Machinehack_Intel_oneapi_hackathon_the_llm_challenge

Competition hosted on Machinehack

About

Generate a response for the question from pre-defined text using LLM(Extracted Question-Answering(QA) Model).

The Final Competition score is 0.25114

Final Leaderboard Rank is 9/35

The Evaluation Metric is Accuracy.

File information

• mh-intel-oneapi-hackathon-the-llm-challenge-eda.ipynb

  Basic Exploratory Data Analysis

  Packages Used,

   * seaborn 
   * Pandas
   * Numpy
   * Matplotlib
   * nltk
   * spacy
   * wordcloud
   * spellchecker
   * sklearn
  

• mh-intel-oneapi-hackathon-the-llm-challenge-model.ipynb

  I have directly used a pre-trained model without fine-tuning it on the training data, primarily due to my limited knowledge in NLP-QA tasks. I loaded and predicted the test data using the transformers inference pipeline.

  Packages Used,

    * Pandas
    * Huggingface