Skip to content
This repository has been archived by the owner on Jul 18, 2024. It is now read-only.

Releases: IBM/pytorchpipe

0.2.1 release

06 Jul 01:13
@tkornuta-ibm tkornuta-ibm
v0.2.1
713e1ca
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
0.2.1 release Latest
Latest

What's new:

  • Refined the readme file
  • Included quick start
  • Included tutorial presentation
Assets 2
Loading

0.2.0 release

03 Jul 00:48
@tkornuta-ibm tkornuta-ibm
v0.2.0
f2d5ef0
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
0.2.0 release

What's new:

  • Refactor: Problems -> Tasks
  • Refactor: DataDict -> DataStreams
  • Code cleanup and reorganization of all components
  • 2 new VQA Tasks: CLEVR and GQA

Most important changes are listed in the associated project:
https://github.com/IBM/pytorchpipe/projects/7

Assets 2
Loading

0.1.0 release

03 Jul 00:47
@tkornuta-ibm tkornuta-ibm
v0.1.0
2462409
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
0.1.0 release

First official release of PTP!

Features:

  • 3 universal, configurable workers: scripts for training/testing of all kinds of pipelines
  • 9 problems from diverse domains (visual question answering, image classification, language classification, language modelling, machine translation)
  • 31 configurable components, including 13 trainable models
  • around 80 pipelines, varying from simple ones with few components for MNIST classification to ones with 50+(!) components for VQA
Assets 2
Loading