Skip to content

Latest commit

 

History

History
13 lines (11 loc) · 912 Bytes

README.md

File metadata and controls

13 lines (11 loc) · 912 Bytes

Templates Overview

Each of notebooks have the objective of train and save a model based on the HAM10000 dataset. There are slighty differences noted below:

  • templateDownload - Download the dataset using a kaggle.json authentication file.
  • templateDrive - When the dataset is saved on Google Drive, it could delay to load all the images into the runtime memory.
  • templateFormDownload - Download the dataset using a kaggle.json authentication file and also the configuration block in form format.

Experiments

Notebooks used to test and watch the behavior of techniques, architectures and hardware.

Trainings

Notebooks used to train and extract metrics to compare techniques and configurations.