ytl0623

Add function in monai.transforms.utils.py

6

Fixes #6704 ### Description Combined `map_classes_to_indices` and `generate_label_classes_crop_centers` to a single function `map_and_generate_sampling_centers` in monai.transforms.utils.py. ### Types of changes - [x] Non-breaking change (fix or new feature that would not...

wholeBody_ct_segmentation model for DICOM input file

2

Hi, I was used wholeBody_ct_segmentation model to test custom clinical image. But the format file is NIFITI (.nii), how can I inference the model with DICOM file format. Is there...

What's the correct segmentation label match the organ name in Wholebody ct segmentation pre-trained model?

2

Hi, I used to execute inference with Wholebody ct segmentation from monai model zoo. But after I finished the process the output image is NOT match the organ list (like...

What's the dataset for training?

1

Hi, I was interest in your work using deep learning model to detect fire. May I ask what is the dataset for training model? Is there any public dataset for...

Print what package should be installed when suitable writer is missing

3

Fixes #7980 ### Description To implement these changes: 1. Add the `from typing import Dict, List` import. 2. Add the `WRITER_PACKAGE_MAP` dictionary after the existing imports. 3. Replace the existing...

timestep scheduling with np.linspace

2

Fixes #8600 ### Description The `np.linspace` approach generates a descending array that starts exactly at 999 and ends exactly at 0 (after rounding), ensuring the scheduler samples the entire intended...

add sigmoid/softmax support and multi-class extension for AsymmetricU…

1

…nifiedFocalLoss Fixes #8603 . ### Description - Added `use_softmax` argument: Allows the loss function to accept raw logits and handle activation (sigmoid or softmax) internally for better numerical stability. -...