Sign in

I skipped a lot of the past days as for one week it was continuous linear algebra, so there wasn’t much value in updating. I then moved on to AI datasets and then image processing, but it went well.

An idea recently was making my own life advice section posts, but then I have to find time in the week to do it.


SAT

09–14:30 I managed to find perfect concenration to revise my neural network notes in my favourite café

14:30–16:30 went home then went for a run

16:30–20:00 Non-work committments

20;00–22:00 Spanish

SUN

09:00–12:00 revised convolutions

12:00–14:00 read spanish book in park

14:00–16:00 run

1600–1700 errands

17:00–20:00 nonwork

20:00–22:00 reading

Do better next time — visit somewhere less familial on Saturdays


08:30 I was finishing off this idea of lack of replication in dermatology trials

11:30 went to histo. Fortunately zero melanocytic lesions. Cases I remember: dowling-Degos, extramammary pagets, ctcl looking like a capillaritis, atrophoderma, amyloid

I then went for a break as I thought it was one of the spr’s last day, but then they there’s a part next week

15:45. Downloading data set failed. Ran out of space. Nightmare

Then started the Andrew Ng course but I’ve already covered most of it and it is on octave so might not be that useful. So that’s good news overall.

Watched…


Today was a better start, 8:30. And it’s easier to study in the rain

Carrying on with the AI papers

  • Reporting on deep learning algorithms in healthcare. An interesting topic but pretty hard to understand. However, I re-read this morning and used some extra sources and it made sense. It is about : Inconsistencies exist in the literature on how predicted results from deep learning are reported. It showed that Esteva et al. paper didn’t report CIs for AUROC, nor accuracy, PPV, NPV, F1 score. It then spoke about Bland Altman plots, which I didn’t really know, but I found…

Late start again 9:50 (I blame the book)

9:50 CS50 AI lecture on knowledge https://cs50.harvard.edu/ai/2020/weeks/1/


From tomorrow I will start these with an update of my current projects, as that has to be my main focus during the ACF time.

Last night was slightly disastrous. I can only think that my matte-leão teabags have more caffeine than I appreciate as I couldn’t sleep until 00:30. So now I have about 3 hours of sleep to catch up on.

What I learned yesterday

  • def __init__ made more sense. I then later saw this in a neural network in PyTorch
  • functions in python are themselves values. Can use as decorators/wrappers (functional programming paradigm)
  • try: ##function else: ##do…

I wanted to keep a log of what I did in my academic block, as I’m particularly worried about not maximising the time. I liked this blog on looking to improve in multiple areas, and so I wanted something to record my progress — as I often forget the work I did, and then look back thinking I didn’t do enough — and create accountability. So far I’ve been happy with how much work I’ve done, my focus, and I’m happy doing it.

//

So today I started later than I would have liked: 09:20, but I have commute of…

LS

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store