Arik Poznanski

icon location My blog: https://arikpoz.github.io/ Enjoyed my work? Treat me to a coffee at https://ko-fi.com/arikpoz ☕

Results 5 issues of Arik Poznanski

Major features are extended siamese network support, visualizing histograms and correlation matrices built from activations and weights, extended deconv options, added grayscale support, improved optimize_image and find_max_acts, improved user interface, refactored and fixed bugs

# Features ## - Siamese network support - Added support for loading Siamese networks, presenting the inputs side-by-side, and properly calculate activations and gradients for visualization - Supports two types...

Fix for quantize fail on init due to undefined static initialization of complex objects

7 comment

Addresses issue #920 Replaced static initialization of complex objects with a initialization on first use. This prevents an undefined behavior on program run, for example, crash in Release build, works...

Added "guided backprop" option for deconv, fixed bugs, updated to latest caffe

I've added another option for making the deconv, namely using "guided backpropagation" instead of "Zeiler-Fergus" deconv. Note that the type of deconv to use is passed as a parameter so...

Improve device discovery

- Enhance device discovery with user-configurable network settings (SSID, network password, subnet address, local IP address) - Refactor smartir_generator.py with type hints and improved error handling

Added Elco Classic climate code for Broadlink