Tensorflow with GPU Installation
Install NVIDIA Graphics Driver
1.Use NVIDIA proprietary driver. Check installation nvidia-smi.
Sun Apr 14 15:23:21 2019
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.39 Driver Version: 418.39 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce 940MX Off | 00000000:01:00.0 Off | N/A |
| N/A 50C P5 N/A / N/A | 460MiB / 4046MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1600 G /usr/lib/xorg/Xorg 225MiB |
| 0 1758 G /usr/bin/gnome-shell 156MiB |
| 0 4577 C python3 24MiB |
| 0 4877 G ...-token=6290163DAE60CC3BAE4735A7678A09B4 49MiB |
+-----------------------------------------------------------------------------+
To install latest driver:
sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt update
sudo apt upgrade
ubuntu-drivers list
sudo apt install nvidia-driver-VERSION_NUMBER_HERE
To solve
E: Unable to correct problems, you have held broken packages.
Try:
sudo apt-get install --fix-broken xorg-video-abi-11 xserver-xorg-core
or
sudo apt-get remove --purge nvidia-*
sudo ubuntu-drivers autoinstall
sudo service lightdm restart
Install CUDA Toolkit
2.Download CUDA installer runfile. Check version
- CUDA 10.0
- 418-10.0
- 390-9.0
3.Stop display manager. ctrl + Alt + F3.
# disable the graphical target
systemctl isolate multi-user.target
# unload Nvidia drivers
sudo modprobe -r nvidia-drm
4.Execute the runfile installer. Install CUDA Toolkit ans samples.
5.Start graphics.
# start graphical env again
systemctl start graphical.target
6.After the installation finishes, configure the runtime library.
sudo bash -c "echo /usr/local/cuda/lib64/ > /etc/ld.so.conf.d/cuda.conf"
sudo ldconfig
7.Append string /usr/local/cuda/bin to system file /etc/environment so that nvcc will be included in $PATH.
sudo vim /etc/environment
Add :/usr/local/cuda/bin at the end before the ".
8.Reboot.
Test installation
9.
cd /usr/local/cuda-10.0/samples/
sudo make
10.After it completes.
cd /usr/local/cuda/samples/bin/x86_64/linux/release
./deviceQuery
./deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "GeForce 940MX"
CUDA Driver Version / Runtime Version 10.1 / 10.0
CUDA Capability Major/Minor version number: 5.0
Total amount of global memory: 4046 MBytes (4242604032 bytes)
( 4) Multiprocessors, (128) CUDA Cores/MP: 512 CUDA Cores
GPU Max Clock rate: 861 MHz (0.86 GHz)
Memory Clock rate: 2505 Mhz
Memory Bus Width: 64-bit
L2 Cache Size: 1048576 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device supports Compute Preemption: No
Supports Cooperative Kernel Launch: No
Supports MultiDevice Co-op Kernel Launch: No
Device PCI Domain ID / Bus ID / location ID: 0 / 1 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 10.1, CUDA Runtime Version = 10.0, NumDevs = 1
Result = PASS
Install cuDNN 7.0
11.Download from cuDNN download page. Download all three .deb files: runtime lib, developer lib, code samples lib.
12.
sudo dpkg -i libcudnn7_7.5.0.56-1+cuda10.0_amd64.deb
sudo dpkg -i libcudnn7-dev_7.5.0.56-1+cuda10.0_amd64.deb
sudo dpkg -i libcudnn7-doc_7.5.0.56-1+cuda10.0_amd64.deb
Verify installation
13.
cp -r /usr/src/cudnn_samples_v7/ ~
cd ~/cudnn_samples_v7/mnistCUDNN
make clean && make
./mnistCUDNN
Configure the CUDA and cuDNN library paths
14.
# put the following line in the end or your .bashrc file
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/extras/CUPTI/lib64"
source ~/.bashrc
Install TensorFlow-GPU
15. pip install --upgrade tensorflow-gpu.
Try it out
16.
python
>>> import tensorflow as tf
>>> sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
2019-04-14 15:04:49.833765: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-04-14 15:04:49.858219: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2400000000 Hz
2019-04-14 15:04:49.858464: I tensorflow/compiler/xla/service/service.cc:150] XLA service 0x5620497c66b0 executing computations on platform Host. Devices:
2019-04-14 15:04:49.858487: I tensorflow/compiler/xla/service/service.cc:158] StreamExecutor device (0): <undefined>, <undefined>
2019-04-14 15:04:49.984415: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:998] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-04-14 15:04:49.985380: I tensorflow/compiler/xla/service/service.cc:150] XLA service 0x56204907b920 executing computations on platform CUDA. Devices:
2019-04-14 15:04:49.985407: I tensorflow/compiler/xla/service/service.cc:158] StreamExecutor device (0): GeForce 940MX, Compute Capability 5.0
2019-04-14 15:04:49.985716: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1433] Found device 0 with properties:
name: GeForce 940MX major: 5 minor: 0 memoryClockRate(GHz): 0.8605
pciBusID: 0000:01:00.0
totalMemory: 3.95GiB freeMemory: 3.61GiB
2019-04-14 15:04:49.985739: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1512] Adding visible gpu devices: 0
2019-04-14 15:04:49.987838: I tensorflow/core/common_runtime/gpu/gpu_device.cc:984] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-04-14 15:04:49.987862: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 0
2019-04-14 15:04:49.987876: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1003] 0: N
2019-04-14 15:04:49.988147: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 3401 MB memory) -> physical GPU (device: 0, name: GeForce 940MX, pci bus id: 0000:01:00.0, compute capability: 5.0)
Device mapping:
/job:localhost/replica:0/task:0/device:XLA_CPU:0 -> device: XLA_CPU device
/job:localhost/replica:0/task:0/device:XLA_GPU:0 -> device: XLA_GPU device
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce 940MX, pci bus id: 0000:01:00.0, compute capability: 5.0
2019-04-14 15:04:49.989223: I tensorflow/core/common_runtime/direct_session.cc:317] Device mapping:
/job:localhost/replica:0/task:0/device:XLA_CPU:0 -> device: XLA_CPU device
/job:localhost/replica:0/task:0/device:XLA_GPU:0 -> device: XLA_GPU device
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce 940MX, pci bus id: 0000:01:00.0, compute capability: 5.0