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

C4 answer-word classifiers, C123 - configs for evaluation #45

Merged
merged 15 commits into from
May 5, 2019
Merged

C4 answer-word classifiers, C123 - configs for evaluation #45

merged 15 commits into from
May 5, 2019

Conversation

@tkornuta-ibm
Copy link
Contributor

@tkornuta-ibm tkornuta-ibm commented May 4, 2019 •
edited
Loading

IMPORTANT:

  • config.py: set required versions of PTP: pytorch (1.0.1) torchvision (0.2.1), tochtext (0.3.1)

New features:

  • new component: ReduceTensor
  • progress bar added to load embeddings

New configs:

  • for experiments with answer-word classifiers
  • for evaluation of C123 pipelines

Fixes:

  • fix in loading embeddings when label has many words
  • fix in MCB: projection matrices are saved/loaded with the model (!), added option so they can also be trainable (they won't be doing "sketching" anymore though)
  • fix in VQA_Attention: exporting calculated output_size to globals

@tkornuta-ibm tkornuta-ibm changed the title C4 answer-word classifiers C4 answer-word classifiers, C123 - configs for evaluation May 5, 2019
@tkornuta-ibm tkornuta-ibm merged commit 98b1f98 into develop May 5, 2019
@tkornuta-ibm tkornuta-ibm deleted the c4_answer branch May 29, 2019 18:36
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@tkornuta-ibm