diff options
Diffstat (limited to 'tensorflow/tensorboard/DEVELOPMENT.md')
| -rw-r--r-- | tensorflow/tensorboard/DEVELOPMENT.md | 123 |
1 files changed, 11 insertions, 112 deletions
diff --git a/tensorflow/tensorboard/DEVELOPMENT.md b/tensorflow/tensorboard/DEVELOPMENT.md index 3ff2c87dab..8e86bf04db 100644 --- a/tensorflow/tensorboard/DEVELOPMENT.md +++ b/tensorflow/tensorboard/DEVELOPMENT.md @@ -2,125 +2,24 @@ ## Launching a Development Instance -The first step is getting a TensorBoard development environment set up. You -should start by making sure you have [nodejs](https://nodejs.org/en/) and -[npm](https://www.npmjs.com/). On Ubuntu, `sudo apt-get install -y nodejs -nodejs-legacy npm`. Ensure your npm version is >=3.0 by running -'npm --version'. If the version is <3.0, run 'sudo npm install npm -g' to -update to the latest version. You may need to open a new terminal window after -updating in order to make use of the newly-installed version. +Run the following to launch a demo of TensorBoard in raw sources mode: -Next, you'll want to install [gulp](http://gulpjs.com/) and -[bower](http://bower.io/), which are used for build tooling and dependency -management respectively. Both must be installed globally: `sudo npm install -g -gulp bower` will do that. +```sh +bazel run third_party/tensorflow/tensorboard/components/tf_tensorboard_d3v4:demo +``` -Then, cd into the TensorBoard directory: - -`cd tensorflow/tensorboard` - -and install dependencies: - -`npm run prep` - -Then, run gulp: `gulp` - -(Don't worry if there are some linter errors.) - -Now you can navigate to -[http://localhost:8000/demo/index.html](http://localhost:8000/demo/index.html) -and play with the demo TensorBoard instance. If you make changes to the source -code, `gulp` should detect it, recompile (if Typescript), and reload your -browser. +Now you can navigate to <http://localhost:6006/demo/index.html> and play with +the demo TensorBoard instance. This will have live source reloading. This demo TensorBoard will have a small amount of demo data generated by [generate_testdata.py](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tensorboard/scripts/generate_testdata.py). You can use [serialize_tensorboard.py](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tensorboard/scripts/serialize_tensorboard.py) to create a realistic demo directory from your own data files. -## Launching TensorBoard with modified source - -If you are developing in open source, and have made some changes to TensorBoard -that you'd like to try out on real data, then you need to regenerate -`dist/tf-tensorboard.html`. - -Run `gulp regenerate`. That will recompile all of the TensorBoard assets, and -produce a new tf-tensorboard.html with your changes. - -Now, you can use `bazel` to launch TensorBoard: - -`bazel run //tensorflow/tensorboard:tensorboard -- --logdir=/path/to/logs`. - -## Updating the vulcanized HTML file (for linux) - -The vulcanized HTML file `dist/tf-tensorboard.html.OPENSOURCE` is the version of -Tensorboard started up by users who install TensorFlow via pip. Today, updating -that file involves using gulp. Future efforts will streamline this process. - -First, `cd` into the `tensorflow/tensorboard` directory within a git repository -(a piper client will not work). Run `npm run prepare`. - -Next, we build some third party JS dependencies via webfiles targets. Run - - bazel build \ - tensorflow/tensorboard/components/tf_imports:d3 \ - tensorflow/tensorboard/components/tf_imports:lodash \ - tensorflow/tensorboard/components/tf_imports:graphlib \ - tensorflow/tensorboard/components/tf_imports:dagre \ - tensorflow/tensorboard/components/tf_imports:plottable - -Users internal to Google should use the internal build tool instead. Move the -output JS binaries into the tf_imports directory. - -Run `gulp vulcanize`. If compilation errors arise (such as those related to -TypeScript), fix them and re-run. This step should update the contents of -`dist/tf-tensorboard.html.OPENSOURCE`. - -Next, we perform some manual find-and-replaces on script `src` paths within -`dist/tf-tensorboard.html.OPENSOURCE`. Manually replace: - -* `<script src="../tf-imports/d3.js"></script>` with `<script src="../d3/d3.js"></script>` -* `<script src="../tf-imports/dagre.js"></script>` with `<script src="../dagre/dist/dagre.core.js"></script>` -* `<script src="../tf-imports/graphlib.js"></script>` with `<script src="../graphlib/dist/graphlib.core.js"></script>` -* `<script src="../tf-imports/lodash.js"></script>` with `<script src="../lodash/lodash.min.js"></script>` -* `<script src="../tf-imports/plottable.js"></script>` with `<script src="../plottable/plottable.js"></script>` - -Also, remove duplicate instances of script includes. Each of those scripts -should only be included once (the first time) within the vulcanized output. - -### Try out the vulcanized Tensorboard HTML output - -To test the vulcanized output, prepare a pip package within a virtualized -environment, and run `tensorboard` after activating the environment. - -To do that, we first create and activate a virtual environment called say -`tf_foo` (Pick your own name.). - - virtualenv --system-site-packages ~/tf_foo - source ~/tf_foo/bin/activate - -Make sure that you have installed `pip` and `virtualenv` beforehand. If not, run - - sudo easy_install pip - sudo pip install --upgrade virtualenv - -Next, we run this command from the `tensorflow directory`. - - tools/google/make_tree.sh --pip_dir=/tmp/pip_dir - -to create a pip package. If you are running within Google, also provide the -`--pending_cl` flag. That script will generate a wheel file (.whl) within -`/tmp/pip_dir`. Lets say that it is -`tensorflow-1.0.0rc2-cp27-none-linux_x86_64.whl`. - -Run - - pip install --upgrade /tmp/pip_dir/tensorflow-1.0.0rc2-cp27-none-linux_x86_64.whl +## Launching TensorBoard Proper -to update the pip installation of TensorFlow within the virtual environment. -Verify that the `tensorboard` command defers to the tensorboard instance -installed within your virtual environment (`tf_foo`) by running -`which tensorboard`. To run tensorboard, start it up as usual within the virtual -environment: +Running TensorBoard automatically asks Bazel to create a vulcanized HTML binary: - tensorboard --logdir=/tmp/my/logdir +```sh +bazel run //tensorflow/tensorboard:tensorboard -- --logdir=/path/to/logs +``` |