Skip to content

Releases: will-rice/denoisers

V0.1.8

11 May 16:41
bb3094e
Compare
Choose a tag to compare

This release adds cropping to the skip connections in the waveform models to enable the use of different max lengths than multples of 16384.

V0.1.7

13 Mar 10:54
e5d6688
Compare
Choose a tag to compare

This release changes model outputs from .logits to .audio

V0.1.6

03 Nov 13:53
bc020bb
Compare
Choose a tag to compare

Remove dependencies from pypi build

V0.1.5

03 Nov 13:50
0c75246
Compare
Choose a tag to compare

Patch release. torchaudio not found via pip during pypi publish.

V0.1.4

03 Nov 13:37
d703781
Compare
Choose a tag to compare

This release introduces a new model UNet1DModel. There is currently a pretrained checkpoint on the huggingface hub. It is still training, but beats the waveunet checkpoints across the board.

V0.1.3

11 Oct 20:16
6f5e0f8
Compare
Choose a tag to compare
Update pyproject.toml

V0.1.2

11 Oct 19:39
9882093
Compare
Choose a tag to compare
Update pyproject.toml

V0.1.0

11 Oct 16:13
94973fc
Compare
Choose a tag to compare
Update pyproject.toml

V0.0.3

06 Feb 22:00
0dea89d
Compare
Choose a tag to compare
V0.0.3 Pre-release
Pre-release
Update pyproject.toml

V0.0.2.1

06 Feb 21:53
7fc6224
Compare
Choose a tag to compare
V0.0.2.1 Pre-release
Pre-release
Merge pull request #2 from will-rice/wr-build

Add pyproject.toml and legacy setup.py