NirajYagnik / Fashion_Recommendation_System

Course: DSC 210: Numerical Linear Algebra for Data Science

Project-Visually Aware Fashion Recommendation Systems

Course: DSC 210: Numerical Linear Algebra for Data Science

Instructor: Dr. Tsui-wei Weng

Team Members:

Atharva Kshirsagar - A59019697

Devam Dave - A59018455

Niraj Yagnik - A59018067

Kartikay Anand - A59020229

Instructions:

• Ensure that the following libraries are installed in python 3 environment:
  • scikit-learn
  • surprise
  • tqdm
  • nltk
  • umap
  • matplotlib
  • seaborn
  • wordcloud
  • plotly
  • tensorflow
  • keras
  • IPython.display
  • requests
  • pytz

The experimentations were conducted on Google Colab Pro and the notebook execution requires around 10-16 GB of free space.

Part_2_SOTA requires GPU support for acceleration.

Results:

• EDA

• SVD

We were able to improve the performance after the presenation and these are out final reproducable results!

• SVD vs SOTA