Cloudbrain Examples

This repository consists mostly of a README with a list of examples as well as a series of JSON configurations files. The goal is to demonstrate how the different modules of CloudBrain work.

Before you start

Make sure you have CloudBrain installed.

cd <where you want to clone cloudbrain>
git clone https://github.com/cloudbrain/cloudbrain
cd cloudbrain
python setup.py develop --user

You can also refer to the Setup section of the CloudBrain wiki

Setting up cloudbrain_examples

If you want to be able to edit cloudbrain_examples without having to re-install the package, then run:

python setup.py develop --user

Otherwise, run:

python setup.py install --user

Learn how to use CloudBrain's modules

All CloudBrain examples are under cloudbrain_examples/modules. CloudBrain modules can be chained together.

Here is a very simple example ``` * --------------* eeg * --------------* filtered_eeg * --------------* | OpenBCISource | ======> | BandFilter | ==============> | StdoutSink | (prints to console via stdout) * --------------* * --------------* * --------------* | | | | eeg * ----------------* alpha * --------------* ====> | FFTTransform | =======> | PyPlotSink | (plots live data via matplotlib.pyplot) * ----------------* * --------------* ```

Sources: modules sending data

Sources are the modules in charge of sending data to CloudBrain.

• cloudbrain.sources.openbci.OpenBCISource: send data from the OpenBCI to CloudBrain.
• cloudbrain.sources.openbci.MockSource: send mock data to CloudBrain.

Send mock data

To send mock data run:

python -m cloudbrain.run --conf cloudbrain_examples/modules/sources/mock.json

Look at cloudbrain_examples/modules/sources/mock.json for more details about how the module is setup.

Send OpenBCI data

To send OpenBCI data run:

python -m cloudbrain.run --conf cloudbrain_examples/modules/sources/openbci.json

Tip: Make sure your OpenBCI board is connected and that the JSON conf file is using the right port.

Look at cloudbrain_examples/modules/sources/openbci.json for more details about how the module is setup.

Filters: modules filtering data

Filters are the modules in charge of filtering the data in CloudBrain.

• cloudbrain.filters.band.BandFilter: filter that can exclude frequencies (if type = bandstop) or keep only certain frequencies (if type = bandpass).

Band stop filter

To filter frequencies outside of a specified range run:

python -m cloudbrain.run --conf cloudbrain_examples/modules/filters/band_stop.json

Tip #1: You can edit the frequencies you filter out by editing start_frequency and stop_frequency in the JSON conf file.

Look at cloudbrain_examples/modules/filters/band_stop.json for more details about how the module is setup.

Tip #2: You can use this filter to create a notch filter. Depending on your country the notch might be 50Hz or 60Hz. Say it is 60Hz; then you should filter out frequencies in the range [59.0, 61.0].

Band pass filter

To keep only the frequencies in aspecified range run:

python -m cloudbrain.run --conf cloudbrain_examples/modules/filters/band_pass.json

Tip: You can edit the frequencies you keep by editing start_frequency and stop_frequency in the JSON conf file.

Look at cloudbrain_examples/modules/filters/band_pass.json for more details about how the module is setup.

Sinks: getting data out of CloudBrain

Sinks are the modules in charge of relaying the data out of CloudBrain.

• cloudbrain.sinks.stdout.StdoutSink: write data to the console (stdout).
• cloudbrain.sinks.pyplot.PyPlotSink: plot a metric in real-time via matplotlib.pyplot.

Stdout sink: see CloudBrain's in the console

To write data to the console (stdout) run:

python -m cloudbrain.run --conf cloudbrain_examples/modules/sinks/stdout.json

Look at cloudbrain_examples/modules/sinks/stdout.json for more details about how the module is setup.

Tip: this module is useful to know if you are sending data to CloudBrain correctly with a source. Make sure that the settings are the same for publishers and subscribers in the modules that you are chaining (particularly, look at the base_routing_key, metric_name and rabbitmq address and credentials.

Sandbox

In cloudbrain_examples/sandbox you will find snippets of code that might not be directly related to CloudBrain's modules. These snippets were often starting points to create new CloudBrain modules so I thought I would include them in the cloudbrain_examples repo.

What's in cloudbrain_examples/sandbox?

• plot_sine_wave.py: plot data generated by CloudBrain's signal generator.
• band_pass_filter.py: experimentation with band pass filters + graphic comparison of the results.
• plot_data_real_time.py: code written to bootstrap the PyPlotSink module in CloudBrain.
• print_data.py: code written to bootstrap the StdoutSink module in CloudBrain.
• send_sine_wave.py: code written to bootstrap the MockSource module in CloudBrain.
• send_basic_message.py: just a mini demo of how the PikaPublisher works.