TensorFlow : Use Docker Image2024/06/21

	Install TensorFlow which is the Machine Learning Library. On this example, Install TensorFlow official Docker Image with GPU support and run it on Containers.
[1]	Install NVIDIA Container Toolkit, refer to here.
[2]	This is the example to use TensorFlow Docker (GPU).

root@dlp:~#

docker pull tensorflow/tensorflow:latest-gpu

root@dlp:~#

docker images

REPOSITORY              TAG                          IMAGE ID       CREATED        SIZE
tensorflow/tensorflow   latest-gpu                   21df1084f706   3 months ago   7.35GB

# verify to run [nvidia-smi]

root@dlp:~#

docker run --gpus all --rm tensorflow/tensorflow:latest-gpu nvidia-smi

Fri Jun 21 06:33:13 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 550.90.07              Driver Version: 550.90.07      CUDA Version: 12.4     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 3060        Off |   00000000:05:00.0 Off |                  N/A |
| 30%   40C    P0             35W /  170W |       1MiB /  12288MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+

# verify to run TensorFlow

root@dlp:~#

docker run --gpus all --rm tensorflow/tensorflow:latest-gpu \
python -c "import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))"

2024-06-21 06:34:05.368713: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-06-21 06:34:07.722550: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.734896: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.735184: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.736129: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.736392: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.736620: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.936372: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.936678: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.937006: I external/local_xla/xla/stream_executor/cuda/cuda_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. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2024-06-21 06:34:07.937295: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1928] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 10394 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3060, pci bus id: 0000:05:00.0, compute capability: 8.6
tf.Tensor(2207.085, shape=(), dtype=float32)

[3]	Install TensorFlow Docker Image with Jupyter Notebook.

root@dlp:~#

docker pull tensorflow/tensorflow:latest-jupyter

root@dlp:~#

docker images

REPOSITORY              TAG                          IMAGE ID       CREATED        SIZE
tensorflow/tensorflow   latest-jupyter               7428e0989e67   3 months ago   2.16GB
tensorflow/tensorflow   latest-gpu                   21df1084f706   3 months ago   7.35GB

# run container as daemon

root@dlp:~#

docker run -dt -p 8888:8888 tensorflow/tensorflow:latest-jupyter

a1b34d93925d27bcd792cbf390ef3481e0d147d20e7b4d565a4462737da25c16

root@dlp:~#

docker ps

CONTAINER ID   IMAGE                                  COMMAND                  CREATED          STATUS          PORTS                                       NAMES
a1b34d93925d   tensorflow/tensorflow:latest-jupyter   "bash -c 'source /et…"   19 seconds ago   Up 18 seconds   0.0.0.0:8888->8888/tcp, :::8888->8888/tcp   eager_brahmagupta

# confirm URL

root@dlp:~#

docker exec a1b34d93925d bash -c "jupyter notebook list"

Currently running servers:
http://a1b34d93925d:8888/?token=5317ae138be568ea78853997a1134b2c4c2b7719a99d2e33 :: /tf

Access to the URL above, then it's possible to use Jupyter Notebook.

Matched Content