README.md

Classroom Examples for Today: 🚀Examples

👋 Two easy ways to turbo boost your AI learning journey! 💻

🌐 AI Pair Programming

Open 2 Browsers to:

1. 🌐 ChatGPT URL or URL2 and
2. 🌐 Huggingface URL in separate browser windows.
3. 🤖 Use prompts to generate a streamlit program on Huggingface or locally to test it.
4. 🔧 For advanced work, add Python 3.10 and VSCode locally, and debug as gradio or streamlit apps.
5. 🚀 Use these two superpower processes to reduce the time it takes you to make a new AI program! ⏱️

Example Starter Prompt:

Write a streamlit program that demonstrates Data synthesis. Synthesize data from multiple sources to create new datasets. Use two datasets and demonstrate pandas dataframe query merge and join with two datasets in python list dictionaries: List of Hospitals that are over 1000 bed count by city and state, and State population size and square miles. Perform a calculated function on the merged dataset.

🎥 YouTube University Method:

1. 🏋️‍♀️ Plan two hours each weekday to exercise your body and brain.
2. 🎬 Make a playlist of videos you want to learn from on YouTube. Save the links to edit later.
3. 🚀 Try watching the videos at a faster speed while exercising, and sample the first five minutes of each video.
4. 📜 Reorder the playlist so the most useful videos are at the front, and take breaks to exercise.
5. 📝 Practice note-taking in markdown to instantly save what you want to remember. Share your notes with others!
6. 👥 AI Pair Programming Using Long Answer Language Models with Human Feedback:

🎥 2023 AI/ML Advanced Learning Playlists:

1. 2023 QA Models and Long Form Question Answering NLP
2. FHIR Bioinformatics Development Using AI/ML and Python, Streamlit, and Gradio - 2022
3. 2023 ChatGPT for Coding Assistant Streamlit, Gradio and Python Apps
4. 2023 BigScience Bloom - Large Language Model for AI Systems and NLP
5. 2023 Streamlit Pro Tips for AI UI UX for Data Science, Engineering, and Mathematics
6. 2023 Fun, New and Interesting AI, Videos, and AI/ML Techniques
7. 2023 Best Minds in AGI AI Gamification and Large Language Models
8. 2023 State of the Art for Vision Image Classification, Text Classification and Regression, Extractive Question Answering and Tabular Classification
9. 2023 AutoML DataRobot and AI Platforms for Building Models, Features, Test, and Transparency

Language Models 🗣️

🏆 Bloom sets new record for most performant and efficient AI model in science! 🌸

Comparison of Large Language Models

Model Name	Model Size (in Parameters)
BigScience-tr11-176B	176 billion
GPT-3	175 billion
OpenAI's DALL-E 2.0	500 million
NVIDIA's Megatron	8.3 billion
Transformer-XL	250 million
XLNet	210 million

ChatGPT Datasets 📚

• WebText
• Common Crawl
• BooksCorpus
• English Wikipedia
• Toronto Books Corpus
• OpenWebText

ChatGPT Datasets - Details 📚

• WebText: A dataset of web pages crawled from domains on the Alexa top 5,000 list. This dataset was used to pretrain GPT-2.
  • WebText: A Large-Scale Unsupervised Text Corpus by Radford et al.
• Common Crawl: A dataset of web pages from a variety of domains, which is updated regularly. This dataset was used to pretrain GPT-3.
  • Language Models are Few-Shot Learners by Brown et al.
• BooksCorpus: A dataset of over 11,000 books from a variety of genres.
  • Scalable Methods for 8 Billion Token Language Modeling by Zhu et al.
• English Wikipedia: A dump of the English-language Wikipedia as of 2018, with articles from 2001-2017.
  • Improving Language Understanding by Generative Pre-Training Space for Wikipedia Search
• Toronto Books Corpus: A dataset of over 7,000 books from a variety of genres, collected by the University of Toronto.
  • Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond by Schwenk and Douze.
• OpenWebText: A dataset of web pages that were filtered to remove content that was likely to be low-quality or spammy. This dataset was used to pretrain GPT-3.
  • Language Models are Few-Shot Learners by Brown et al.

Big Science Model 🚀

• 📜 Papers:

  1. BLOOM: A 176B-Parameter Open-Access Multilingual Language Model Paper
  2. Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism Paper
  3. 8-bit Optimizers via Block-wise Quantization Paper
  4. Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation Paper
  5. Other papers related to Big Science
  6. 217 other models optimized for use with Bloom

• 📚 Datasets:

Datasets:

1. • Universal Dependencies: A collection of annotated corpora for natural language processing in a range of languages, with a focus on dependency parsing.

• Universal Dependencies official website.

2. • WMT 2014: The fourth edition of the Workshop on Statistical Machine Translation, featuring shared tasks on translating between English and various other languages.

• WMT14 website.

3. • The Pile: An English language corpus of diverse text, sourced from various places on the internet.

• The Pile official website.

4. • HumanEval: A dataset of English sentences, annotated with human judgments on a range of linguistic qualities.

• HumanEval: An Evaluation Benchmark for Language Understanding by Gabriel Ilharco, Daniel Loureiro, Pedro Rodriguez, and Afonso Mendes.

5. • FLORES-101: A dataset of parallel sentences in 101 languages, designed for multilingual machine translation.

• FLORES-101: A Massively Multilingual Parallel Corpus for Language Understanding by Aman Madaan, Shruti Rijhwani, Raghav Gupta, and Mitesh M. Khapra.

6. • CrowS-Pairs: A dataset of sentence pairs, designed for evaluating the plausibility of generated text.

• CrowS-Pairs: A Challenge Dataset for Plausible Plausibility Judgments by Andrea Madotto, Zhaojiang Lin, Chien-Sheng Wu, Pascale Fung, and Caiming Xiong.

7. • WikiLingua: A dataset of parallel sentences in 75 languages, sourced from Wikipedia.

• WikiLingua: A New Benchmark Dataset for Cross-Lingual Wikification by Jiarui Yao, Yanqiao Zhu, Ruihan Bao, Guosheng Lin, Lidong Bing, and Bei Shi.

8. • MTEB: A dataset of English sentences, annotated with their entailment relationships with respect to other sentences.

• Multi-Task Evaluation Benchmark for Natural Language Inference by Michał Lukasik, Marcin Junczys-Dowmunt, and Houda Bouamor.

9. • xP3: A dataset of English sentences, annotated with their paraphrase relationships with respect to other sentences.

• xP3: A Large-Scale Evaluation Benchmark for Paraphrase Identification in Context by Aniket Didolkar, James Mayfield, Markus Saers, and Jason Baldridge.

10. • DiaBLa: A dataset of English dialogue, annotated with dialogue acts.

• A Large-Scale Corpus for Conversation Disentanglement by Samuel Broscheit, António Branco, and André F. T. Martins.

• 📚 Dataset Papers with Code

  1. Universal Dependencies
  2. WMT 2014
  3. The Pile
  4. HumanEval
  5. FLORES-101
  6. CrowS-Pairs
  7. WikiLingua
  8. MTEB
  9. xP3
  10. DiaBLa

Deep RL ML Strategy 🧠

The AI strategies are:

• Language Model Preparation using Human Augmented with Supervised Fine Tuning 🤖
• Reward Model Training with Prompts Dataset Multi-Model Generate Data to Rank 🎁
• Fine Tuning with Reinforcement Reward and Distance Distribution Regret Score 🎯
• Proximal Policy Optimization Fine Tuning 🤝
• Variations - Preference Model Pretraining 🤔
• Use Ranking Datasets Sentiment - Thumbs Up/Down, Distribution 📊
• Online Version Getting Feedback 💬
• OpenAI - InstructGPT - Humans generate LM Training Text 🔍
• DeepMind - Advantage Actor Critic Sparrow, GopherCite 🦜
• Reward Model Human Prefence Feedback 🏆

For more information on specific techniques and implementations, check out the following resources:

• OpenAI's paper on GPT-3 which details their Language Model Preparation approach
• DeepMind's paper on SAC which describes the Advantage Actor Critic algorithm
• OpenAI's paper on Reward Learning which explains their approach to training Reward Models
• OpenAI's blog post on GPT-3's fine-tuning process