Skip to content

Latest commit

 

History

History
73 lines (64 loc) · 1.83 KB

README.md

File metadata and controls

73 lines (64 loc) · 1.83 KB

torch_soft

Version MIT License

torch_soft is a high level implementation of pytorch for some of claasification implementation with ease...

Installation

You can use pip to install torch_soft using pip.

pip install torch-soft==0.1.1

features

TensorNet currently supports the following features

  • Model architectures
    • ResNet18
    • A custom model called naiveresnet
  • Model utilities
    • Loss functions
      • Cross entropy loss
      • nll_loss
    • Optimizers
      • Stochastic Gradient Descent
    • Regularizers
      • L1 regularization
      • L2 regularization
    • LR Schedulers
      • Step LR
      • Reduce LR on Plateau
      • One Cycle Policy
    • LR Range Test
  • Model training and validation
  • Datasets (data is is returned via data loaders)
    • MNIST
    • CIFAR10
    • TinyImageNet
  • Data Augmentation
    • Resize
    • Padding
    • Random Crop
    • Horizontal Flip
    • Vertical Flip
    • Gaussian Blur
    • Random Rotation
    • CutOut
  • GradCAM and GradCAM++ (Gradient-weighted Class Activation Map)
  • Result Analysis Tools
    • Plotting changes in validation accuracy and loss during model training
    • Displaying correct and incorrect predictions of a trained model
    • Plotting images in a batch for visualization
    • Plotting gradcam outputs

How to Use

For examples on how to use torch_soft, refer to the examples directory.

Dependencies

torch_soft has the following third-party dependencies

  • torch
  • torchvision
  • torchsummary
  • tqdm
  • matplotlib
  • albumentations
  • opencv-python