Skip to content

Conversation

@Kamil-Kisielewicz
Copy link
Collaborator

No description provided.

Kamil-Kisielewicz and others added 10 commits August 1, 2019 15:17
basic playground
minimal difference from v1, just more experimentation and extra dropout and relu layers here and there
minimal changes, some experimentation with adding relu and dropouts here and there
heavy experimentation, added BC model, added GPU capabilities, full train and test on the v1.1 model with a 99.5 test accuracy and 99.79 validation accuracy.
switch legend to 'best' and add abstraction of history object
densenet architecture used for poster, evaluation, and confusion matrix without heuristics.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants