sophgo/LLM-TPU
Dec 2024 – Dec 2024
1 Month active
Languages Used
C++Python
Technical Skills
C++ DevelopmentComputer VisionDeep LearningLarge Language ModelsMachine LearningModel Export
fangz-ai
PROFILE
Fangz-ai
During December 2024, Fangzheng Wang enhanced the sophgo/LLM-TPU repository by implementing multimodal input support and video processing for the Qwen2_VL model. He refined the model’s forward pass in both C++ and Python, enabling robust handling of image and video data. His work included updating position ID logic and adapting C++ processing to accommodate multimodal inputs, addressing the challenge of integrating diverse data types. Additionally, he developed an ONNX export script to facilitate video input processing and deployment. This feature leveraged his expertise in C++ development, computer vision, and deep learning, demonstrating depth in model export and multimodal AI integration.
Overall Statistics
Feature vs Bugs
100%Features
Repository Contributions
2Total
Bugs
0
Commits
2
Features
1
Lines of code
2,494
Activity Months1
Your Network
25 people
Work History
December 2024
2 Commits • 1 Features
Dec 1, 2024December 2024—sophgo/LLM-TPU: Delivered Qwen2_VL multimodal input support and video processing with ONNX export, enabling robust image and video data handling and deployment readiness.
2 Commits • 1 Features
Dec 1, 2024December 2024—sophgo/LLM-TPU: Delivered Qwen2_VL multimodal input support and video processing with ONNX export, enabling robust image and video data handling and deployment readiness.
December 2024
Activity
Loading activity data...
Quality Metrics
Correctness80.0%
Maintainability80.0%
Architecture80.0%
Performance70.0%
AI Usage20.0%
Skills & Technologies
Programming Languages
C++Python
Technical Skills
C++ DevelopmentComputer VisionDeep LearningLarge Language ModelsMachine LearningModel ExportMultimodal AIONNXPython DevelopmentTransformer Architecture
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline