Skip to content

Commit 518478e

Browse files
haifeng-jincopybara-github
haifeng-jin
authored and
copybara-github
committed
update the TF 2.12 pip install guide.
PiperOrigin-RevId: 519210355
1 parent fc6d145 commit 518478e

File tree

1 file changed

+39
-47
lines changed

1 file changed

+39
-47
lines changed

‎site/en/install/pip.md

Lines changed: 39 additions & 47 deletions
Original file line numberDiff line numberDiff line change
@@ -26,9 +26,10 @@ step-by-step instructions.
2626
for more information about this collaboration.
2727

2828
```bash
29-
conda install -c conda-forge cudatoolkit=11.2.2 cudnn=8.1.0
30-
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/
31-
python3 -m pip install tensorflow
29+
conda install -c conda-forge cudatoolkit=11.8.0
30+
python3 -m pip install nvidia-cudnn-cu11==8.6.0.163 tensorflow==2.12.*
31+
CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))
32+
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib
3233
# Verify install:
3334
python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
3435
```
@@ -72,9 +73,10 @@ step-by-step instructions.
7273
for CUDA in WSL.
7374

7475
```bash
75-
conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0
76-
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/
77-
python3 -m pip install tensorflow
76+
conda install -c conda-forge cudatoolkit=11.8.0
77+
python3 -m pip install nvidia-cudnn-cu11==8.6.0.163 tensorflow==2.12.*
78+
CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))
79+
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib
7880
# Verify install:
7981
python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
8082
```
@@ -155,8 +157,8 @@ The following NVIDIA® software are only required for GPU support.
155157
156158
* [NVIDIA® GPU drivers](https://www.nvidia.com/drivers){:.external}
157159
version 450.80.02 or higher.
158-
* [CUDA® Toolkit 11.2](https://developer.nvidia.com/cuda-toolkit-archive){:.external}.
159-
* [cuDNN SDK 8.1.0](https://developer.nvidia.com/cudnn){:.external}.
160+
* [CUDA® Toolkit 11.8](https://developer.nvidia.com/cuda-toolkit-archive){:.external}.
161+
* [cuDNN SDK 8.6.0](https://developer.nvidia.com/cudnn){:.external}.
160162
* *(Optional)*
161163
[TensorRT](https://docs.nvidia.com/deeplearning/tensorrt/archives/index.html#trt_7){:.external}
162164
to improve latency and throughput for inference.
@@ -233,17 +235,19 @@ The following NVIDIA® software are only required for GPU support.
233235
nvidia-smi
234236
```
235237

236-
Then install CUDA and cuDNN with conda.
238+
Then install CUDA and cuDNN with conda and pip.
237239

238240
```bash
239-
conda install -c conda-forge cudatoolkit=11.2.2 cudnn=8.1.0
241+
conda install -c conda-forge cudatoolkit=11.8.0
242+
pip install nvidia-cudnn-cu11==8.6.0.163
240243
```
241244

242245
Configure the system paths. You can do it with the following command every time
243246
you start a new terminal after activating your conda environment.
244247

245248
```bash
246-
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/
249+
CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))
250+
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib
247251
```
248252

249253
For your convenience it is recommended that you automate it with the following
@@ -252,7 +256,8 @@ The following NVIDIA® software are only required for GPU support.
252256

253257
```bash
254258
mkdir -p $CONDA_PREFIX/etc/conda/activate.d
255-
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/' > $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
259+
CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))
260+
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib' > $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
256261
```
257262

258263
### 5. Install TensorFlow
@@ -271,7 +276,7 @@ The following NVIDIA® software are only required for GPU support.
271276
PyPI.
272277
273278
```bash
274-
pip install tensorflow==2.11.*
279+
pip install tensorflow==2.12.*
275280
```
276281
277282
### 6. Verify install
@@ -312,7 +317,7 @@ The following NVIDIA® software are only required for GPU support.
312317
conda install -c nvidia cuda-nvcc=11.3.58
313318
# Configure the XLA cuda directory
314319
mkdir -p $CONDA_PREFIX/etc/conda/activate.d
315-
printf 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/\nexport XLA_FLAGS=--xla_gpu_cuda_data_dir=$CONDA_PREFIX/lib/\n' > $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
320+
printf 'export XLA_FLAGS=--xla_gpu_cuda_data_dir=$CONDA_PREFIX/lib/\n' >> $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
316321
source $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
317322
# Copy libdevice file to the required path
318323
mkdir -p $CONDA_PREFIX/lib/nvvm/libdevice
@@ -598,17 +603,19 @@ The following NVIDIA® software are only required for GPU support.
598603
nvidia-smi
599604
```
600605
601-
Then install CUDA and cuDNN with conda.
606+
Then install CUDA and cuDNN with conda and pip.
602607
603608
```bash
604-
conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0
609+
conda install -c conda-forge cudatoolkit=11.8.0
610+
pip install nvidia-cudnn-cu11==8.6.0.163
605611
```
606612
607613
Configure the system paths. You can do it with following command everytime
608614
your start a new terminal after activating your conda environment.
609615
610616
```bash
611-
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/
617+
CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))
618+
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib
612619
```
613620
614621
For your convenience it is recommended that you automate it with the following
@@ -617,7 +624,8 @@ The following NVIDIA® software are only required for GPU support.
617624
618625
```bash
619626
mkdir -p $CONDA_PREFIX/etc/conda/activate.d
620-
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/' > $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
627+
CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))
628+
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib' > $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
621629
```
622630
623631
### 5. Install TensorFlow
@@ -636,7 +644,7 @@ The following NVIDIA® software are only required for GPU support.
636644
PyPI.
637645

638646
```bash
639-
pip install tensorflow
647+
pip install tensorflow==2.12.*
640648
```
641649

642650
### 6. Verify install
@@ -667,73 +675,57 @@ The value you specify depends on your Python version.
667675
<table>
668676
<tr><th>Version</th><th>URL</th></tr>
669677
<tr class="alt"><td colspan="2">Linux</td></tr>
670-
<tr>
671-
<td>Python 3.7 GPU&nbsp;support</td>
672-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-2.11.0-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
673-
</tr>
674-
<tr>
675-
<td>Python 3.7 CPU-only</td>
676-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.11.0-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
677-
</tr>
678678
<tr>
679679
<td>Python 3.8 GPU&nbsp;support</td>
680-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-2.11.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
680+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-2.12.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
681681
</tr>
682682
<tr>
683683
<td>Python 3.8 CPU-only</td>
684-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.11.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
684+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.12.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
685685
</tr>
686686
<tr>
687687
<td>Python 3.9 GPU&nbsp;support</td>
688-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-2.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
688+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-2.12.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
689689
</tr>
690690
<tr>
691691
<td>Python 3.9 CPU-only</td>
692-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
692+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.12.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
693693
</tr>
694694
<tr>
695695
<td>Python 3.10 GPU&nbsp;support</td>
696-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-2.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
696+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-2.12.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
697697
</tr>
698698
<tr>
699699
<td>Python 3.10 CPU-only</td>
700-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
700+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow_cpu-2.12.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</td>
701701
</tr>
702702

703703
<tr class="alt"><td colspan="2">macOS (CPU-only)</td></tr>
704-
<tr>
705-
<td>Python 3.7</td>
706-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.11.0-cp37-cp37m-macosx_10_14_x86_64.whl</td>
707-
</tr>
708704
<tr>
709705
<td>Python 3.8</td>
710-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.11.0-cp38-cp38-macosx_10_14_x86_64.whl</td>
706+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.12.0-cp38-cp38-macosx_10_15_x86_64.whl</td>
711707
</tr>
712708
<tr>
713709
<td>Python 3.9</td>
714-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.11.0-cp39-cp39-macosx_10_14_x86_64.whl</td>
710+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.12.0-cp39-cp39-macosx_10_15_x86_64.whl</td>
715711
</tr>
716712
<tr>
717713
<td>Python 3.10</td>
718-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.11.0-cp310-cp310-macosx_10_14_x86_64.whl</td>
714+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-2.12.0-cp310-cp310-macosx_10_15_x86_64.whl</td>
719715
</tr>
720716

721717
<tr class="alt"><td colspan="2">Windows</td></tr>
722-
<tr>
723-
<td>Python 3.7 CPU-only</td>
724-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.11.0-cp37-cp37m-win_amd64.whl</td>
725-
</tr>
726718
<tr>
727719
<td>Python 3.8 CPU-only</td>
728-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.11.0-cp38-cp38-win_amd64.whl</td>
720+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.12.0-cp38-cp38-win_amd64.whl</td>
729721
</tr>
730722
<tr>
731723
<td>Python 3.9 CPU-only</td>
732-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.11.0-cp39-cp39-win_amd64.whl</td>
724+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.12.0-cp39-cp39-win_amd64.whl</td>
733725
</tr>
734726
<tr>
735727
<td>Python 3.10 CPU-only</td>
736-
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.11.0-cp310-cp310-win_amd64.whl</td>
728+
<td class="devsite-click-to-copy">https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow_cpu-2.12.0-cp310-cp310-win_amd64.whl</td>
737729
</tr>
738730

739731
</table>

0 commit comments

Comments
 (0)