forked from NVIDIA-Merlin/HugeCTR
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge branch 'master' into v2.2.1-integration
- Loading branch information
Showing
2 changed files
with
15,602 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,57 @@ | ||
# HugeCTR Jupyter demo notebooks | ||
This folder contains demo notebooks for HugeCTR. | ||
|
||
## 1. Requirements | ||
|
||
The most convenient way to run these notebooks is via a docker container, which provides a self-contained, isolated and re-producible environment for all experiments. | ||
|
||
First, clone the repository: | ||
|
||
``` | ||
git clone https://github.com/NVIDIA/HugeCTR | ||
``` | ||
|
||
Next, follow the step in the [README](../README.md) to build the NVIDIA HugeCTR container. Briefly, the steps are as follows. | ||
|
||
From repo root: | ||
|
||
``` | ||
docker build -t hugectr:devel -f ./tools/dockerfiles/dev.Dockerfile . | ||
``` | ||
|
||
Then launch the container in interaction mode (mount the home directory of repo into container for easy development): | ||
|
||
``` | ||
docker run --runtime=nvidia --rm -it -u $(id -u):$(id -g) -v $(pwd):/hugectr -w /hugectr hugectr:devel bash | ||
``` | ||
|
||
Within the docker interactive bash session, build HugeCTR: | ||
``` | ||
mkdir -p build | ||
cd build | ||
cmake -DCMAKE_BUILD_TYPE=Release -DSM=70 .. # Asuming the target GPU is Volta-generation GPU, such as the Tesla V100 | ||
make -j | ||
``` | ||
|
||
Then finally, install and start Jupyter with | ||
|
||
``` | ||
pip3 install jupyter | ||
cd /hugectr | ||
jupyter notebook --allow-root --ip 0.0.0.0 --port 8888 | ||
``` | ||
|
||
Navigate a web browser to the IP address or hostname of the host machine | ||
at port 8888: ```https://[host machine]:8888``` | ||
|
||
Use the token listed in the output from running the jupyter command to log | ||
in, for example: | ||
|
||
```https://[host machine]:8888/?token=aae96ae9387cd28151868fee318c3b3581a2d794f3b25c6b``` | ||
|
||
|
||
Within the container, the notebooks themselves are located at `/hugectr/notebooks`. | ||
|
||
## 2. Notebook list | ||
|
||
- [movie-lens-example.ipynb](movie-lens-example.ipynb): Training and inference demo on the movie lens dataset. |
Oops, something went wrong.