Sometimes I get too excited about my projects and just want to talk about them all, all the time, so that’s what this is. A chance to catch you all up on everything that’s going on, but also to spill about everything.

What’s almost done:

Curveballs v Fastballs (https://github.com/Katherine89Lee/Curveball-Data)

It’s kind of wild to consider that this is almost done. There’s so much in the jupyter notebook for this that sometimes I just miss what I’ve done already.

This started mostly as a fun, “small” project to measure efficiency of curveballs versus fastballs, and I thought it wouldn’t take much. However, I’ve really pushed a lot of my knowledge with this one, and it’s really helped me learn just by finding new things to do and questioning what the best way is to show my findings.

All I have left for this is to visualize a few things in a way that’s easy to understand, and then I’m going to take a deep dive into creating both a presentation *and* an infographic for it. There’s a lot I’ve found from this analysis, and even though I’m not a big baseball nut, I’ve had a lot of fun digging.

Shrimpnalysis – En Memoriam for our late sweet kitty (https://github.com/Katherine89Lee/Shrimp-Analysis-in-Memory-of-Lilo)

This was a project I created to work through my grief after our cat passed away. She loved shrimp, so I went with a shrimp time series analysis, with the hopes of coming up with a final analysis that pushes the envelope of what I know and how to visualize information.

The hope is that I can make something either animated, or interactive, and add her little face in the corner. I know it’s possible, but it’s a good project to get me to dig into functions and annotations and more involved programming concepts.

Tours and Travels ( https://github.com/Katherine89Lee/Tour-and-Travel-Data-Analysis )

This was a dataset that I had downloaded for my portfolio builder class, in case the original one I went with didn’t work out. I wanted to continue working on it anyway just to get more practice with ML concepts, and I’m basically at the modeling stage now.

The only thing left is to figure out what kind of model would be best. I don’t want to go with the same model I used for the customer churn set, and I want to learn something. So I need to look into other models that would work, and then I’ll do a full blog post/report on that.

What’s up next:

Grocery Trends through the Pandemic ( https://github.com/Katherine89Lee/Personal-Grocery-Project )

Originally I was going to say this was almost done, but I still have more analysis to do for this overall before I can get to a final point. I’m struggling with sorts and how to show an accurate trend of the larger grocery bills, but I’ve made bigger leaps lately in getting to where I want to be with this.

I don’t know what the final result of this is going to be, because I never figured I’d be done. I think of this more as an ongoing project, but i think the bulk of it is almost finished. I may just do a powerpoint presentation for this one, and consider that a full report of my findings.

I know that I do want to add this dataset to a public directory, like Kaggle or somewhere, for other people to mess with. I don’t know how helpful it’ll be for other people, but I know I went through a lot of work putting this whole dataset together, I should get it out in the world.

Frognalysis – Calaveras Jumping Frog Contest Winners (https://github.com/Katherine89Lee/Calaveras-Jumping-Frog-Analysis)

The roadblock for this one comes with trying to visualize categorical data without overplotting. I want to try to do a map bubble plot (all of the cities/regions the winning owners are from), but that’s going to take some poking around. It’s not simple code.

I have come up with a good visualization for this one and the overall distances jumped across time with a scatterplot.

The dataset for this one is small. There’s also pictures involved with this one, but that was more for record keeping than for any visual analysis of any kind.

Ice Cream Code (https://github.com/Katherine89Lee/Ice-Cream-Code)

For this one, I’ve only uploaded it so far and then explored a little bit of the datasets. There are images as well with everything that I downloaded, so I’m considering maybe doing a machine learning thing with those. That would be brand new for me to tackle. Which is scary at the same time that it’s fun.

I could do a number of things with this. Some fun analyses with final vizzes, or some modeling or forecasting.

IMDB Movie Ratings (https://github.com/Katherine89Lee/IMDB-Movie-Rating-Project)

This one is tough. I have a wordcloud finished on this one that took me a while to get to and get finished, but I’m wondering how else I can visualize the rest of the information in the dataset and what to do with it.

I love the wordcloud, and I’ll probably end up customizing it a little and using it as one of the final visualizations.

The other idea is finding another, or two more other, datasets to put together with this one and then create a larger analysis with all of that information. I may actually end up doing that, but unsure if it’ll all add up.

Ah, right, the original end goal for this was to build a recommender system! Another of the ML project ideas I had wanted to tackle and understand due to the portfolio builder class.

Video Game Sales Time Series Analysis (https://github.com/Katherine89Lee/Video-Game-Sales-Time-Series-Analysis)

This one… This one is giving me issues. There are a bunch of rows in this dataset that are just straight up 0s. On the other hand, the dataset is one of the largest I’ve dealt with, at 20,000+ entries in total. I need to do some more explorative analysis of this one just to see what I can come up with and what would be the best way forward.

My hope for this one… I don’t know that I have an end goal. I think I just wanted to play around with it, and for that, the hopeful result will probably end up being just a basic viz. That may change as I get into it a bit more.

What I’m hoping to dig into soon:

Final Fantasy Chatbot

This one, I’m really excited about but also super intimidated to work on. I need to learn more about chatbots in general and how to make one work, what the essential code is and how to get it functional.

This is… A massive project in my eyes, but with it being Final Fantasy, I’m already super interested in it. Getting through the ML portfolio builder gave me a bit more confidence in doing things with ML concepts, as I mentioned earlier, and I’d love to put those lessons into practice with something completely different.

Also it would just be straight up fun to create this and let it loose on my friends who are also Final Fantasy buffs.

Open Source project contribution

This is something I’ve only considered doing recently, because I’m unsure if there’s anything I can contribute to, and do so in a meaningful way at that.

I’m just in the searching phase right now, but what I’ve found is that it’s mostly ML projects that are open sourced. No surprise there, but my knowledge in that end of data science is a little limited there. I’ll keep looking, though, and hopefully I can call myself a contributor soon!

Thanks so much for reading through all of this. I’m not sure what next week’s post will be, but if I finish one of the projects, it will likely be about that!

Until then, I hope that all your data cleaning is swift, and your vizzes come out gorgeous.