Playing around with Jupyter
I’ve written some python scripts that talk to a BG7TBL signal generator and to my Tektronix 11801B oscilloscope. I can set the output frequency for the BG7TBL and then dump the measured waveform from the scope over a serial port or GPIB. With a bit of signal processing (FFT and some calculations) it’s possible to calculate the frequency and amplitude of the signal. Combine the signal generator with the amplitude measurements and you have a poor man’s network analyzer. It’s very slow, it takes a couple of minutes to do a sweep of 400 pointer, and it’s probably not very accurate, but it’s good enough to be useful.
I’ve been thinking of ways to create some kind fo GUI for these scripts, and finally found something called Project Jupyter. Jupyter allows you to basically have a virtual notebook where you can enter text and calculations. The killer though, is that you can enter code and do things such as plot the output from a function. I abused this interface a bit to run my Python code and now I can do things like this:
This could be really, really powerful. Imagine jotting down notes and making measurements and then postprocessing data on the fly. All the raw measurements can be be saved so if I want to reprocess the measurements later on I can do that. It should also be fairly easy to take a notebook and convert it into a blog post.
I could really like this.