Skip to content

Latest commit

 

History

History
20 lines (18 loc) · 824 Bytes

File metadata and controls

20 lines (18 loc) · 824 Bytes

LLM-Data-Engineering

  1. Execute the provided notebooks in Google Colab.
  2. Find screenshots of the output in the "output" folder.
  3. The Python files associated with Textract can be run in Lambda. Ensure you configure roles and permissions for Lambda to interact with S3 and Textract.

Topics Covered

  1. What is LLM?
  2. SageMaker
  3. Textract and use cases ( Demo) a. Pdf document to text with confidence score b. Pdf table to text with confidence score
  4. Overview of Haystack + Mixtral model (Demo) a. Web scraping + prompt template + Q & A b. AP and CP websites c. Video to + prompt template + Q & A
  5. Logs NLP Analysis - Not progressed
  6. Text to Image - Bedrock Titan
  7. Case notes - NLP Analysis - Didnt get good ouput- lack of data Text summarisation , topic modelling, sentiment analy