What did you do this past week?
I spent the majority of my time working on the Netflix project with my partner. We built the function signatures and file structures by mimicking the Collatz directory and I spent some time figuring out how to use getline to parse in the “probe.txt” file after that. On Tuesday night, this Github page helped us lot in figuring out what is going on in Checktestdata although MATCH(<value> str) displays some interesting behaviors we still can’t understand …
Boost serialization really frustrated us a lot, since it kept throwing at us this std::bad_alloc error. So, on Wednesday night, Yuhan actually re-downloaded the entire Docker and pulled the Docker image again, which magically solved the issue. ¯\_(ツ)_/¯ I was trying to scp/rsync the training data so we could build a cache based on that, but I found out later that we can just build our cache on the cs machine without moving that gigantic dataset around …
Thursday was when things started to get exciting since the deadline was approaching. Yuhan and I actually had a 7-hour Zoom call to build the cache and push our project to completion. We made the strategic mistake of waiting for other people’s caches instead of building our own early on.
Anyway, we greatly appreciated the AnswerCache built by Colette and Kevin since we built a big cache representing a customer's attitude towards movies in a specific decade [CustomerID(int) -> (decadeIndex(short) -> Avg Rating that Customer has given for movies in that decade)] by extracting data out of that answer cache (which was a great idea proposed by Yuhan). And this approach successfully brought our RSME for probe.txt down to around 0.80 in the end.
What’s in your way?
I have to be more focused and free myself from distractions (there are too many) when working on projects and assignments to boost efficiency. I wish my Wi-Fi signal could stay strong!
What will you do next week?
I will study for the upcoming exam and get started on the next project. I am definitely not positive on doing a 7-hour Zoom call again, so we should probably start earlier and encounter issues earlier.
What was your experience of RMSE, lambdas, function objects, Boost serialization, and iterators?
I know Root Mean Square Error pretty well since I have been taking many Math and Statistic courses.
Lambdas, function objects, and iterators are the subjects I know but never got the chance to get a lot of practice on. Thus, I can’t really say much about my experience with these things. I didn’t know Boost serialization before, but now I see how it could be very valuable when handling big files.
By the way, this statement below is so rich.
“When designing algorithms, design them with the WEAKEST possible iterator; when designing containers, design them with the STRONGEST possible iterator.”
How are you doing and holding up? What’s been most helpful for you in terms of support at this time?
I am doing well. Actually, I am running out of things to say here. In general, it’s just good to stay in touch with family and friends. I can’t really imagine what the situation would be like if I am still on West Campus right now. ¯\_(ツ)_/¯ In terms of this class, helps from Professor Downing and classmates are really awesome supports!
What made you happy this week?
I can’t stop playing this happy video and I want to share it here.
What’s your pick-of-the-week or tip-of-the-week?
This process of building the makefile, the header files, and integrating the unit tests is really worth learning and practicing. Sometimes I don’t know if it works correctly or not, so I would just use a Foo or a Bar to test the code locally. Right setups at the start of a project could prevent later struggles. I see experienced programmers building little tools for themselves to convenience themselves later. I guess that must be the spirit (or one of the many spirits) underlying the programmers’ community.
