Last Dry Run
in Metis
I have been working at the Metis site today.
I caught an earlier train immediately after dropping my daughter off at high school. It is still rather close and depends on finding parking and walking to the train stop in time. But that’s the express train. It is actually about two-thirds the travel time. If I miss that one, the next available train should get me down here in time before things start for the day.
Last night I caught a curiosity. I find a remark that a particular event shouldn’t be deemed terribly noteworthy because it’s still within the customary 5% confidence level. That seemed like something I could try to replicate. I continued that effort today. For a while I was conviced the comment was wrong because the significance seemed much greater. The trouble was that I had missed an implicit assumption about the data table I was reading. I replicated the work properly and had to agree with the comment.
What was good about this endeavor, this curiosity, was that I was able to go “out there” and find the data and bring it back through data cleaning, munging, through Pandas and use the same techniques I’ve been learning in these MOOCs.
This is the sort of thing I foresee we’ll be doing a lot in this Metis Bootcamp. Take the techniques and go out in the wild to see what fun we can have.
Today, the actual instructors are down here. They’re working through their overall plans since it will be a team taught experience. I had a bit of time to talk. But like Wednesday, I’m not really supposed to be here. So when or if they leave, I’ll have to pack up and head out.
I also made good progress in the MOOC. I should have no problem finishing up the second course before the bootcamp. But I may not be able to keep up even a slowed pace on the next two courses. David (one of the instructors) was rather skeptical of that idea. No worries. The main idea behind the Harvard series was to prepare me for the bootcamp. If the bootcamp demands preempts it, this will be fine.
The recent work in this MOOC was interesting. I had to learn a new Python module as I mirror the work. The module is patsy. It was rather specifically built to mimic or match the functionality in R. But… it is slightly different enough to have given me a bit of work to understand it. The interesting thing is apparently there are some ways it improved upon the behavior in R.
EDIT…
Now, there are a couple of things for this dry run that are simply different from previous trips.
How could I have truly tested the main experiences of the commute if I didn’t do of the typical commuter things? It’d be like going to the Alps and not any skiing. Or going to South Africa and visiting no wildlife.
What experience have I been avoiding?
Using the laptop while on the train.
To be honest, I am not certain I’ll be doing more if it. I’ve seen folk doing it. Maybe I’ll be more like everyone else when I have stuff I want to wrap up or explore. But, you just end up being so cramped and squished. But I’ve proven it in, so to speak. I’m typing this up as others board the train.
The other thing is that this was the first time I’m returning in the same pattern as I’ll be doing for the next few weeks. I made the walk in 18 minutes. Doable at a fast clip. But pointless. There’s an oddity with the train schedules. There are lots of trains outbound between five and six. But half or more have express segments, passing many stops including mine. If I don’t get on a train shortly after 5, the next viable one is at about a quarter to six.