Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bratao
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
26 ms
·
331.
▲
Conda now supports PyPy builds
(conda-forge.org)
64 points
by
bratao
7y ago
|
30 comments
332.
▲
Vespa.ai February Product Updates
(blog.vespa.ai)
2 points
by
bratao
7y ago
|
0 comments
333.
▲
by
bratao
7y ago
For Vue.js fans, try https://github.com/nuxt/nuxt.js We are very happy with it!
334.
▲
by
bratao
7y ago
Every time people say that if they had how to pay for the site they would use it. Now is the time to show the truth
335.
▲
by
bratao
7y ago
The MyRocks is a major trade secret in my company. We can very quickly store TBs of highly compressed text in Hundreds of GB. More than 10x compression
336.
▲
A Hidden Gem in MySQL: MyRocks
(percona.com)
9 points
by
bratao
7y ago
|
2 comments
337.
▲
Cursedfs – Make a disk image formatted with both ext2 and FAT at once
(github.com)
23 points
by
bratao
7y ago
|
2 comments
338.
▲
by
bratao
7y ago
I migrated from ES and for me, I do not agree about the feeling that it doesn't feel modern. The Middleware logic container and Live reconfiguration it is mind blowing. About those two things: - Nested (For my use cases, this is a prob
339.
▲
by
bratao
7y ago
Shameless plug from someone who want this project to flourish. Check https://vespa.ai as an alternative to Elasticsearch. Migrating from a ES to it, I got a faster search, never had to face a unhealthy node and native tensor sup
340.
▲
by
bratao
7y ago
Here is the full changelog https://docs.python.org/3.8/whatsnew/changelog.html#python-3... It is a signal for many people to move to 3.8
341.
▲
Python 3.8.1 Released
(python.org)
6 points
by
bratao
7y ago
|
3 comments
342.
▲
by
bratao
7y ago
I was impressed by the quality of PyPy errors!
343.
▲
Getting started with machine-learned search ranking using Vespa.ai
(medium.com)
1 points
by
bratao
7y ago
|
0 comments
344.
▲
Skip YouTube video sponsors(browser extension)
(github.com)
2 points
by
bratao
7y ago
|
0 comments
345.
▲
Force Matlab to use a fast codepath on AMD CPUs – up to 250% performance gains
(reddit.com)
1 points
by
bratao
7y ago
|
0 comments
346.
▲
Elastic.co vs. Vespa.ai dense tensor benchmark
(github.com)
3 points
by
bratao
7y ago
|
0 comments
347.
▲
by
bratao
7y ago
Thank you, I will read and try it. Looks very interesting!
348.
▲
by
bratao
7y ago
I did not understood. BERT it is not a similarity measure. For our use case we did use a simple cosine similarity to find the similar documents. But we have to represent those documents in a vector space. From our test representing the docu
349.
▲
by
bratao
7y ago
BERT is truly amazing. Almost all inovation in NLP uses BERT and transformers somehow. ALBERT will be the next HUGE thing for the next months, as it show results better than BERT with a small fraction of parameters. We did a "Semantic
350.
▲
by
bratao
7y ago
This confirm my real-world usage. We had to do a "Semantic Similarity search" for some documents, where we represent a document as a vector using BERT, and had to look for documents close to a reference document. First we did on t
351.
▲
Google is improving searches by using BERT and understanding language context
(theverge.com)
12 points
by
bratao
7y ago
|
4 comments
352.
▲
by
bratao
7y ago
There is https://github.com/python/mypy/tree/master/mypyc that I think is a great idea and approach
353.
▲
by
bratao
7y ago
According to the Table 6, the performance of not instrumented cube is 20% for applying half of a fair scramble and 0% for a full scramble. Right?
354.
▲
by
bratao
7y ago
I felt that the communication in PR was not clear. Even about this, the cube was heavily instrumented.
355.
▲
OpenAI's blog post about "solving the Rubik's cube" and what they actually did
(twitter.com)
111 points
by
bratao
7y ago
|
50 comments
356.
▲
PyPy v7.2 Released
(morepypy.blogspot.com)
5 points
by
bratao
7y ago
|
3 comments
357.
▲
How to Make Friends as an Adult: 5 Secrets Backed by Research
(bakadesuyo.com)
4 points
by
bratao
7y ago
|
1 comments
358.
▲
The MySQL 8.0.18 Is Released
(mysqlserverteam.com)
2 points
by
bratao
7y ago
|
0 comments
359.
▲
by
bratao
7y ago
I try PyPy every 3 months. It has improved greatly and for some periods I migrated to it. But for this particular project, most of time is spent inside lxml, pandas, scikit-learn and other extensions. CPython is actually faster than PyPy fo
360.
▲
by
bratao
7y ago
I do not know exactly what caused this, but my internal project test-suite got a much appreciated 5% speedup! Thank you Python team!
More ›