TensorFlow 2.1.0-rc2 released, TensorFlow 2.1 will be the last to support Python version TF 2. Python 2 support will be 1 January 2020 officially ended, TensorFlow also from that date to stop supporting Python 2, and will no longer be expected to release a new version in 2019.
Key features and improvements are as follows:
- tensorflow pip package is now included by default for Linux and Windows GPU support (and tensorflow-gpu same). It can run on machines with and without NVIDIA GPU's. tensorflow-gpu is still available for users concerned about the size of the package, the package can be downloaded only CPU on tensorflow-cpu.
- Windows users : In order to take advantage of new / d2ReducedOptimizeHugeFunctions compiler flags, now use Visual Studio 2019 version 16.4 build tensorflow Pip package officially released. To use these new packages, you must install the "Visual Studio 2015,2017 and 2019 of Microsoft Visual C ++ redistributable."
- This does not change on the Windows build the minimum required TensorFlow from the source code, but in the absence of this flag is enabled by constructing EIGEN_STRONG_INLINE may require more than 48 hours to build.
- If you lack the required DLL on your computer (new) one msvcp140.dll (old) or msvcp140_1.dll, the import tensorflow displays a warning message.
- tensorflow pip package uses CUDA 10.1 and cuDNN 7.6 build.
- ……
More See update:
https://github.com/tensorflow/tensorflow/releases/tag/v2.1.0-rc2