rathapech

Pinned Repositories

• DTI_LMP

  The codes for predicting interactions between drugs and targets based on low-rank matrix projection on heterogeneous biological data.

  Language:MATLAB53

• LOMDA

  The codes aim at predicting the associations between miRNAs and diseases based on linear optimization.

  Language:MATLAB40

• fraud_detection

  The study of fraud detection in mobile money transfer as binary classification

  Language:Python20

• Link-prediction-via-matrix-completion

  Link prediction via matrix completion (published in EPL, (2017): 38002)

  Language:MATLAB21

• customer_churn_prediction

  The Python source code for predicting the customer churn based on Scikit Learn.

  Language:Python10

• split-worksheets-to-workbook-in-Excel

  Splitting multiple worksheets and save them to different workbooks in Excel

  10

• Bank-Marketing

  The data I used in this report is obtained from a publication entitled "A data-driven approach to predict the success of bank telemarketing". Moro, Sérgio, Paulo Cortez, and Paulo Rita. "A data-driven approach to predict the success of bank telemarketing." Decision Support Systems 62 (2014): 22-31. Available at: https://archive.ics.uci.edu/ml/datasets/Bank+Marketing

  Language:Jupyter Notebook00

• Bees-vs-Ants-Classification-Pytorch

  Classify between bees and ants using deep learning and convolutional neural network with Pytorch.

  Language:Jupyter Notebook00

• Boston_Crimes

  Boston crimes analysis using different visualization techniques and clustering

  Language:Jupyter Notebook00

• Cats_vs_Dogs_DL_augmentati_pre-trained

  In this study, we aim at classifying two classes of images of cats and dogs respectively by using limited data. There are 1000 images of each class and there are 400 images for validation. In total, we have 2000 training data and 800 validating data. It is commonly believed that deep learning works well with large amount of training data especially in image classification. It is not completely wrong. In this study, we will train the deep learning classifier with limited data and try to improve the performance by using two techniques, namely data augmentation and pre-trained network.

  Language:Jupyter Notebook00

rathapech's Repositories