Skip to content

Aayushktyagi/DeepLearning_Resources

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Deep Learning Resources

Repository contains variour Machine learning , Deep learning and computer vision approaches

Image Processing

Notebook Covers following

  • Loading image using opencv
  • Using filter for Edge computation
  • Feature extraction and visualization using HOG.

PyRadiomics

Notebook Cover following

  • Loading images
  • Image visualization with segemenation
  • Feature extraction using Pyradiomics

Logistic and linear regression

Notebook Covers following

  • Loading dataset into csv format(Heart disease dataset for logistic regression and diabetes dataset for linear regression)
  • converting pandas dataframe to tf.data format
  • Convarience matrix calculation
  • Training linear regression using Sklearn
  • Training logistic regresison using tensorflow

SVM

Notebook Covers following

  • Loading dataset in csv format
  • Training SVM with Grid Search
  • Model evaluation

ResNet 50 Baseline

Notebook covers following

  • Loading data from directory using flow_from_dir function
  • Data augmentation using ImageDatagenerator
  • Validation split from same dataset
  • Train ResNet50 as Backbone
  • Evaluating the Model

Aptos challenge

Notebook covers following

  • Loading data with csv file and directory
  • Data Visualization
  • ResNet 50 , Xception Net and NASNet as backbone for Fine Tuning
  • Data loader with Augmentation
  • Confusion metrics for evaluation

UefficientNet

Notebook covers following

  • Loading images from folder custom data loader
  • EfficientNet as Unet backbone
  • Loss function IOU Dice
  • Custom callback for evalauting on test set while training
  • Segmentation visualization
  • evalaution

Vae

Notebook covers following

  • Loading image from directory
  • Custom Encoder and Decoder
  • KL and mse loss
  • visualization
  • evalaution

Hyper-parameter tuning

Notebook covers following

  • Uses cancer stage classification dataset
  • Training DNN model
  • Hyper-parametert tuning
  • visualization
  • Evaluation

About

Deep Learning Resources

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published