https://github.com/alexisvassquez/juniper2.0
Conversational AI model and educational tool
Science Score: 13.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (14.1%) to scientific vocabulary
Keywords
Repository
Conversational AI model and educational tool
Basic Info
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
⚛️ Juniper2.0
README.md
Project: AI LLM based on GPT-4
- Main source: OpenAI GPT-2 (base), -4
- Languages: Python, JSON, JavaScript
Goal
Develop a conversational LLM model based on my personalized GPT-4 model, Juniper. The model will be able to: - Interact with users and answer questions. - Provide coding and other tech-related lessons and examples. - Assess the user's knowledge through quizzes. - Offer career guidance (interview prep, resume-building, job application tracking, send reminders, and job sourcing).
This model is designed to be a learning tool for beginners to start a career in tech, with a target audience that includes: - Continuing adult education learners. - Prisoners or individuals with criminal records. - GED students. - Those experiencing financial instability or hardship. - Other disadvantaged, novice, or late learners.
Description
Juniper 2.0 is an LLM based on OpenAI's GPT-2 model and my interactions with my assistant, Juniper (based on the GPT-4 model). This personalized assistant is built to simplify complex tech concepts and provide clear, easy-to-understand responses.
Step 1: Project setup
- Install PyTorch and Hugging Face Transformers packages
- Test basic functionality
- Source base model dataset:
- OpenAI GPT-2 (open source)
- Conversational
- Provides simplified responses to complex tech concepts
Step 2: Data Collection/Preparation
Collect/Create Datasets:
- Combine the following datasets:
- OpenAI GPT-2 (base)
- Conversational tone/context
- StackExchange
- Professional tone/context
- Tech-related conversations (Q&A)
- Kaggle
- Tech knowledge and factoids
- Tutorials
- Coding examples
- Quizzes
- GitHub Jobs/LinkedIn Jobs/Indeed APIs
- Job sourcing and career guidance
- Resume-building
- Interview prep
- Job application tracking/reminders
- Custom Dataset
- Samples of conversations between Juniper (GPT-4) and myself
Data Preprocessing:
- Format and clean the datasets
- Tokenize using mySQL/Excel/VSCode
- Store datasets as JSON or CSV with
'inputtext' and 'outputtext'
Step 3: Fine-Tuning
- Use the Hugging Face Transformers library for fine-tuning
Step 4: Model Evaluation and Tuning
- Evaluate the model's performance.
- Optimize the model for:
- Accuracy
- Response quality
- Data-specific goals
Step 5: Deployment
- Build an API for access (FastAPI/Flask).
- Deploy onto Hugging Face model hub and GitHub.
License
This project is licensed under the GNU General Public License v3.0. See the LICENSE file for details.
For more information, refer to the OpenAI GPT-2 repository.
Updates - README.md
⚙️ Updates:
Step 1 Completion: - Successful installation of key packages: - PyTorch - Hugging Face Transformers - OpenAI DistilGPT-2 datasets
Jump to Step 3 (Fine-Tuning): - Fine-tuned DistilGPT-2 using the OpenWebText dataset.
System Improvements: - Added extra storage, routed microSD card to Linux environment to resolve temp storage issues. - Addressed storage limitations that initially prevented testing fine-tuned model.
Next Steps:
Model Refinement: - Cont improving fine-tuned model's performance.
Custom Dataset Creation: - Develop custom samples from Juniper (GPT-4) dataset.
Further Testing: - Begin testing more input/output scenarios with the fine-tuned model.
System Optimization: - Apply additional sys optimizations to improve efficiency as necessary.
Owner
- Name: alexis marie vasquez
- Login: alexisvassquez
- Kind: user
- Location: Fort Lauderdale, FL
- Company: Joseph John Dougherty LLC
- Repositories: 1
- Profile: https://github.com/alexisvassquez
project manager. data analytics.
GitHub Events
Total
- Watch event: 1
- Push event: 5
Last Year
- Watch event: 1
- Push event: 5
Committers
Last synced: 7 months ago
Top Committers
| Name | Commits | |
|---|---|---|
| Alexis M Vasquez | v****2@g****m | 12 |
Issues and Pull Requests
Last synced: 7 months ago
All Time
- Total issues: 0
- Total pull requests: 1
- Average time to close issues: N/A
- Average time to close pull requests: 15 minutes
- Total issue authors: 0
- Total pull request authors: 1
- Average comments per issue: 0
- Average comments per pull request: 0.0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 1
- Average time to close issues: N/A
- Average time to close pull requests: 15 minutes
- Issue authors: 0
- Pull request authors: 1
- Average comments per issue: 0
- Average comments per pull request: 0.0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
- alexisvassquez (1)