Tuesday Very Quiet

Tuesday started with me rising after 8 and feeling more rested. After putting on my slippers, I could easily reach the kitchen; it had not moved. I have a new rule that I must wear shoes or slippers. My toes are numb to various degrees (chemo, diabetes, hammer toe, and flat feet all contribute), and I can easily injure my toes and not know how bad. I have found I slam them into things harder with the reduced feeling. Shoes and slippers are required now.

I made liberal coffee from Equal Exchange’s French Roast grounds. Tuesday promised a sunny fall day in Oregon, but that will soon be a rare event as the rain will be all day, and sunshine will only be in my cup in the mornings. I take a deep drink of the dark, liberal caffeinated beverage, tasting the sunshine of helping people and creating a fair tax system in the USA. Dear readers, have some liberal coffee; it will help!

I had oatmeal with my coffee for breakfast.

On Tuesday morning, I wrote slowly and carefully. Still, when I published the blog, I reread it and corrected five make-me-wince errors. I was distracted by texting, news stories, and a few calls. This was not bad; I have no plans and am retired.

I made a late lunch. I boiled some yellow potatoes that were mashed with butter once soft. I took frozen Brussels sprouts and microwaved them (flipping them to the correct “This Side Up” side, oops). These were mushy and green. Not great. I put butter with a dash of salt (I use unsalted butter) in the non-stick fry pan (thanks, Steve). I fried them until brown and then black on the sides, stirring often to change sides. Sprouts are good, slightly burned. I microwaved a frozen piece of eggplant parmesan I made a few weeks ago. I did not get it hot enough, but it was at least warm, and I ate it. It was excellent.

I finally did the dishes and put away the laundry. The orchids get misted all day. I watched Kaos on Netflix for two more episodes. While the show is clever, it is hard to find any sympathy for the characters. The sense that no character is really controlling their fate, a very Greek myth theme, makes the show seem a bit artificial. It is well done and set in modern times but revised to be a Greek god-run society.

I read but rose and then returned to Python coding for the Kaggle coding contest. I have poor results using a random forest of trees AI classifier, 60%, but I am not surprised as I am using only some of the data supplied. I also made no effort to reduce the features; there are less than sixty, and some likely overlap and bring little value, but with these few, the results are likely good for the limited data.

I changed some of the parameters of the AI classifier and did not get better results (again, suggesting I have good-as-it-gets results). The primary data is insufficient to produce the required results—no surprise. The time series data must be consumed and provide more information to create a better outcome. No matter. I decided to submit the notebook to get credit for trying on the contest, but all the coding is done in a Jupiter Notebook in Python, and I cannot figure out what I need to do. I am missing a step or an approval. I decided to work that out on Wednesday (I forgot to write the submission file, which is rather basic).

More information on AI stuff: I have a confusion matrix that shows how my solution fits, and I see two areas where I am missing. I am interested in looking through the time series data and seeing if there are ample data sets to match incorrect results. In my twisted thinking, I might just count and update my features with a simple count value.

The Hillsboro Machine Learning Python Meetup meets at 6 on the first Tuesday of each month (today) for a Zoom call. Today, Ernest, the group leader, presents on neural networks and large feature number solutions with text processing. Again, the small amount of code required to produce a working result is surprising. Ernest reminds me that our last meeting discussed time series processing (quite tricky) and some options. I should look back at the notes and video. It was an excellent and better-attended meeting.

After the meeting, I coded for a while, then read and made a late dinner while watching Kaos. I opened a can of baked beans. I have a case of the smaller cans; they were cheap on Amazon. I heated the beans on the stove and enjoyed them while relaxing. Yes, there is no meat on Tuesday, just veggies (but with butter and maybe a tiny bit of bacon that comes in the beans); thus, there is no vegan credit for me.

I read more Slow Horses, and the book matches the show in terms of words and story. I’m unsure if I need to read this, as I have seen it, and it is a near match to the show. I will try to finish the first book.

I shower, put on my PJs, and sleep early. I wake at 2, not 3, for proof of hydration. I wake at 4, not 5, and roll over and manage to sleep.

Thanks for reading.