TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: A high-performance TensorFlow library for quantitative finance

153 pointsby partingshotsover 4 years ago

11 comments

dvtover 4 years ago
Is there any actual value in doing quantitative finance via TF as opposed to just Pandas? I get that calculating a zillion option prices on a GPU is faster, but does anyone actually do that? Not to mention that dataframes (both conceptually, and in practice) seem much more mature and easier to grok than tensors.
评论 #25830640 未加载
评论 #25830484 未加载
评论 #25830903 未加载
评论 #25830145 未加载
评论 #25837902 未加载
评论 #25831534 未加载
akhilcacharyaover 4 years ago
Appears to be the result of an Area120 incubated project: <a href="https:&#x2F;&#x2F;avera.area120.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;avera.area120.com&#x2F;</a>
reallymentalover 4 years ago
I understand that this is google&#x27;s library, and they&#x27;re promoting TensorFlow&#x27;s usage, and subsequently the usage of TPU&#x27;s.<p>But this can just as easily be done in Pytorch right? Yes we can&#x27;t have GPU&#x27;s there but having your model train 50% slower (exaggeration) is better than having to spend 150% of the time taken in Pytorch to debug anything in TensorFlow.<p>I may be beating a dead horse here, but why doesn&#x27;t google just accept that TensorFlow needs to be redesigned for more ease of use?<p>Can anyone point out the benefits of TensorFlow over Pytorch (besides TPU&#x27;s) ? It&#x27;s been a few years since I&#x27;ve used TF and I may have missed something.
评论 #25830536 未加载
评论 #25831415 未加载
proverbialbunnyover 4 years ago
Is this using floating point numbers under the hood? There is a lot you can do with floating point, but for some of what I&#x27;ve worked on you really do want fixed precision types.
评论 #25835270 未加载
评论 #25831969 未加载
sandGorgonover 4 years ago
I wonder why they are not using the in-built numpy api - <a href="https:&#x2F;&#x2F;www.tensorflow.org&#x2F;guide&#x2F;tf_numpy" rel="nofollow">https:&#x2F;&#x2F;www.tensorflow.org&#x2F;guide&#x2F;tf_numpy</a><p>This will be accelerated by Tensorflow - which means acceleration on Macbook M1 as well
secondcomingover 4 years ago
We tried to use Tensorflow&#x27;s native C++ library to run some predictive model in real time on CPU-only. The performance sucked, even though the model was not particularly complex.<p>My advice would be to only use Tensorflow with the python wrappers. It&#x27;s just too complex.
vmceptionover 4 years ago
Hi, I have a need to know the price of Liquidity Pool shares without relying on many oracles and multiple ABI calls<p>Liquidity pool shares are asset backed by at least two other assets<p>Can Tensorflow help me with keeping pricing?
pvitzover 4 years ago
Right now, it looks like a fun project of students who got their hands on Andersen&#x2F;Piterbarg, Brigo&#x2F;Mercurio and the QMC part of Glasserman. I wonder what their real goal is with this.
评论 #25840132 未加载
评论 #25834138 未加载
person_of_colorover 4 years ago
Why would Google employees be doing quant finance on company time?
评论 #25831962 未加载
neologover 4 years ago
Was this developed for Google internal finance projects?
评论 #25834150 未加载
leoplctover 4 years ago
Are still someone using Black-Scholes for option pricing nowadays?
评论 #25831701 未加载