CogitoNTNU/DeepTactics-Muzero
Feb 2025 – Apr 2025
3 Months active
Languages Used
JavaScriptPythonCSSHTMLSVGTypeScript
Technical Skills
Algorithm DesignBackend DevelopmentConfiguration ManagementDeep LearningGame AIMachine Learning
adisinghstudent
PROFILE
Adisinghstudent
Over a three-month period, contributed to CogitoNTNU/DeepTactics-Muzero by integrating a unified MuZero system, overhauling neural network architecture, and modernizing the frontend experience. Developed a modular backend in Python and PyTorch, establishing a base Network class and components for representation, prediction, and dynamics, which streamlined experimentation and reproducibility. Addressed a critical bug in the network’s representation layer, improving model reliability and evaluation stability. On the frontend, delivered a responsive navigation bar, enhanced home page visuals, and implemented global theming using React, Next.js, and Tailwind CSS, resulting in improved UI performance, maintainability, and a more cohesive user experience.
Overall Statistics
Feature vs Bugs
67%Features
Repository Contributions
30Total
Bugs
2
Commits
30
Features
4
Lines of code
2,454
Activity Months3
Your Network
12 peopleWork History
April 2025
20 Commits • 3 Features
Apr 1, 2025April 2025 frontend delivery for CogitoNTNU/DeepTactics-Muzero focused on navigation modernization, MuZero home-page storytelling, and a cohesive design system, while stabilizing builds and optimizing UI performance. Key outcomes include a responsive animated stacking navigation bar with a sticky header; enhanced MuZero visuals and neural-network explanations on the home page (Hero45 component); global theming with gradient backgrounds and consistent typography; and targeted UI performance improvements. Bug fixes addressed build/logic issues and pixel trail performance to ensure smooth user interactions. These efforts improve user onboarding, content comprehension, and cross-page readability, while reducing maintenance overhead.
20 Commits • 3 Features
Apr 1, 2025April 2025 frontend delivery for CogitoNTNU/DeepTactics-Muzero focused on navigation modernization, MuZero home-page storytelling, and a cohesive design system, while stabilizing builds and optimizing UI performance. Key outcomes include a responsive animated stacking navigation bar with a sticky header; enhanced MuZero visuals and neural-network explanations on the home page (Hero45 component); global theming with gradient backgrounds and consistent typography; and targeted UI performance improvements. Bug fixes addressed build/logic issues and pixel trail performance to ensure smooth user interactions. These efforts improve user onboarding, content comprehension, and cross-page readability, while reducing maintenance overhead.
April 2025
March 2025
1 Commits
Mar 1, 2025March 2025 monthly summary for CogitoNTNU/DeepTactics-Muzero focusing on network representation correctness and reliability. Delivered a targeted forward-pass fix to the Network Representation Layer, correcting a mis-specification in the representation path and removing a placeholder for random hidden state generation. The change ensures accurate representation outputs and stabilizes downstream evaluation. Impact: improved model reliability, reproducibility of results, and faster debugging for representation-related issues in the Muzero-based pipeline.
March 2025
1 Commits
Mar 1, 2025March 2025 monthly summary for CogitoNTNU/DeepTactics-Muzero focusing on network representation correctness and reliability. Delivered a targeted forward-pass fix to the Network Representation Layer, correcting a mis-specification in the representation path and removing a placeholder for random hidden state generation. The change ensures accurate representation outputs and stabilizes downstream evaluation. Impact: improved model reliability, reproducibility of results, and faster debugging for representation-related issues in the Muzero-based pipeline.
February 2025
9 Commits • 1 Features
Feb 1, 2025February 2025 monthly summary for CogitoNTNU/DeepTactics-Muzero: Delivered a consolidated MuZero system integration and neural network architecture overhaul, establishing a cohesive framework that unifies the algorithm, networks, action representation, and configuration. Built a base Network class and modular components for representation, prediction, and dynamics; introduced training/configuration scaffolding and debugging utilities to accelerate experimentation and improve maintainability. The foundation enables faster iterations, reproducible experiments, and scalable training pipelines.
9 Commits • 1 Features
Feb 1, 2025February 2025 monthly summary for CogitoNTNU/DeepTactics-Muzero: Delivered a consolidated MuZero system integration and neural network architecture overhaul, establishing a cohesive framework that unifies the algorithm, networks, action representation, and configuration. Built a base Network class and modular components for representation, prediction, and dynamics; introduced training/configuration scaffolding and debugging utilities to accelerate experimentation and improve maintainability. The foundation enables faster iterations, reproducible experiments, and scalable training pipelines.
February 2025
Activity
Loading activity data...
Quality Metrics
Correctness86.6%
Maintainability86.6%
Architecture82.0%
Performance79.8%
AI Usage21.4%
Skills & Technologies
Programming Languages
CSSHTMLJavaScriptPythonSVGTypeScript
Technical Skills
Algorithm DesignAnimationBackend DevelopmentCSSComponent DesignComponent DevelopmentComponent-Based ArchitectureConfiguration ManagementCustom HooksDeep LearningFrontend DevelopmentGame AIMachine LearningNeural NetworksNext.js
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline