| ... | ... | @@ -13,29 +13,37 @@ |
|
|
|
### Get the Fidle Docker image
|
|
|
|
|
|
|
|
- Get Fidle image : ...it takes some time... (~5mn) and the image is quiet huge (~7 GB) <br>
|
|
|
|
`$ docker pull gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle_cpu:python3.7`
|
|
|
|
`$ docker pull gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle-cpu_oct22:python3.8`
|
|
|
|
|
|
|
|
- Check the image has been pulled :<br>
|
|
|
|
`$ docker images`
|
|
|
|
|
|
|
|
- On standard output, you should have a line of the form <br>
|
|
|
|
`gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle_cpu python3.7 [...] 6.98GB`
|
|
|
|
`gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle-cpu_oct22 python3.8 [...] 6.98GB`
|
|
|
|
|
|
|
|
- You can test it :<br>
|
|
|
|
`$ docker run --rm gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle_cpu:python3.7 python3 /root/fidle_env_test.py`
|
|
|
|
`$ docker run --rm gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle-cpu_oct22:python3.8 fid check`
|
|
|
|
- You should see printed the following lines:
|
|
|
|
<code>
|
|
|
|
FIDLE_DATASETS_DIR = /data/fidle-datasets
|
|
|
|
Python version = 3.7
|
|
|
|
Tensorflow version = 2.8.0
|
|
|
|
Tensorflow GPU/CUDA available = False
|
|
|
|
Pytorch version = 1.11.0+cu102
|
|
|
|
Pytorch GPU/CUDA available = False</code>
|
|
|
|
Check environment :
|
|
|
|
|
|
|
|
Fidle module : Ok (2.0b42)
|
|
|
|
tensorflow : Ok (2.9.0)
|
|
|
|
tensorflow.keras : Ok (2.9.0)
|
|
|
|
numpy : Ok (1.23.4)
|
|
|
|
sklearn : Ok (1.1.2)
|
|
|
|
skimage : Ok (0.19.3)
|
|
|
|
matplotlib : Ok (3.6.1)
|
|
|
|
plotly : Ok (5.10.0)
|
|
|
|
pandas : Ok (1.5.0)
|
|
|
|
jupyterlab : Ok (3.4.8)
|
|
|
|
pytorch : Ok (1.12)
|
|
|
|
torchvision : Ok (0.13.1+cu102)</code>
|
|
|
|
|
|
|
|
|
|
|
|
### Launch Fidle Jupyterlab
|
|
|
|
- Launch Jupyterlab for Fidle : <br>
|
|
|
|
`$ docker run --rm -p 8888:8888 gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle_cpu:python3.7`
|
|
|
|
`$ docker run --rm -p 8888:8888 gricad-registry.univ-grenoble-alpes.fr/talks/fidle/fidle-cpu_oct22:python3.8`
|
|
|
|
|
|
|
|
- Copy the URL printed on screen beginning with `http://127.0.0.1:8888/?token`...
|
|
|
|
- Open a Web Browser and paste the URL
|
| ... | ... | |
| ... | ... | |
