Comments (8)
We filter the data used to under a certain length. If you look at the librispeech.py script, for example, you can see that there is a flag there for doing filtering when the manifest is created.
from deepspeech.pytorch.
Baidus' internal code is much more memory optimised than Torch (atleast for RNNs), allowing them to run much larger batch sizes per GPU. This has always been a crux of using RNNs! I'm honestly not sure where to exactly look to handle this, but since you have pascal cards this PR is something to look at. Persistent RNNs speed up RNNs a lot when using smaller batch sizes because there is less transfer of memory!
from deepspeech.pytorch.
I've seen many references to their work on being efficient with memory. I think I'm just surprised that it's making a ~4x difference in batch size. Impressive (and good for them) I suppose. Will close if this behavior is roughly consistent with others are seeing.
from deepspeech.pytorch.
Yes, I have observed the same behaviour
from deepspeech.pytorch.
I'm going to tie #55 to this, with successful FP16 we should be allowed to double batch sizes (with slight performance loss due to cublas doing some funky stuff with pseudo FP16 that I was notified about)
from deepspeech.pytorch.
As explained here, sadly FP16 won't be viable for some time (atleast for training).
from deepspeech.pytorch.
@ryanleary @SeanNaren : Excuse me! I've been seeing you guys training on full set of Librispeech. I wonder if you do something to constrain the length of the audio files?
I'm also doing experiments on Librispeech (100h), and I just cannot train with the batch sizes as big as yours.
I know that in the dataset there are quite a lot files longer than 30 seconds. I don't know if this is the main the reason that keeps me from training with bigger batch size? Now I'm training with the batch size of 8, and it's extremely slow.
Did you guys do something with the dataset?
Thanks in advance!
from deepspeech.pytorch.
@ryanleary : Ok I see. Thank you so much!
from deepspeech.pytorch.
Related Issues (20)
- Question about data shuffling HOT 3
- cant train! HOT 1
- could you please provide a docker image with an established environment? HOT 1
- loss (Loss 0.00000) for all along the training HOT 4
- weights_summary and weights_path on lightning_config not working. HOT 1
- omegaconf.errors.ValidationError: Unexpected object type: tuple HOT 1
- About omegaconf.errors.ValidationError: Unexpected object type: tuple at train_config.py HOT 2
- error in training HOT 1
- Does batch size change while traininig with elastic agent? HOT 1
- deepspeech2 need how much Flops? HOT 1
- How to support microphone vad by deepspeech.pytorch? HOT 1
- partition_activations produces no activation memory improvement with zero3 HOT 1
- TypeError("__init__() got an unexpected keyword argument 'num_processes'") HOT 3
- What should we do if we have segments for dataset? HOT 1
- it did not generates the models/deepspeech.pth file HOT 2
- Training
- Got too high WER and CER on AN4 HOT 1
- How can I train or test without spectrogram? HOT 1
- Support for newest version of pytorch-lightning right now (2.1)
- SequenceWise Operation
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 deepspeech.pytorch.