Create a FastAPI service that utilizes a pre-trained Hugging Face NLP model to perform text summarization on user-provided text inputs. The service should expose an API to receive text input and return a summary in a JSON response.

1. Environment Setup:

   • Set up a Python environment with the necessary dependencies (FastAPI, Hugging Face Transformers, or any other library you find appropriate)

2. Model Selection:

   • Choose a pre-trained NLP model from the Hugging Face Model Hub that is suitable for text summarization.

3. API Endpoint:

   • Implement an endpoint /summarize that:
     • Accepts a JSON input containing the user-provided text and a desired summary length.
     • Returns a JSON response with the summarized text.

   Example Request:

   {
     "text": "This is a long text input that needs to be summarized. The FastAPI service should be able to handle lengthy inputs and generate a coherent summary based on the user-defined summary length...",
     "summary_length": 50
   }

   Example Response:

   {
     "summary": "A brief summary of the lengthy input text, generated by the pre-trained Hugging Face model."
   }

4. Containerization:

   • Containerize the FastAPI service using Docker.
   • Provide a Dockerfile and instructions on how to build and run the Docker container.

5. Documentation:

   • Include a README file with clear instructions on how to set up and run the service.

For candidates who wish to further demonstrate their technical abilities, implement the following additional functionality:

1. Compressed Summary Endpoint:

   • Add a POST endpoint /compress_summary that:
     • Accepts the same input as /summarize.
     • Returns a compressed version of the summarized text.

2. Compression Algorithm:

   • For each group of consecutive repeating characters in the summarized text:
     • If the group's length is 1, append the character to the compressed string.
     • Otherwise, append the character followed by the group's length.

   Example:

   • Original Summary:
     This committee meets annually to assess the effectiveness of policies on environmental protection.
   • Compressed Summary:
     This com2it2e2 me2ts an2ual2y to as2es2 the ef2ectivenes2 of policies on environmental protection.
   • Sample Output:

   {
     "summary": "This com2it2e2 me2ts an2ual2y to as2es2 the ef2ectivenes2 of policies on environmental protection."
   }