Connor-Shen's repositories
Visual-Analysis-System-for-Rumor-Propagation-under-Covid-19
In this project I design a rumor visualization system for visualizing rumors in social networks during the epidemic. I conduct topic modeling of the rumor dataset and explore the changes in topics over time and region. Geographical rumor distribution is also performed and visualized in the form of China map.
all-of-statistics
Self-study on Larry Wasserman's "All of Statistics"
BP-Neural-Network-Exploration
Using python(start from scratch) to build a basic neural network in identificating handwriting digits
code-evaluator
A multi-language code evaluation tool.
connor-shen.github.io
A beautiful, simple, clean, and responsive Jekyll theme for academics
Data_Science_Template
Templates of different python packages evaluating the language model's data science ability. Experiments are generated from GPT-4.
Exploration-on-Batch-Normalization
In this PJ I build a simple Conv2d network and implement some beneficial modifications based on it. Meanwhile, by researching into the BatchNorm algorithm, I verify its benefits on a variety of procedures.
LAgent_ToolBench
A lightweight framework for building LLM-based agents
opencompass
OpenCompass is an LLM evaluation platform, supporting a wide range of models (LLaMA, LLaMa2, ChatGLM2, ChatGPT, Claude, etc) over 50+ datasets.
Pruning-Transformer-via-Sparsity
Neural network pruning is a method that revolves around the intuitive idea of removing superfluous parts of a network that performs well but costs a lot of resources. The idea is to eliminate these parts without impacting the network’s performance.
Novel-Image-Captioning
A pytorch repetition of Novel Object Captioner (NOC)
QWE-dataset
In this project I tried to explore the QWE dataset and build a model to predict the potential churning customers. EDA and business analysis were also conducted apart from building the predict model.