REIP: IoT Data Pipelines

REIP is a reconfigurable framework that allows users to quickly build data streaming pipelines for IoT-based sensor research projects.

It is designed for mid-tier IoT devices that are small enough to be placed in the field, but still can run more complex operations like video processing and machine learning.

Quickstart

Chain Interface – Javascript-esque method chaining.

import reip
import reip.blocks as B
import reip.blocks.audio

# Define your pipeline
########################

# record audio and buffer into chunks of 10
audio = B.audio.Mic(block_duration=1)
audio10s = (
   audio.to(B.Debug('Audio'))
   .to(B.Rebuffer(duration=10)))
# plot a spectrogram
spec_img = (
   audio10s.to(B.audio.Stft())
   .to(B.Debug('Spec'))
   .to(B.Specshow('plots/{time}.png')))
# to wavefile
wav = (
   audio10s.to(B.Debug('Audio 222'))
   .to(B.audio.AudioFile('audio/{time}.wav')))

# Execute your pipeline
########################

reip.default_graph().run()

Function Interface - Mimicking a Keras interface.

# Define your pipeline
########################

# record audio and buffer into chunks of 10
audio = B.audio.Mic(block_duration=1)
audio10s = B.Rebuffer(duration=10)(audio)

# to spectrogram
stft = B.audio.Stft()(audio)
specshow = B.Specshow('plots/{time}.png')(spec)
# to wavefile
wav10s = B.audio.AudioFile('audio/{time}.wav')(audio10s)

# Execute your pipeline
########################

reip.default_graph().run()

Feel free to mix and match!

audio = B.audio.Mic(block_duration=1)
# functional
stft = B.audio.Stft()(audio)
# chain
spl = audio1s.to(B.audio.SPL(calibration=72.54, weighting='ZAC'))

Custom Data Processing Blocks

Above we were using blocks built-in to the library. It’s also extremely easy to create your own blocks that can process data. For example:

class MyOwnProcessingBlock(reip.Block):
   def init(self):
      print("I run when we start the thread")
      self.log.info("Hi!")

   def process(self, audio, meta):
      # we'll calculate audio MFCCs
      # docs: https://librosa.org/doc/latest/generated/librosa.feature.mfcc.html#librosa.feature.mfcc
      mfccs = librosa.feature.mfcc(y=audio, sr=meta['sr'], dct_type=2)
      return [mfccs], meta

   def finish(self):
      print("I run when we finish the thread")

Getting Started:

Examples

API