Develop the backend functionality to handle incoming user input from the frontend, process it, and return a response generated by a model. For testing and demonstration purposes, we will use the Hugging Face model (Salesforce/codet5-base) (https://huggingface.co/Salesforce/codet5-base) as a mock model. This backend API will enable interaction between the frontend’s user input and the model's response, allowing the UI to simulate conversation with an AI model in a chat format.

Acceptance Criteria:

• Endpoint Setup:

  • Create an API endpoint that accepts a POST request with JSON data
  • Validate the request to ensure the "text" field is present and non-empty.

• Process Input with Mock Model:

  • Use the Hugging Face model Salesforce/codet5-base as a mock to generate responses. This model will serve as a placeholder to test the system until the actual model is integrated.
  • Integrate the Hugging Face API or Transformers library to interact with the model and process the user’s input.

• Model Processing Script:

  • Set up a Python script using the transformers library that takes in the text input, runs it through the Salesforce/codet5-base model, and returns the generated output.(You can find the model script in the link with description)
  • This script will serve as a backend utility that can be called by the API to generate responses.

• Response Handling:

  • Return the model’s output in a JSON format so the frontend can display it.
  • If there is any processing error, return an appropriate error message with a clear HTTP status code.

• Error Handling:

  • Handle possible errors such as invalid input, model processing failures, and server errors.
  • Provide meaningful error messages for each error type.

Additional Information:

• This setup is intended as a mock to test the chat interface and backend flow.
• Later, the mock model can be replaced with the actual bioinformatics model once it is ready for integration.