Suraj Pai

Add Automated Google Image Downloader #35

1

Add third party feature #35 for Downloading Bulk images from Google Image search

CPU-Memory keeps accumulating during `trainer.predict`

3

### Bug description This is very similar to closed issue #15656 I am working on predicting using PL Trainer on 3D images and these are huge, my process keeps getting...

EMP-SSL

5

Hi Team, I recently came across this paper (Link: [EMP-SSL Paper](https://arxiv.org/abs/2304.03977)). The authors have provided their implementation on Github [(EMP-SSL Repository)](https://github.com/tsb0601/EMP-SSL). TLDR of the method: The key towards efficient self-supervised...

[MONAI Bundle] Very long errors during config instantiation - Clean up raised errors

2

**Is your feature request related to a problem? Please describe.** When there is an error in the monai bundle config during instantiation, and the config is large, the `RuntimeError` ends...

`reverse_indexing` crashes for `ITKReader` when using `.nrrd`

2

**Describe the bug** Setting `reverse_indexing=True` works well for NIFTI files but crashes for NRRD files when setting `reader="ITKReader"` in `LoadImage` **To Reproduce** _Case 1: Load NIFTI without setting `reverse_indexing` -...

**Is your feature request related to a problem? Please describe.** Currently, classes to generate saliency maps like `VanillaGrad`, `SmoothGrad`, etc. support only running on a single input sample. Giving a...

Thanks for the beautiful project! We use Telegram and Pytorch extensively in our team and it would be a shame not to integrate this with Pytorch. I'm thinking of transforming...

Provide ability to override `map_items=True` set by default for `Dataset`

2

**Is your feature request related to a problem? Please describe.** https://github.com/Project-MONAI/MONAI/blob/bff4b1559a147f570fc881717eadc7b3e8a98b96/monai/data/dataset.py#L98 In the `Dataset` class, for any `transform` applied, `map_items=True` is the default behavior in the `apply_transform` helper function. I'm...

Make dictionary transforms support different random parameters across keys

2

**Is your feature request related to a problem? Please describe.** For current dictionary transforms, the random parameters are shared across keys. E.g: see https://github.com/Project-MONAI/MONAI/blob/ec63e068fec85dff5f043d73f2c029b359332707/monai/transforms/intensity/dictionary.py#L1401 https://github.com/Project-MONAI/MONAI/blob/ec63e068fec85dff5f043d73f2c029b359332707/monai/transforms/croppad/dictionary.py#L405 **Describe the solution you'd like**...