Completed
- Breaking up text into chunks for indexing
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Answer Complex Questions From an Arbitrarily Large Set of Documents With Vector Search and GPT-3
Automatically move to the next video in the Classroom when playback concludes
- 1 - The need for answering questions from arbitrary data sources
- 2 - The benefits of a multi-document answering system
- 3 - Building a chatbot to answer questions and summarize documents
- 4 - Breaking up text into chunks for indexing
- 5 - Building an index
- 6 - Converting unicode to ascii to prevent gpt3 errors
- 7 - Saving data as a JSON file
- 8 - Building the index
- 9 - Building a Knowledge Base
- 10 - Building the index
- 11 - Searching for memories
- 12 - Using GPT-3 to answer questions about a text
- 13 - The majority's decision to allow states to ban abortion
- 14 - Answering a question with a superintelligence
- 15 - Generating answers to questions with GPT-3
- 16 - Joining answers into one big block
- 17 - Creating a detailed summary of chunks
- 18 - Trying to fix a broken search
- 19 - Fixing the bug in the gpt3 completion function
- 20 - GPT3's difficulty with complex questions
- 21 - The gpt3 log
- 22 - The Supreme Court's decision on abortion
- 23 - The Supreme Court's decision on abortion
- 24 - The Supreme Court overturns a lower ruling banning abortion
- 25 - The final result of the superintelligence question
- 26 - The Supreme Court overturns Roe v. Wade
- 27 - The Supreme Court overturns Roe v. Wade
