splike

That's funny, I've just built the same thing for Asda in the UK https://github.com/markDunne/asdabot

It can search for items, add them to the basket, picks a delivery slot and does the checkout.

With a little more scaffolding in markdown files, this now takes care of my weekly shopping.

I'm interested, could you expand on that?

You can use openai embeddings in elastic if you don't want to use their elser sparse embeddings

I don't have the link, but I also remember reading this story. I tried googling but didn't have any luck either. I just wanted to let you know that you're not misremembering!

For a while I've wanted to build a big red "Ship it!" button (think emergency stop button). When slammed, it runs `git add -u . && git commit -m "Shipped!" && git push`

SEEKING WORK

Location: UK/France/Switzerland

Remote: Yes or hybrid

Willing to relocate: Yes

Technologies: Python, Pandas, Numpy, PyTorch, ML/AI, etc. with a healthy sprinkling of devops

Résumé/CV: https://markd.ie/public/Resume.pdf

Email: markdunne02 [at] gmail.com

Summary: Most valuable when bridging the gap between data science and software engineering. Expertise in broad spectrum of industries, and a passion for jumping into brand new problems. Have considerable expertise in genomics (CRISPR/TALEN gene editing on-target and off-target search) as well GIS development. Absolutely love solving problems that combine complexities of the real world with computer science.

Open to full time or contract work

Location: UK/France/Switzerland

Remote: Yes or hybrid

Willing to relocate: Yes

Technologies: Python, Pandas, Numpy, PyTorch, etc. with a healthy sprinkling of devops

Résumé/CV: https://markd.ie/public/Resume.pdf

Email: markdunne02 [at] gmail.com

Summary: Most valuable when bridging the gap between data science and software engineering. Expertise in broad spectrum of industries, and a passion for jumping into brand new problems. Have considerable expertise in genomics (CRISPR/TALEN gene editing on-target and off-target search) as well GIS development. Absolutely love solving problems that combine complexities of the real world with computer science.

Open to full time or contract work

Someone with more in depth knowledge can tell me why this isn't a good idea, but I'm pretty sure you can get something very close to this with graphql subscription query against hasura backed with postgres

One factor of this battle against trickle down economics is language. 'Trickle down economics' is term that is intuitive, explains itself well, and is easy to remember. It's a meme in Dawkin's original definition of the word.

The alternative term, 'bubble up economics', needs to the popularised more.

This looks like an incredible trip.

Can I ask how did you safely organise a tour like that without getting scammed/putting yourself in danger? How did you get in contact with your driver?

I agree that Discord seems like an odd place to find this information, but I'm curious to know what prevents you from creating a Discord account?

Where might I learn why that is?

Could you give examples of megaprojects happening outside of the west?

I find comments like this one really strange.

What do you think is more likely? That there is not a single stream of water hitting it, or there are plenty of fire fighters at the scene you just happen not to have seen any pictures of them?

I would really like to see something like this for probability

I love typora, but this is open source which makes it extendable.

Am I wrong in thinking that this algorithm could only be applied to continuous variables/hyperparameters?

The most farcical part of Formula E for me will always be the Fan Boost [0]. I almost couldn't believe it when I saw it on TV for the first time. Fans vote for their favourite driver, and that driver then remotely receives a boost to their car. I thought it was a joke.

[0] http://fanboost.fiaformulae.com/

I wonder if this was in response to the top comment on this HN post from 3 days ago https://news.ycombinator.com/item?id=16237026

I just wanted to point out Typora as a nice interface for a markdown editor, it's not really a note taking app.

I'm using Dynalist, which is definitely an improvement over Google Keep, but I'm starting to find it quite restricting that everything has to be in bullet point lists. What I really want is something thats between Dynalist with its Zooming mechanism, and Typora with its super nice markdown editor.

Indeed it is.

My workaround for the failures of pickle is to use pathos. The API for the multiprocessing module is almost identical to the default multiprocessing module which makes it easy to use as a drop in replacement. Pathos uses Dill to serialise, which can serialise many more things, but you have to pay the price in performance. Although again, you can't swap out Dill for the serialiser of your choice.

We'll cross that bridge when we come to it I guess, but we work almost exclusively with human and mouse genomes for now.

In any case, I imagine the Ensembl ID is still safer than other encodings in the case of invertebrates. For example, genes IDs in the Fruit fly genome look like FBgn0034730.

That's useful, thanks!

Interestingly, genetic biologists are probably more aware of this problem than most. When importing a CSV containing gene names such as SEPT2 or MARCH1, they automatically get converted to dates by Excel. This has potentially had a fairly large effect on research in the area [1]. One of the many reasons we insist on only using Ensembl IDs for genes at my company.

[1] https://genomebiology.biomedcentral.com/articles/10.1186/s13...

I've only seen Edward for the first time today, and it looks quite nice.

Could you (or anyone else) comment on why would you choose Edward over something PyMC3? How does it compare?

I think a word2vec guided brute force approach to the riddles would be the winning combination.

I had assumed that the AR side of the problem would be the easy bit and that the real issue would be solving the riddles.