tiny-cnn v1.0.0.a3 Release Notes

Release Date: 2016-11-29 // over 7 years ago
  • ๐Ÿš€ Now we are announcing v1.0.0a3. Thanks for all great contributors! This release includes the following changes from v1.0.0a2:

    ๐Ÿ› Bug fix

    • ๐Ÿ›  Convolutional layer with padding::same mode doesn't work #332 fixed by @nyanp
    • ๐Ÿ— Segmentation fault at MinGW build #203 #281 fixed by @nyanp
    • ๐Ÿ›  NNPACK backend doesn't work #398 fixed by @azsane

    ๐Ÿ‘Œ Improvements

    • โœ‚ Remove compiler warnings & improve CMakeLists #387 by @beru
    • ๐Ÿ‘Œ Improve memory consumption #410 by @beru
    • ๐Ÿ‘Œ Improve unit tests #408 by @Randl
    • Subtle speed optimization #419 by @beru
    • ๐Ÿ”จ Refactor serialization type & size type #407, #422 by @Randl and @edgarriba
    • ๐Ÿ‘Œ Improve compilation time by splitting serialiation/deserialization #421 by @beru

    ๐Ÿ“„ Docs&Comments

    Toward v1.0.0

    ๐Ÿ”€ The first version of tensor is merged into tiny-dnn (#411 #417 #418 by @pansk @Randl and @edgarriba). It isn't integrated with tiny-dnn layers yet, but it's the starting point of the GPU tiny-dnn.

Previous changes from v1.0.0.a2

  • ๐Ÿ› Bug Fixes

    • ๐Ÿ›  Fix SEGV errors on AVX Optimized code (#353) by @nyanp
    • ๐Ÿ›  Fix compiler error on msvc2013 (#320) by @nyanp
    • ๐Ÿ›  Fix AVX backend slowdown on convolutional layer (#322) by @nyanp
    • ๐Ÿ›  Fix throwing error when we load weights manually (#330) by @nyanp
    • ๐Ÿ›  Fix returning infinity in tan_h (#347) by @nyanp
    • ๐Ÿ›  Fix portability issues on serialization (#377) by @nyanp

    ๐Ÿ”‹ Features

    • ๐Ÿ‘ Provides compile option to disable serialization support to speedup compilation time (#316) by @nyanp
    • โž• Adds set_trainable method to freeze layers (#346) by @nyanp
    • โž• Adds power layer to caffe converter by @goranrauker
    • ๐Ÿ‘ double precision support (#332) by @nyanp
    • Provides pad_type and non-square input to pooling layers (#374) by @nyanp
    • โž• Adds public predict method for vector of tensors (#396) by @reunanen
    • โž• Adds Auto engine selection (#339) by @edgarriba
    • โž• Adds basic image utilities, and remove OpenCV dependencies (#337) by @nyanp