-
Hi, I'm trying to train mini_librispeech with a custom dataset.
The dataset consists of [LibriTTS](http://www.openslr.org/60/) speakers, [ARU](http://datacat.liverpool.ac.uk/681/) speakers, and […
-
Looks like a bug to me. Here's a simple script:
```
if __name__ == '__main__':
rows = [{'foo': 1, 'bar': 2}, {'foo': 3, 'bar': 4}]
import pandas as pd
import pyarrow as pa
import py…
pwais updated
4 years ago
-
I am afraid my conjecture in [#448](https://github.com/OCR-D/core/pull/448#pullrequestreview-393997881) that we could become slower than before has now proven correct. I have a workspace here with mor…
-
For example:
The first match is not what you want nor is it on the way to what you wan…
-
Hi, thanks for making your code available! I'm trying to produce similar results to what you report on the CoNLL03 German NER task using the 0.5 release. When I run the following script, though, I onl…
-
tsalo updated
4 years ago
-
On the perkeep mailing list thread where this project got started, @bobg noted some observed statistical weaknesses in the `rrs1` hash in the spec and argued for defining and recommending a stronger h…
-
am using mallet for topic modeling of data. All things are good but the topics getting generated are inconsistent. Its generating weightage every time different on running the model on the same data I…
-
@jonathanmetzman
I dont want to spam the announcement thread so I think communicating in a dedicated issue is better?
you wanted to run the next batch today. Will that happen? And if so - when do…
-
Objective
===
The current version of the EEBO site could do with some design improvements -- there is a disconnect between the pages and their content, many of the visualizations are confusing and s…