The team at TensorFlow announced the first release of TensorFlow 1.13.0-rc0 that explores major improvements, bug fixes, and other changes. Major improvements in the new TensorFlow 1.13.0-rc0 includes TensorFlow Lite has been moved from contrib to core. That means Python modules are now under tf.lite and the source code is now under tensorflow/lite instead of tensorflow/contrib/lite, TensorFlow GPU binaries have now been built against CUDA 10, and NCCL has been moved to core in TensorFlow 1.13.0-rc0.
In TensorFlow 1.13.0-rco features behavioral and other changes that comprise conversion of python floating types to uint32/64, i.e. matching behavior of other integer types in tf.constant has been disallowed in TensorFlow 1.13.0-rc0; Doc consisting of details about the rounding mode utilized in quantize_and_dequantize_v2 has been updated; the performance of GPU cumsum/cumprod has been enhanced by up to 300x; support has been added for weight decay in most TPU embedding optimizers like AdamW and MomentumW; an experimental Java API has been added for injecting TensorFlow Lite delegates; the latest support for strings in TensorFlow Lite Java API is added; and tf.spectral has been merged into tf.signal for TensorFlow 2.0.
TensorFlow’ TensorFlow 1.13.0-rc0 features major Bug fixes, include tensorflow::port::InitMain() now gets called before utilizing the TensorFlow library. Programs that fail to perform this are not portable to all platforms; saved_model.loader.load has been deprecated and is replaced by saved_model.load; Saved_model.main_op has also been deprecated and is replaced by saved_model.main_op in V2; tf.QUANTIZED_DTYPES has been deprecated and is transformed into tf.dtypes.QUANTIZED_DTYPES; sklearn imports have been updated for deprecated packages; confusion_matrix op is now exported as tf.math.confusion_matrix instead of tf.train.confusion_matrix; an ignore_unknown argument is added in TensorFlow 1.13.0-rc0 to parse_values that suppresses ValueError for unknown hyperparameter types. Such * Add tf.linalg.matvec convenience function; the XRTCompile op is can now reverse the ProgramShape resulted from the XLA compilation as a second return argument; XLA HLO graphs are rendered as SVG/HTML in TensorFlow 1.13.0-rc0.