v2.12.0rc0

README.md

@@ -19,30 +19,30 @@ Prebuilt binary with Tensorflow Lite enabled. For RaspberryPi. Since the 64-bit
 |RaspberryPi3/4|Ubuntu 22.04|Jammy|aarch64 / armv8|3.10.x|64bit, glibc2.35|
 |???|Debian|Bookworm|aarch64 / armv8|3.11.x|64bit, glibc2.36|
 
-Minimal configuration stand-alone installer for Tensorflow Lite.  
+Minimal configuration stand-alone installer for Tensorflow Lite.
 **https://github.com/PINTO0309/TensorflowLite-bin.git**
 
-The official installer including only the Tensorflow Lite runtime can be obtained from the following URL.  
+The official installer including only the Tensorflow Lite runtime can be obtained from the following URL.
 **https://www.tensorflow.org/lite/guide/python**
 
-Bazel's pre-build binay is below.  
+Bazel's pre-build binay is below.
 **https://github.com/PINTO0309/Bazel_bin.git**
 
-Procedure for building a remote caching environment by Bazel.  
-**[Ultra-fast build of Tensorflow with Bazel Remote Caching [Google Cloud Storage version]](https://qiita.com/PINTO/items/eef2eccc21dc72227d0f)**  
-**[Ultra-fast build of Tensorflow with Bazel Remote Caching [Docker version]](https://qiita.com/PINTO/items/49ce77a6ee03b404c156)**  
+Procedure for building a remote caching environment by Bazel.
+**[Ultra-fast build of Tensorflow with Bazel Remote Caching [Google Cloud Storage version]](https://qiita.com/PINTO/items/eef2eccc21dc72227d0f)**
+**[Ultra-fast build of Tensorflow with Bazel Remote Caching [Docker version]](https://qiita.com/PINTO/items/49ce77a6ee03b404c156)**
 ![001](media/001.gif)
 
-Use the Tensorflow Lite FlexDelegate shared library below to execute .tflite models that contain custom operations that cannot be performed by Tensorflow Lite. **`libtensorflowlite.so`**  
+Use the Tensorflow Lite FlexDelegate shared library below to execute .tflite models that contain custom operations that cannot be performed by Tensorflow Lite. **`libtensorflowlite.so`**
 **https://github.com/PINTO0309/TensorflowLite-flexdelegate.git**
 
-A repository that shares tuning results of trained models generated by Tensorflow. Post-training quantization (Weight Quantization, Integer Quantization, Full Integer Quantization), Quantization-aware training.  
+A repository that shares tuning results of trained models generated by Tensorflow. Post-training quantization (Weight Quantization, Integer Quantization, Full Integer Quantization), Quantization-aware training.
 **https://github.com/PINTO0309/PINTO_model_zoo.git**
 
-Cross compilation recommends using **`lhelontra`** repository.  
+Cross compilation recommends using **`lhelontra`** repository.
 **https://github.com/lhelontra/tensorflow-on-arm.git**
 
-Prebuilt binary for Jetson Nano by **`Michael`**.  
+Prebuilt binary for Jetson Nano by **`Michael`**.
 **https://dl.photoprism.org/tensorflow/**
 
 **[Add a custom OP to the TFLite runtime to build the whl installer (for Python)](https://zenn.dev/pinto0309/articles/a0e40c2817f2ee)**, **`MaxPoolingWithArgmax2D`**, **`MaxUnpooling2D`**, **`Convolution2DTransposeBias`**
@@ -72,11 +72,11 @@ Prebuilt binary for Jetson Nano by **`Michael`**.
 |tensorflow-2.12.0rc0-cp310-none-linux_aarch64.whl|○||○||1.24|Ubuntu 22.04 glibc 2.35|
 |tensorflow-2.12.0rc0-cp311-none-linux_aarch64.whl|○||○||1.24|Debian  Bookworm glibc 2.36|
 
-**【Appendix】 C Library + Tensorflow v1.x.x / v2.x.x**  
-The behavior is unconfirmed because I do not have C language implementation skills.  
+**【Appendix】 C Library + Tensorflow v1.x.x / v2.x.x**
+The behavior is unconfirmed because I do not have C language implementation skills.
 **[Official tutorial on Tensorflow C binding generation](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/lib_package/README.md)**
 
-**Appx1. C-API build procedure**  
+**Appx1. C-API build procedure**
 **[Native build procedure of Tensorflow v2.0.0 C API for RaspberryPi / arm64 devices (armhf / aarch64)](https://qiita.com/PINTO/items/4f26f63bf40014bec146)**
 
 **Appx2. C-API Usage**
@@ -128,13 +128,15 @@ $ sudo pip3 install pybind11==2.9.2
 $ pip3 install -U --user six wheel mock
 $ sudo pip3 uninstall tensorflow
 
-$ TFVER=2.10.0
+$ TFVER=2.12.0rc0
 
 $ PYVER=39
 or
 $ PYVER=38
 or
 $ PYVER=310
+or
+$ PYVER=311
 
 $ ARCH=`python -c 'import platform; print(platform.machine())'`
 $ echo CPU ARCH: ${ARCH}
@@ -148,7 +150,7 @@ https://github.com/PINTO0309/Tensorflow-bin/releases/download/v${TFVER}/tensorfl
 **Example of Python 3.x series**
 ```bash
 $ python -c 'import tensorflow as tf;print(tf.__version__)'
-2.10.0
+2.12.0rc0
 ```
 
 **Sample of MultiThread x4**

download_tensorflow-2.12.0rc0-cp310-none-linux_aarch64.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+TFVER=2.12.0rc0
+curl -L https://github.com/PINTO0309/Tensorflow-bin/releases/download/v${TFVER}/tensorflow-${TFVER}-cp310-none-linux_aarch64.whl -o tensorflow-${TFVER}-cp310-none-linux_aarch64.whl
+echo Download finished.

download_tensorflow-2.12.0rc0-cp311-none-linux_aarch64.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+TFVER=2.12.0rc0
+curl -L https://github.com/PINTO0309/Tensorflow-bin/releases/download/v${TFVER}/tensorflow-${TFVER}-cp310-none-linux_aarch64.whl -o tensorflow-${TFVER}-cp310-none-linux_aarch64.whl
+echo Download finished.

download_tensorflow-2.12.0rc0-cp38-none-linux_aarch64.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+TFVER=2.12.0rc0
+curl -L https://github.com/PINTO0309/Tensorflow-bin/releases/download/v${TFVER}/tensorflow-${TFVER}-cp38-none-linux_aarch64.whl -o tensorflow-${TFVER}-cp38-none-linux_aarch64.whl
+echo Download finished.

download_tensorflow-2.12.0rc0-cp39-none-linux_aarch64.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+TFVER=2.12.0rc0
+curl -L https://github.com/PINTO0309/Tensorflow-bin/releases/download/v${TFVER}/tensorflow-${TFVER}-cp39-none-linux_aarch64.whl -o tensorflow-${TFVER}-cp39-none-linux_aarch64.whl
+echo Download finished.