metadata

library_name: transformers
tags:
  - text-classification
  - sentiment-analysis
  - depression
  - BERT
  - mental-health
model-index:
  - name: Sentiment Classifier for Depression
    results:
      - task:
          type: text-classification
        dataset:
          name: Custom Depression Tweets Dataset
          type: custom
        metrics:
          - name: Accuracy
            type: accuracy
            value: 99.87
          - name: Precision
            type: precision
            value: 99.91
          - name: Recall
            type: recall
            value: 99.81
          - name: F1 Score
            type: f1
            value: 99.86
license: apache-2.0
language:
  - en
base_model: google-bert/bert-base-uncased
metrics:
  - Accuracy
  - Recall
  - Percision
  - F1 Score
widget:
  - text: >-
      RT EichinChangLim In Talking About Adolescence Book you'll discover key
      strategies to tackle self-harm panic attacks bullies child
    example_title: Depression
  - text: >-
      SharronS Hello there Thanks for reaching out I can understand your
      frustration I would feel the same Id be happy to take a closer look Please
      feel free to send me a DM with your full name phone number and email
      address Lena
    example_title: Non-depression

Model Card for Sentiment Classifier for Depression

This model is a fine-tuned version of BERT (bert-base-uncased) for classifying text as either Depression or Non-depression. The model was trained on a custom dataset of mental health-related social media posts and has shown high accuracy in sentiment classification.

Training Data

The model was trained on a custom dataset of tweets labeled as either depression-related or not. Data pre-processing included tokenization and removal of special characters.

Training Procedure

The model was trained using Hugging Face's transformers library. The training was conducted on a T4 GPU over 3 epochs, with a batch size of 16 and a learning rate of 5e-5.

Evaluation and Testing Data

The model was evaluated on a 20% holdout set from the custom dataset.

Results

Accuracy: 99.87%
Precision: 99.91%
Recall: 99.81%
F1 Score: 99.86%

Environmental Impact

The carbon emissions from training this model can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Hardware Type: T4 GPU
Hours used: 1 hour
Cloud Provider: Google Cloud (Colab)
Carbon Emitted: Estimated at 0.45 kg CO2eq

Technical Specifications

Architecture: BERT (bert-base-uncased)
Training Hardware: T4 GPU in Colab
Training Library: Hugging Face transformers

Citation

BibTeX:

@misc{poudel2024sentimentclassifier,
  author = {Poudel, Ashish},
  title = {Sentiment Classifier for Depression},
  year = {2024},
  url = {https://huggingface.co./poudel/sentiment-classifier},
}