metadata

base_model: INSAIT-Institute/BgGPT-7B-Instruct-v0.2
library_name: peft
license: apache-2.0
language:
  - en
tags:
  - propaganda

Model Card for identrics/BG_propaganda_detector

Model Description

Developed by: Identrics
Language: English
License: apache-2.0
Finetuned from model: google-bert/bert-base-cased
Context window : 512 tokens

Model Description

This model consists of a fine-tuned version of google-bert/bert-base-cased for a propaganda detection task. It is effectively a binary classifier, determining wether propaganda is present in the output string. This model was created by Identrics, in the scope of the Wasper project.

Uses

To be used as a binary classifier to identify if propaganda is present in a string containing a comment from a social media site

Example

First install direct dependencies:

pip install transformers torch accelerate

Then the model can be downloaded and used for inference:

from transformers import AutoModelForSequenceClassification, AutoTokenizer

model = AutoModelForSequenceClassification.from_pretrained("identrics/EN_propaganda_detector", num_labels=2)
tokenizer = AutoTokenizer.from_pretrained("identrics/EN_propaganda_detector")

tokens = tokenizer("Our country is the most powerful country in the world!", return_tensors="pt")
output = model(**tokens)
print(output.logits)

Training Details

Trained on a corpus of 200 human-generated comments, augmented with 200 more synthetic comments...

Achieved an f1 score of x%

PEFT 0.11.1