py311-timm-vis
2.0.0_1Visualizer for PyTorch image models
timm-vis is a PyTorch image models visualizer. Implementation of various visualization techniques for pytorch image classifiers. This library can be used to visualize and understand any PyTorch image classifier. This is NOT an official PyTorch library, nor is it affiliated with Ross Wightman's pytorch-image-models library. details.ipynb has visual examples of all methods implemented. Currently, the following methods are implemented: * Filter visualization * Activations visualization * Maximally activated patches * Saliency maps * Synthetic image generation * Adversarial attacks to fool models * Feature inversion * Grad-CAM * Deep Dream
Origin: misc/py-timm-vis
Category: misc
Size: 78.7KiB
License: MIT
Maintainer: yuri@FreeBSD.org
Dependencies: 6 packages
Required by: 0 packages
Website: github.com/novice03/timm-vis
$
pkg install py311-timm-visDependencies (6)
More in misc
shared-mime-info2.4_2
MIME types database from the freedesktop.org projectlibuuid2.41.1_1
Library to generate or parse 128-bit Universally Unique IDs (UUIDs)iso-codes4.19.0
Lists of ISO standards, e.g. country, language, language scripts and currency namespy311-tqdm4.67.1
Fast, extensible progress bar for Pythonpy311-pytorch2.8.0_4
PyTorch: Tensors and dynamic neural networks in Pythonrubygem-mime-types3.7.0
Ruby gem which provides a mailcap-like MIME Content-Type lookupen-freebsd-doc20260213,1
Documentation from the FreeBSD Documentation Projectkf6-purpose6.22.0
Offers available actions for a specific purposeSoapySDR0.8.1_1
Vendor and platform neutral SDR support libraryhicolor-icon-theme0.18
High-color icon theme shell from the FreeDesktop project