diff options
Diffstat (limited to 'tensorflow/g3doc/get_started/os_setup.md')
| -rw-r--r-- | tensorflow/g3doc/get_started/os_setup.md | 99 |
1 files changed, 56 insertions, 43 deletions
diff --git a/tensorflow/g3doc/get_started/os_setup.md b/tensorflow/g3doc/get_started/os_setup.md index 0d98427edd..430bcc38da 100644 --- a/tensorflow/g3doc/get_started/os_setup.md +++ b/tensorflow/g3doc/get_started/os_setup.md @@ -61,58 +61,54 @@ $ sudo easy_install pip $ sudo easy_install --upgrade six ``` -We have also uploaded the CPU version of the binaries to Pypi, so you can -simply install on Linux, Mac or Windows with: +We have also uploaded the binaries to Pypi, so you can +simply install tensorflow on Linux, Mac or Windows with pip install. Note you will need pip version 8.1 or later for the following commands to work on Linux : ```bash $ pip install tensorflow ``` -Note that you will need pip version 8.1 or later for the above command to work on Linux. +For installing the version with GPU support, please use: -For Windows users, you can also install the GPU version of the binary with: ```bash $ pip install tensorflow-gpu ``` -Unfortunately, this command is not yet available for Linux or Mac GPU binaries -due to their sizes exceeding the Pypi limit. -If the above commands do not work on your system or you want to install the GPU version -of the binary on Linux or Mac, you can follow these instructions: +If the above commands do not work on your system, you can follow these instructions: ```bash # Ubuntu/Linux 64-bit, CPU only, Python 2.7 -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp27-none-linux_x86_64.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp27-none-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp27-none-linux_x86_64.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp27-none-linux_x86_64.whl # Mac OS X, CPU only, Python 2.7: -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py2-none-any.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py2-none-any.whl # Mac OS X, GPU enabled, Python 2.7: -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py2-none-any.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0-py2-none-any.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.4 -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp34-cp34m-linux_x86_64.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp34-cp34m-linux_x86_64.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.5 -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp35-cp35m-linux_x86_64.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp35-cp35m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp35-cp35m-linux_x86_64.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp35-cp35m-linux_x86_64.whl # Mac OS X, CPU only, Python 3.4 or 3.5: -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py3-none-any.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py3-none-any.whl # Mac OS X, GPU enabled, Python 3.4 or 3.5: -$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py3-none-any.whl +$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0-py3-none-any.whl ``` Install TensorFlow: @@ -154,14 +150,14 @@ Both distributions include pip. To install the CPU-only version of TensorFlow, enter the following command at a command prompt: ```bat -C:\> pip install --upgrade https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-0.12.0rc1-cp35-cp35m-win_amd64.whl +C:\> pip install --upgrade https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-0.12.0-cp35-cp35m-win_amd64.whl ``` To install the GPU version of TensorFlow, enter the following command at a command prompt: ```bat -C:\> pip install --upgrade https://storage.googleapis.com/tensorflow/windows/gpu/tensorflow_gpu-0.12.0rc1-cp35-cp35m-win_amd64.whl +C:\> pip install --upgrade https://storage.googleapis.com/tensorflow/windows/gpu/tensorflow_gpu-0.12.0-cp35-cp35m-win_amd64.whl ``` You can now [test your installation](#test-the-tensorflow-installation). @@ -216,37 +212,37 @@ Now, install TensorFlow just as you would for a regular Pip installation. First ```bash # Ubuntu/Linux 64-bit, CPU only, Python 2.7 -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp27-none-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp27-none-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp27-none-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp27-none-linux_x86_64.whl # Mac OS X, CPU only, Python 2.7: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py2-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py2-none-any.whl # Mac OS X, GPU enabled, Python 2.7: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py2-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0-py2-none-any.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.4 -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp34-cp34m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp34-cp34m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.5 -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp35-cp35m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp35-cp35m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp35-cp35m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp35-cp35m-linux_x86_64.whl # Mac OS X, CPU only, Python 3.4 or 3.5: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py3-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py3-none-any.whl # Mac OS X, GPU enabled, Python 3.4 or 3.5: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py3-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0-py3-none-any.whl ``` Finally install TensorFlow: @@ -368,37 +364,37 @@ select the correct binary to install: ```bash # Ubuntu/Linux 64-bit, CPU only, Python 2.7 -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp27-none-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp27-none-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp27-none-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp27-none-linux_x86_64.whl # Mac OS X, CPU only, Python 2.7: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py2-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py2-none-any.whl # Mac OS X, GPU enabled, Python 2.7: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py2-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0-py2-none-any.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.4 -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp34-cp34m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp34-cp34m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.5 -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp35-cp35m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0-cp35-cp35m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5 # Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below. -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp35-cp35m-linux_x86_64.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0-cp35-cp35m-linux_x86_64.whl # Mac OS X, CPU only, Python 3.4 or 3.5: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py3-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py3-none-any.whl # Mac OS X, GPU enabled, Python 3.4 or 3.5: -(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py3-none-any.whl +(tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0-py3-none-any.whl ``` Finally install TensorFlow: @@ -466,7 +462,7 @@ code. code. We also have tags with `latest` replaced by a released version (e.g., -`0.12.0-rc1-gpu`). +`0.12.0-gpu`). With Docker the installation is as follows: @@ -651,6 +647,12 @@ sudo cp -P cuda/lib64/libcudnn* /usr/local/cuda/lib64/ sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn* ``` +##### Install other dependencies + +```bash +$ sudo apt-get install libcupti-dev +``` + #### Optional: Install OpenCL (Experimental, Linux only) In order to build or run TensorFlow with OpenCL support, both OpenCL (>= 1.2) @@ -755,7 +757,7 @@ export PATH="$CUDA_HOME/bin:$PATH" ``` Finally, you will also want to install the [CUDA Deep Neural -Network](https://developer.nvidia.com/cudnn) (cuDNN v5) library which currently +Network](https://developer.nvidia.com/cudnn) (cuDNN v5.1) library which currently requires an [Accelerated Computing Developer Program](https://developer.nvidia.com/accelerated-computing-developer) account. Once you have it downloaded locally, you can unzip and move the header and @@ -868,7 +870,7 @@ $ bazel build -c opt --config=cuda //tensorflow/tools/pip_package:build_pip_pack $ bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg # The name of the .whl file will depend on your platform. -$ sudo pip install /tmp/tensorflow_pkg/tensorflow-0.12.0rc1-py2-none-any.whl +$ sudo pip install /tmp/tensorflow_pkg/tensorflow-0.12.0-py2-none-any.whl ``` ## Optimizing CPU performance @@ -1112,6 +1114,17 @@ Step #1 might already solve the problem, however if it still persists, execute s This issue occurs with new Anaconda installations when `pip` tries to remove `easy-install.pth`. This file is not included in Anaconda packages, which causes the `pip` installation to fail. +#### Cupti_wrapper.cc: Could not find cuptiActivityRegisterCallbacksin libcupti DSO + +If, when running a TensorFlow Python script, you encounter the following error: +``` +c:\tf_jenkins\home\workspace\nightly-win\device\gpu\os\windows\tensorflow\core\platform\default\gpu\cupti_wrapper.cc:59] Check failed: ::tensorflow::Status::OK() == (::tensorflow::Env::Default()->GetSymbolFromLibrary( GetDsoHandle(), kName, &f )) (OK vs. Not found: cuptiActivityRegisterCallbacks not found)could not find cuptiActivityRegisterCallbacksin libcupti DSO +``` + +Add `<path-to-cuda-folder>\NVIDIA GPU Computing Toolkit\CUDA\v8.0\extras\CUPTI\libx64` to your `PATH`. + +This issue occurs because on CUDA 8.0 the location of the file `cupti64_80.dll` is not on `PATH` by default. + ### Linux issues |