Comments (4)
For the filtering stage, on CIRR with one A100 80G, the model takes 4 minutes to train an epoch (incl. per-epoch validation), in total it takes about 2-3 hours till its peak performance. Likewise for Fashion-IQ.
from candidate-reranking-cir.
Hi,
To answer your question, on one A100 80G, when training with the CIRR dataset, the candidate re-ranking stage takes around 40 minutes per epoch, similar for training on Fashion-IQ.
In total we trained for 80 epochs on CIRR (50 on Fashion-IQ) until it peaks so the entire training phase took a bit over 2 days for CIRR (around 1 day for Fashion-IQ).
Of course, in development we usually don't have to train till peak to gauge its performance so if you are keen on a faster turn-around for development just examine the curve upfront and decide if it's worth your time to keep going.
Here is how it goes for CIRR (we extend the epochs to 100 just to verify its peak at ~80):
from candidate-reranking-cir.
Oh I see. But this is the reranker training time?
I want to know the time of dual encoder training time(first stage), thanks!!
from candidate-reranking-cir.
Thanks!!
from candidate-reranking-cir.
Related Issues (1)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from candidate-reranking-cir.