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 use only. If you are not affiliated with an academic institution, please provide a rationale for using our models. Please allow us a few business days to manually review subscriptions.
If you use our models for your work or research, please cite this paper: Sebők, M., Máté, Á., Ring, O., Kovács, V., & Lehoczki, R. (2024). Leveraging Open Large Language Models for Multilingual Policy Topic Classification: The Babel Machine Approach. Social Science Computer Review, 0(0). https://doi.org/10.1177/08944393241259434

Log in or Sign Up to review the conditions and access this model content.

xlm-roberta-large-german-parlspeech-cap-v4

Model description

An xlm-roberta-large model fine-tuned on german training data containing parliamentary speeches (oral questions, interpellations, bill debates, other plenary speeches, urgent questions) labeled with major topic codes from the Comparative Agendas Project. To augment the lower-performing labels and the No Policy Content label, we used English parliamentary speeches that were translated into German using OpenNMT. This model is an improved version of xlm-roberta-large-german-parlspeech-cap-v3.

We follow the master codebook of the Comparative Agendas Project, and all of our models use the same major topic codes.

How to use the model

from transformers import AutoTokenizer, pipeline

tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-large")
pipe = pipeline(
    model="poltextlab/xlm-roberta-large-pooled-cap-v4",
    task="text-classification",
    tokenizer=tokenizer,
    use_fast=False,
    token="<your_hf_read_only_token>"
)

text = "Entwicklung der Regionalisierungsmittel fuer den Schienenpersonennahverkehr nach der ersten Revision zum 31. Dezember 1997"
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 1576 examples.
Model accuracy is 0.74. Model performance comparison (v3, v4)

Fine-tuning procedure

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

  • Number of Training Epochs: 10
  • Batch Size: 10
  • Learning Rate: 5e-06
  • Early Stopping: enabled with a patience of 1 epochs

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.

Reference

Sebők, M., Máté, Á., Ring, O., Kovács, V., & Lehoczki, R. (2024). Leveraging Open Large Language Models for Multilingual Policy Topic Classification: The Babel Machine Approach. Social Science Computer Review, 0(0). https://doi.org/10.1177/08944393241259434

Debugging and issues

This architecture uses the sentencepiece tokenizer. In order to use 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.

Downloads last month
1,441
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support