https://github.com/devzhk/ds_projects
Group projects for Data Science Foundation
https://github.com/devzhk/ds_projects
Last synced: 7 months ago
JSON representation
Group projects for Data Science Foundation
- Host: GitHub
- URL: https://github.com/devzhk/ds_projects
- Owner: devzhk
- Created: 2019-04-03T08:23:21.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2019-07-30T17:31:49.000Z (over 6 years ago)
- Last Synced: 2025-07-09T10:43:38.067Z (7 months ago)
- Language: Python
- Size: 6.32 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# DS_Projects
## 1. Dimensionality Reduction
1. Use linear SVM for image classification based on the deep learning features. You can use K-fold cross-validation within the training set to determine hyper-parameters, such as C in SVM.
2. Reduce the dimensionality of deep learning features using at least three methods (i.e., one feature selection method, one feature projection method, one feature learning method) and perform image classification again based on the obtained low-dimensional features. Record the performance variance w.r.t. different feature dimensionality.
3. Explore the optimal dimensionality reduction method and the optimal dimensionality.
4. The project report should contain experimental setting (i.e., dataset, feature, training/testing split), the dimensionality reduction methods you tried, the experimental results you obtained, and the experimental observations based on your experimental results.
## 2. K-Nearest Neighbors (KNN) Classification with Different Distance Metrics
## 3. Feature Encoding for Image Classification
## 4. Domain Adaptation