Metadata-Version: 2.1
Name: flashtorch
Version: 0.0.6
Summary: Visualisation toolkit for neural networks in PyTorch
Home-page: https://github.com/MisaOgura/flashtorch
Author: Misa Ogura
Author-email: misa.ogura01@gmail.com
License: UNKNOWN
Description: # FlashTorch :flashlight:
        
        Visualisation toolkit implemented in PyTorch for inspecting what neural networks learn in image recognition tasks (feature visualisation).
        
        The project is very much work in progress, and I would appreciate your feedback!
        
        It currently supports visualisation of saliancy maps for all the models available under [torchvision.models](https://pytorch.org/docs/stable/torchvision/models.html).
        
        ## Installation
        
        ```bash
        $ (sudo) pip install flashtorch
        ```
        
        ## Usage (example notebooks)
        
        - [Image handling](./examples/image_handling.ipynb)
        
        - [Image-specific class saliency map with backpropagation](./examples/visualise_saliency_with_backprop.ipynb)
        
        ## Papers
        
        - Using backpropagation for gradient visualisation: [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://arxiv.org/pdf/1312.6034.pdf)
        
        - Guided backprobagation: [Striving for Simplicity: The All Convolutional Net](https://arxiv.org/pdf/1412.6806.pdf)
        
        ## Inspiration
        
        - [pytorch-cnn-visualizations](https://github.com/utkuozbulak/pytorch-cnn-visualizations) by utkuozbulak
        
        - [keras-vis](https://github.com/raghakot/keras-vis) by raghakot
        
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
