Skip to content

Latest commit

 

History

History
23 lines (15 loc) · 774 Bytes

File metadata and controls

23 lines (15 loc) · 774 Bytes

Kaggle Second Annual Data Science Bowl

Preprocessing

Read the preprocessing readme on how to preprocess the data.

Evaluation of Segmentation Models

Although Kaggle does not provide the ground truth segmentation masks, we can still evaluate the segmentation models and compare the segmentation-derived ejection fraction (EF) with the ground truth EF. The predicted segmentation can also be used to extract the frames of ED and ES phases used for EF calculation.

kaggle_seg_eval --data_dir ~/.cache/cinema_datasets/kaggle/processed --ckpt_path

Evaluation

The following command can be used to evaluate the model (please adjust the model.ckpt_path accordingly):

kaggle_clf_eval --data_dir --ckpt_path