Comments (3)
Hi,
That's an interesting task. It's 536h training data? Or how much? I don't quite understand how to read the table.
I would recommend a new data preparation pipeline. It's kind of similar to 2018-asr-attention (to the latest version, in Ogg Zip format), but cleaned up, and more generic (it uses the OggZipDataset
, and not LibriSpeechCorpus
).
It's 2020-librispeech-data-prepare.
How did you do the data preparation? Which dataset in RETURNN did you use? Different dataset might use different data-keys. E.g. in the SprintDataset
, it's called "bpe"
, and in many other datasets, it's called "classes"
. Depending what the dataset provides, you need to configure the corresponding data-keys in your config (via num_outputs
, or via extern_data
), and then also use the right target.
More specifically, in your current config:
- You probably have
target = "bpe"
somewhere, but your extern-data is configured such that there is"classes"
, not"bpe"
. So just change it totarget = "classes"
. - Or maybe, you configured your extern-data (via
num_outputs
orextern_data
) wrong. You maybe havenum_outputs = {..., "classes": ...}
. Just change"classes"
to"bpe"
.
Which variant is correct depends on your dataset class (some provide "bpe"
, most provide "classes"
).
from returnn-experiments.
Btw, what is your attention baseline config? The exp3.ctc is old (from 2018). I would suggest to use some of the configs from 2019-asr-e2e-trafo-vs-lstm.
from returnn-experiments.
No answer. Closed for now.
from returnn-experiments.
Related Issues (20)
- local attention with unidirectional lstm not converging HOT 5
- Implement a unidirectional variant of local attention HOT 10
- Loading a saved Returnn model from its .meta file HOT 16
- query regarding LM data preprocessing HOT 2
- Reusing parameters inside rec layer HOT 5
- Training Configuration for TEDLIUMv2 HOT 3
- specAugment policy and schedules HOT 3
- Question about 2020-rnn-transducer HOT 16
- Question about 2018-asr-librispeech dev = get_dataset("dev", subset=3000) HOT 2
- loss nan and cost nan while running my own corpus using librispeech sets HOT 10
- Hierarchical layer name not captured correctly
- Problem with retrieving source layer from a hierarchical definition
- Multi Stage Training
- Questions on librispeech transformer lm HOT 10
- Transducer error in GetFilteredScoreOp HOT 4
- Big files in repo HOT 5
- Git commit/push rule to not allow big files HOT 3
- Could you please provide a script that could run lsh-attention for translation? HOT 4
- Assert Error when running 2022-lsh-attention HOT 7
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 returnn-experiments.