desku

7 ms·

1.
by desku 3y ago
I’d doubt it’s even that complicated, a few if-statements would be enough.
2.
by desku 4y ago
How is £$165/year not nutty? That is more than I pay per year for Amazon Prime, Spotify, Disney+, MLB or The Economist. All for basically being able to just write text in a browser and store it.
3.
by desku 4y ago
Ah yes, can't wait for a 250 page book that should've been a blog post full of high-level content that you can easily find online, padded out with tenuously linked anecdotes.
4.
GloVedle – Wordle using GloVe embeddings(bentrevett.com)
1 points by desku 5y ago|0 comments
5.
by desku 5y ago
And how does this company make money? Harvesting and selling all of my data?
6.
by desku 5y ago
I’ve been interested in making an open source version of the Seeing AI app ( https://www.microsoft.com/en-us/ai/seeing-ai ) with a more limited scope. Contact me if this sounds inline with your ideas.
7.
by desku 5y ago
Can’t figure out a way to contact you from your About section —- I’m a ML engineer with an early interest in algorithmic trading, would be interested in brainstorming ideas.
8.
by desku 5y ago
I’m looking for collaborators for re-implementing “modern” machine learning and deep learning models/papers. Modern is in quotes as I’d actually like to focus less on the super recent, and more on those around ~5 years old, as the comp
9.
by desku 5y ago
Nobody should feel alone at Christmas. If anyone needs someone to talk to, about anything at all, feel free to contact me. Details in my profile.
10.
by desku 5y ago
This looks great. Is it possible to get this as an RSS feed instead of an e-mail newsletter?
11.
by desku 6y ago
Here's a paper on how BERT (a large Transformer model trained using self-supervised learning) implicitly learns the traditional NLP pipeline: https://arxiv.org/abs/1905.05950
12.
by desku 6y ago
Is there actually a new PyTorch version? I can't seem to find anything about it online.
13.
by desku 6y ago
What niche libraries do you think PyTorch is lacking? Do you have some examples of ones that exist in Tensorflow with no PyTorch equivalent?
14.
by desku 6y ago
I very much enjoyed, and would recommend, White Noise and Mao II. As for adjacent authors, I think I'd probably list: David Foster Wallace, Pynchon, Franzen and Roth.
15.
by desku 8y ago
The link to the .pdf seems to be broken though.
16.
by desku 8y ago
I'm not sure how to interpret the argument of the article or the results in the appendix here. The first table shows AdamW having the best results, which follows the argument of the article. However, the following three tables all have
17.
Tutorials on Sentiment Analysis Using PyTorch and TorchText(github.com)
2 points by desku 8y ago|0 comments
18.
by desku 9y ago
It's probably domain/industry/company dependent, but the vast majority (>90%) of NLP work I do nowadays is sequence-to-sequence models.
19.
by desku 9y ago
I actually found this to be one of the best explanations on this topic I've read. Fully recommend the author's book too. Also recommend this as it follows it/is an alternative: https://arxiv.org/abs/1703.
20.
by desku 9y ago
gameaibook.org
21.
by desku 9y ago
It's bloat due to 'introns' (useless statements that don't effect the output, like x = x * 1). And yes, just adding a fitness function to shorten program length isn't optimal. I've found it easier to evolve suc
22.
by desku 9y ago
I don't think it'll ever be finished.
23.
by desku 9y ago
I get this too. Haven't been able to get references to work for a single paper yet.
24.
by desku 9y ago
I'd argue that's exactly what computer science is.
25.
by desku 9y ago
If you're into AI/ML: Import AI and The Wild Week in AI.
26.
by desku 10y ago
MIT had a short, week-long 'Intro to Deep Learning' course that had some labs in Tensorflow. http://introtodeeplearning.com/
27.
by desku 10y ago
I believe there's bindings for: C++, Java, Rust, Haskell and Go.