metadata

language:
  - en
license: lgpl-3.0
library_name: transformers
tags:
  - text-classification
  - transformers
  - pytorch
  - generated_from_keras_callback
datasets:
  - m-newhauser/senator-tweets
metrics:
  - accuracy
  - f1
widget:
  - text: >-
      This pandemic has shown us clearly the vulgarity of our healthcare system.
      Highest costs in the world, yet not enough nurses or doctors. Many
      millions uninsured, while insurance company profits soar. The struggle
      continues. Healthcare is a human right. Medicare for all.
    example_title: Bernie Sanders (D)
  - text: >-
      Team Biden would rather fund the Ayatollah's Death to America regime than
      allow Americans to produce energy for our own domestic consumption.
    example_title: Ted Cruz (R)
base_model: distilbert-base-uncased

distilbert-political-tweets 🗣 🇺🇸

This model is a fine-tuned version of distilbert-base-uncased on the m-newhauser/senator-tweets dataset, which contains all tweets made by United States senators during the first year of the Biden Administration. It achieves the following results on the evaluation set:

Accuracy: 0.9076
F1: 0.9117

Model description

The goal of this model is to classify short pieces of text as having either Democratic or Republican sentiment. The model was fine-tuned on 99,693 tweets (51.6% Democrat, 48.4% Republican) made by US senators in 2021.

Model accuracy may not hold up on pieces of text longer than a tweet.

Training hyperparameters

The following hyperparameters were used during training:

optimizer: Adam
training_precision: float32
learning_rate = 5e-5
num_epochs = 5

Framework versions

Transformers 4.16.2
TensorFlow 2.8.0
Datasets 1.18.3
Tokenizers 0.11.6