ChainerUI is a visualization and management tool for Chainer.
To install ChainerUI, use pip
$ pip install chainerui
Then initialize ChainerUI database.
$ chainerui db create
$ chainerui db upgrade
To install ChainerUI from source
$ git clone https://github.com/chainer/chainerui.git
$ cd chainerui/frontend
$ npm install && npm run build && cd ..
$ python setup.py install
- ChainerUI uses
sqlite3
module which is included in the Python standard library. To enablesqlite3
module, need to install SQLite library before building Python.- on Ubuntu, install "libsqlite3-dev" (
apt-get install libsqlite3-dev
) before building Python
- on Ubuntu, install "libsqlite3-dev" (
- On Windows, install Visual C++ Build Tools with the Default Install setting, before ChainerUI install.
Try examples.
$ git clone https://github.com/chainer/chainerui.git
$ cd chainerui
$ # create your first project
$ chainerui project create -d examples -n example-project
$ # run ChainerUI server
$ chainerui server
Open http://localhost:5000/ and select "example-project".
$ chainerui project create -d PROJECT_DIR [-n PROJECT_NAME]
The ChainerUI server watches the below files under the project directory recursively.
log
: a JSON file created byLogReport
extension.args
: (optional) a JSON file, which includes key-value pairs you want to see on ChainerUI along with logs. Seesave_args
, util function to dump command line arguments or dictionary toargs
file.commands
: (optional) a JSON file, which is automatically created byCommandsExtension
. This is a file for ChainerUI server to communicate withCommandsExtension
.
path/to/result
|--- log
|--- args
|--- commands
examples/train_mnist.py
, based on chainer/examples/mnist/train_mnist.py
, is a useful example to see how to set save_args
and CommandsExtension
on training script.
$ chainerui server
Open http://localhost:5000/ .
To unregister a result, click X
button at the right end of the result table. If unregister a result once, cannot restore the same name result on the result table (known problem).
To unregister all results, drop and create a new database as follows.
$ chainerui db drop
$ chainerui db create
$ chainerui db upgrade
ChainerUI is supported by the latest stable version of the following browser.
- Firefox
- Chrome
Any contribution to ChainerUI is welcome!
- Python codes follow the Chainer contribution guide.
Install "pytest" package and run test
$ CHAINERUI_ENV=test pytest
$ cd frontend
$ npm install
$ npm run build:watch