational activities	0.8	0.55	0.65	29
(504) Educational needs and support	0.76	0.74	0.75	50
(601) Partnership status and history	0.88	0.92	0.9	50
(602) Sexual orientation and behaviour	0.88	0.93	0.9	15
(603) Family relations and parenting	0.65	0.6	0.62	50
(604) Friendship and social ties	0.78	0.6	0.68	30
(605) Conflicts, cooperation and contact frequency	0.75	0.82	0.78	50
(606) Sexual, domestic and interpersonal violence and abuse	0.96	0.9	0.93	50
(607) Household composition and living arrangements	0.71	0.72	0.71	50
(701) Ethnic and cultural identity	0.78	0.78	0.78	50
(702) Religious identity	0.85	0.88	0.86	50
(703) Political views, participation in national and international institutions and processes	0.65	0.6	0.62	50
(704) Macro-level issues and processes	0.5	0.5	0.5	50
(705) Gender	0.8	0.8	0.8	10
(706) Age, age group and aging	0.89	0.8	0.84	10
(707) Human qualities and personality traits	1	1	1	11
(801) Sports and physical activities	0.96	0.92	0.94	26
(802) Culture and media consumption and activities	0.94	0.9	0.92	50
(803) Leisure activities and hobbies	0.85	0.85	0.85	13
(901) Wages and salaries	0.94	0.78	0.85	37
(902) Benefits and allowances	0.45	0.6	0.51	30
(903) Financial situation	0.7	0.74	0.72	50
(904) Private transfers (voluntary, unrequited)	0.8	0.57	0.67	21
(905) Expenses (personal, family, household)	0.63	0.78	0.7	37
(1001) Crime events and criminal behavior	0.88	0.79	0.84	29
(1002) Legal system and sanctions	0.89	0.98	0.93	50
(1100) TECHNICAL	0.81	0.78	0.8	50
(1200) OTHER	0.78	0.7	0.74	50

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.

Downloads last month: 25

Safetensors

Model size

0.6B params

Tensor type

F32

Model tree for poltextlab/xlm-roberta-large-ontolisst-minor-v2

Base model

FacebookAI/xlm-roberta-large

Finetuned

(1007)

this model

Evaluation results

Accuracy
self-reported

75%
F1-Score
self-reported

74%