Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Merge dev into geometric #8026

Closed
wants to merge 148 commits into from
Closed

Merge dev into geometric #8026

wants to merge 148 commits into from

Conversation

KumoLiu
Copy link
Contributor

@KumoLiu KumoLiu commented Aug 19, 2024

Fixes # .

Description

A few sentences describing the changes proposed in this pull request.

Types of changes

  • Non-breaking change (fix or new feature that would not break existing functionality).
  • Breaking change (fix or new feature that would cause existing functionality to change).
  • New tests added to cover the changes.
  • Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
  • Quick tests passed locally by running ./runtests.sh --quick --unittests --disttests.
  • In-line docstrings updated.
  • Documentation updated, tested make html command in the docs/ folder.

KumoLiu and others added 30 commits February 24, 2024 00:49
@KumoLiu
Use int16 instead of int8 in LabelStats (#7489)
f6f9e81 
Use uint8 instead of int8 in `LabelStats`.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
@monai-bot
auto updates (#7495)
20512d3 
Signed-off-by: monai-bot <[email protected]>

Signed-off-by: monai-bot <[email protected]>
@bakert1
Add sample_std parameter to RandGaussianNoise. (#7492)
7cfa2c9 
Fixes issue #7425

### Description

Add a `sample_std` parameter to `RandGaussianNoise` and
`RandGaussianNoised`. When True, the Gaussian's standard deviation is
sampled uniformly from 0 to std (i.e., what is currently done). When
False, the noise's standard deviation is non-random and set to std. The
default for sample_std would be True for backwards compatibility.

Changes were based on RandRicianNoise which already has a `sample_std`
parameter and is similar to RandGaussianNoise in concept and
implementation.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Timothy Baker <[email protected]>
@MathijsdeBoer @KumoLiu
Add __repr__ and __str__ to Metrics baseclass (#7487)
9830525 
### Description

When training a model using MONAI metrics for experiment tracking, I
tend to log which metrics I am using. Unfortunately, just sending the
metrics objects to Tensorboard will result in a list like
[CustomMetric1, CustomMetric2, <monai.metrics.Metric ...>, etc.]

Adding `__repr__` and `__str__` methods to the base class will solve
this small annoyance. The current implementation will only return the
class name, but if a certain metric would wish to report more data for
its `__repr__` string, this can be easily overridden in any subclass.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Signed-off-by: Mathijs de Boer <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: Mathijs de Boer <[email protected]>
@dependabot
Bump al-cheb/configure-pagefile-action from 1.3 to 1.4 (#7510)
02c7f53 
Bumps
[al-cheb/configure-pagefile-action](https://github.com/al-cheb/configure-pagefile-action)
from 1.3 to 1.4.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/al-cheb/configure-pagefile-action/releases">al-cheb/configure-pagefile-action's
releases</a>.</em></p>
<blockquote>
<h2>v1.4: Update task node version to 20</h2>
<h1>configure-pagefile-action</h1>
<p>This action is intended to configure Pagefile size and location for
Windows images in GitHub Actions.</p>
<h1>Available parameters</h1>
<table>
<thead>
<tr>
<th>Argument</th>
<th>Description</th>
<th>Format</th>
<th>Default value</th>
</tr>
</thead>
<tbody>
<tr>
<td><code>minimum-size</code></td>
<td>Set minimum size of Pagefile</td>
<td><code>2048MB</code>, <code>4GB</code>, <code>8GB</code> and etc</td>
<td><code>8GB</code></td>
</tr>
<tr>
<td><code>maximum-size</code></td>
<td>Set maximum size of Pagefile</td>
<td>The same like <code>minimum-size</code></td>
<td><code>minimum-size</code></td>
</tr>
<tr>
<td><code>disk-root</code></td>
<td>Set disk root where Pagefile will be located</td>
<td><code>C:</code> or <code>D:</code></td>
<td><code>D:</code></td>
</tr>
</tbody>
</table>
<h1>Usage</h1>
<pre><code>name: CI
on: [push]
jobs:
  build:
    runs-on: windows-latest
    steps:
    - name: configure Pagefile
      uses: al-cheb/[email protected]
      with:
        minimum-size: 8
<pre><code>- name: configure Pagefile
  uses: al-cheb/[email protected]
  with:
    minimum-size: 8
    maximum-size: 16
    disk-root: &amp;quot;D:&amp;quot;
</code></pre>
<p></code></pre></p>
<h1>License</h1>
<p>The scripts and documentation in this project are released under the
<a
href="https://github.com/al-cheb/configure-pagefile-action/blob/HEAD/LICENSE">MIT
License</a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/al-cheb/configure-pagefile-action/commit/a3b6ebd6b634da88790d9c58d4b37a7f4a7b8708"><code>a3b6ebd</code></a>
Merge pull request <a
href="https://redirect.github.com/al-cheb/configure-pagefile-action/issues/20">#20</a>
from mikehardy/mikehardy-patch-1</li>
<li><a
href="https://github.com/al-cheb/configure-pagefile-action/commit/850626ff7d6f5665848a60cdc3b0ace19fa0b094"><code>850626f</code></a>
build(deps): bump javascript dependencies / forward-port as needed</li>
<li><a
href="https://github.com/al-cheb/configure-pagefile-action/commit/e7aac1b891d29f440ea916231d51d085206f9106"><code>e7aac1b</code></a>
fix: use node 20</li>
<li><a
href="https://github.com/al-cheb/configure-pagefile-action/commit/d940d2434c38e1a228580edc8f9421aafca9800d"><code>d940d24</code></a>
build(deps): use v4 of setup-node action, use node 20</li>
<li><a
href="https://github.com/al-cheb/configure-pagefile-action/commit/dfdc038768fdb21f91ed75a5fa40890fb92f5919"><code>dfdc038</code></a>
build(deps): bump actions/checkout from 3 to 4</li>
<li>See full diff in <a
href="https://github.com/al-cheb/configure-pagefile-action/compare/v1.3...v1.4">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=al-cheb/configure-pagefile-action&package-manager=github_actions&previous-version=1.3&new-version=1.4)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
@KumoLiu @ericspod
Add arm support (#7500)
e9e2738 
Fixes # .

### Description

Add arm support

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
@KumoLiu
Fix error in "test_bundle_trt_export" (#7524)
95f69de 
Fixes #7523


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <[email protected]>
@SomeUserName1
Fix typo in the PerceptualNetworkType Enum (#7548)
6b7568d 
Fixes #7547 

### Description
Previously it was 'medical_resnet50_23datasets' for both identifier and
string, which doesn't correspond to the name in the hubconf.py of
Warvito's repo. Now it is the correct version (according to Warvitos
repo) 'medicalnet_resnet50_23datasets'.

The docs state it correctly already. 

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ x] New tests added to cover the changes.
- [ x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Fabian Klopfer <[email protected]>
@KumoLiu
Update to use log_sigmoid in FocalLoss (#7534)
ec63e06 
Fixes #7533

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
@KumoLiu @yiheng-wang-nv
Update integration_segmentation_3d result for PyTorch2403 (#7551)
35c93fd 
Fixes #7550


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: Yiheng Wang <[email protected]>
@Lucas-rbnt @ericspod
@KumoLiu @pre-commit-ci
Add Barlow Twins loss for representation learning (#7530)
c649934 
### Description

Addition of the BarlowTwinsLoss class. This cost function is introduced
in the http://proceedings.mlr.press/v139/zbontar21a/zbontar21a.pdf paper
with the aim of disentangling the representations learned on two views
of the same sample, making it a powerful tool for multimodal and
unsupervised learning.
This cost function is similar to the InfoNCE Loss function already
implemented in MONAI
(https://docs.monai.io/en/latest/_modules/monai/losses/contrastive.html#ContrastiveLoss).
However, it differs in several respects: there is no l2-normalisation,
but rather a z-normalisation. In addition, rather than working between
pairs of embeddings, Barlow Twins seeks to decorrelate the components of
the representations.

```math
\mathcal{L}_{BT} := \sum_i (1 - \mathcal{C}_{ii})^2 + \lambda \sum_i \sum_{i\neq j} \mathcal{C}_{ij}^2
```
with $\lambda$ a positive hyperparameters and $\mathcal{C}$ the
cross-correlation matrix

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Lucas Robinet <[email protected]>
Signed-off-by: Lucas Robinet <[email protected]>
Co-authored-by: Lucas Robinet <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@cxlcl @pre-commit-ci
@KumoLiu @ericspod
Stein's Unbiased Risk Estimator (SURE) loss and Conjugate Gradient (#…
c3a7383 
…7308)

### Description

Based on the discussion topic
[here](#7161 (comment)),
we implemented the Conjugate-Gradient algorithm for linear operator
inversion, and Stein's Unbiased Risk Estimator (SURE) [1] loss for
ground-truth-date free diffusion process guidance that is proposed in
[2] and illustrated in the algorithm below:

<img width="650" alt="Screenshot 2023-12-10 at 10 19 25 PM"
src="https://github.com/Project-MONAI/MONAI/assets/8581162/97069466-cbaf-44e0-b7a7-ae9deb8fd7f2">

The Conjugate-Gradient (CG) algorithm is used to solve for the inversion
of the linear operator in Line-4 in the algorithm above, where the
linear operator is too large to store explicitly as a matrix (such as
FFT/IFFT of an image) and invert directly. Instead, we can solve for the
linear inversion iteratively as in CG.

The SURE loss is applied for Line-6 above. This is a differentiable loss
function that can be used to train/giude an operator (e.g. neural
network), where the pseudo ground truth is available but the reference
ground truth is not. For example, in the MRI reconstruction, the pseudo
ground truth is the zero-filled reconstruction and the reference ground
truth is the fully sampled reconstruction. The reference ground truth is
not available due to the lack of fully sampled.

**Reference**
[1] Stein, C.M.: Estimation of the mean of a multivariate normal
distribution. Annals of Statistics 1981 [[paper
link](https://projecteuclid.org/journals/annals-of-statistics/volume-9/issue-6/Estimation-of-the-Mean-of-a-Multivariate-Normal-Distribution/10.1214/aos/1176345632.full)]
[2] B. Ozturkler et al. SMRD: SURE-based Robust MRI Reconstruction with
Diffusion Models. MICCAI 2023
[[paper link](https://arxiv.org/pdf/2310.01799.pdf)]

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: chaoliu <[email protected]>
Signed-off-by: cxlcl <[email protected]>
Signed-off-by: chaoliu <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
@monai-bot
auto updates (#7577)
c86e790 
Signed-off-by: monai-bot <[email protected]>

Signed-off-by: monai-bot <[email protected]>
@surajpaib @ericspod
Remove nested error propagation on ConfigComponent instantiate (#7569)
97678fa 
Fixes #7451

### Description
Reduces the length of error messages and error messages being propagated
twice. This helps debug better when long `ConfigComponent`s are being
instantiated. Refer to issue #7451 for more details

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Suraj Pai <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
@juampatronics @pre-commit-ci
@monai-bot @elitap @Shadow-Devil @YanxuanLiu @ytl0623 @dzenanz @ericspod @KumoLiu @ishandutta0098 @dependabot @kvttt @heyufan1995 @binliunls @atbenmurray @vlaminckaxel @mingxin-zheng @ibro45 @drbeh @bakert1 @MathijsdeBoer @SomeUserName1 @yiheng-wang-nv @Lucas-rbnt @cxlcl
2872 implementation of mixup, cutmix and cutout (#7198)
e5bebfc 
Fixes #2872 

### Description

Implementation of mixup, cutmix and cutout as described in the original
papers.
Current implementation support both, the dictionary-based batches and
tuples of tensors.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Juan Pablo de la Cruz Gutiérrez <[email protected]>
Signed-off-by: monai-bot <[email protected]>
Signed-off-by: elitap <[email protected]>
Signed-off-by: Felix Schnabel <[email protected]>
Signed-off-by: YanxuanLiu <[email protected]>
Signed-off-by: ytl0623 <[email protected]>
Signed-off-by: Dženan Zukić <[email protected]>
Signed-off-by: KumoLiu <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Signed-off-by: Ishan Dutta <[email protected]>
Signed-off-by: dependabot[bot] <[email protected]>
Signed-off-by: kaibo <[email protected]>
Signed-off-by: heyufan1995 <[email protected]>
Signed-off-by: binliu <[email protected]>
Signed-off-by: axel.vlaminck <[email protected]>
Signed-off-by: Ibrahim Hadzic <[email protected]>
Signed-off-by: Behrooz <[email protected]>
Signed-off-by: Timothy Baker <[email protected]>
Signed-off-by: Mathijs de Boer <[email protected]>
Signed-off-by: Fabian Klopfer <[email protected]>
Signed-off-by: Lucas Robinet <[email protected]>
Signed-off-by: Lucas Robinet <[email protected]>
Signed-off-by: chaoliu <[email protected]>
Signed-off-by: cxlcl <[email protected]>
Signed-off-by: chaoliu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: monai-bot <[email protected]>
Co-authored-by: elitap <[email protected]>
Co-authored-by: Felix Schnabel <[email protected]>
Co-authored-by: YanxuanLiu <[email protected]>
Co-authored-by: ytl0623 <[email protected]>
Co-authored-by: Dženan Zukić <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: Ishan Dutta <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Kaibo Tang <[email protected]>
Co-authored-by: Yufan He <[email protected]>
Co-authored-by: binliunls <[email protected]>
Co-authored-by: Ben Murray <[email protected]>
Co-authored-by: axel.vlaminck <[email protected]>
Co-authored-by: Mingxin Zheng <[email protected]>
Co-authored-by: Ibrahim Hadzic <[email protected]>
Co-authored-by: Dr. Behrooz Hashemian <[email protected]>
Co-authored-by: Timothy J. Baker <[email protected]>
Co-authored-by: Mathijs de Boer <[email protected]>
Co-authored-by: Mathijs de Boer <[email protected]>
Co-authored-by: Fabian Klopfer <[email protected]>
Co-authored-by: Yiheng Wang <[email protected]>
Co-authored-by: Lucas Robinet <[email protected]>
Co-authored-by: Lucas Robinet <[email protected]>
Co-authored-by: cxlcl <[email protected]>
@KumoLiu @ericspod
Remove device count cache when import monai (#7581)
2716b6a 
workaround for #7575


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
@Lucas-rbnt @KumoLiu
Fixing gradient in sincos positional encoding in monai/networks/block…
c9fed96 
…s/patchembedding.py (#7564)

### Description

When you choose to put the argument `pos_embed_type='sincos'` in the
`PatchEmbeddingBlock` class, it still return a learnable positional
encoding

To reproduce:
```python
from monai.networks.blocks import PatchEmbeddingBlock
patcher = PatchEmbeddingBlock(
            in_channels=1,
            img_size=(32, 32, 32),
            patch_size=(8, 8, 8),
            hidden_size=96,
            num_heads=8,
            pos_embed_type="sincos",
            dropout_rate=0.5,
        )
print(patcher.position_embeddings.requires_grad) 
>>> True
```

In the literature, we sometimes use either positional encoding in sincos
which are fixed and non-trainable as in the original Attention Is All
You Need [paper](https://arxiv.org/abs/1706.03762) or a learnable
positional embedding as in the ViT
[paper](https://arxiv.org/abs/2010.11929).
If you choose to use a sincos, then it seems that is must be fixed which
is not the case here.
I'm not completely sure of the desired result in MONAI since there's
already a learnable possibility, so if we choose sincos we'd like
gradient-free parameters. However the documentation of
`build_sincos_position_embedding`in the `pos_embed_utils.py`files
stipulate: "The sin-cos position embedding as a learnable parameter"
which seems a bit confusing. Especially as the encoding construction
function seems to aim to set the require gradient to False (see below)

```python
pos_embed = nn.Parameter(pos_emb)
pos_embed.requires_grad = False

return pos_embed
```
But these changes are not maintained by torch's `copy_` function, which
does not copy gradient parameters (see the cpp code
https://github.com/pytorch/pytorch/blob/148a8de6397be6e4b4ca1508b03b82d117bfb03c/torch/csrc/lazy/ts_backend/tensor_aten_ops.cpp#L51).
This `copy_`is used in the `PatchEmbeddingBlock` class to instantiate
the positional embedding.

I propose a small fix to overcome this problem as well as test cases to
ensure that positional embedding behaves correctly.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Lucas Robinet <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@johnzielke @KumoLiu
Fix inconsistent alpha parameter/docs for RandGibbsNoise/RandGibbsNoi…
ba3c72c 
…sed (#7584)

Fixes inconsistent alpha parameter/docs for
RandGibbsNoise/RandGibbsNoised

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: John Zielke <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@mingxin-zheng @KumoLiu
Fix bundle_root for NNIGen (#7586)
7c0b10e 
Fixes #7585  .

### Description

Because the NNI test takes too much time, the previous behavior did not
get caught with the dry-run mode of HPO Gen

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).

---------

Signed-off-by: Mingxin Zheng <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@monai-bot
Auto3DSeg algo_template hash update (#7589)
2d463a7 
Signed-off-by: monai-bot <[email protected]>

Signed-off-by: monai-bot <[email protected]>
@KumoLiu
Utilizing subprocess for nnUNet training. (#7576)
15d2abf 
Workaround for #7575 

### Description
- Due to the impact of #7575, the operation to set the device within
nnUNetV2Runner will become ineffective. This PR is intended to resolve
this issue.
- Add a version check for #7575, will revisit after the update from
pytorch team.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
@scalyvladimir @KumoLiu
typo fix (#7595)
ec4d946 
Fixes # 1.

### Description

Fixed typo.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] In-line docstrings updated.

Signed-off-by: Vladimir Chernyi <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@monai-bot
auto updates (#7599)
a7c2589 
Signed-off-by: monai-bot <[email protected]>

Signed-off-by: monai-bot <[email protected]>
@yiheng-wang-nv @pre-commit-ci @KumoLiu
7540 change bundle workflow args (#7549)
c885100 
Fixes #7540  .

### Description

This PR:
1. add logging file and meta file into BundleWorkflow
2. add the sequence form of meta files check for ConfigWorkflow

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <[email protected]>
@KumoLiu @pre-commit-ci
Add "properties_path" in BundleWorkflow (#7542)
264b9e4 
Fixes #7541


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@monai-bot
Auto3DSeg algo_template hash update (#7603)
bbaaf4c 
Signed-off-by: monai-bot <[email protected]>

Signed-off-by: monai-bot <[email protected]>
@lorinczszabolcs @KumoLiu
ENH: generate_label_classes_crop_centers: warn only if ratio of missi…
5ec7305 
…ng class is not set to 0 (#7602)

Fixes #7594 

### Description

Only warn users that the ratio of a missing class is set to 0, when it
wasn't already set to 0 by the user, in
`generate_label_classes_crop_centers`, function being used by
`RandCropByLabelClasses` transform.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`. ⚠️ See notes
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`. ⚠️ See notes
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

### Notes regarding tests
Some tests were failing, see details:
<details>
======================================================================
ERROR:
test_cuda_0_2_batches_1_dimensions_1_channels_2_classes_2_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_1_2_1_Linux_3_10_12_22_12_1'

======================================================================
ERROR:
test_cuda_1_1_batches_1_dimensions_5_channels_2_classes_1_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_5_2_1_Linux_3_10_12_22_12_1'

======================================================================
ERROR:
test_cuda_2_1_batches_2_dimensions_2_channels_4_classes_4_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_2_4_1_Linux_3_10_12_22_12_1'

======================================================================
ERROR:
test_cuda_3_1_batches_3_dimensions_1_channels_2_classes_1_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension
'gmm_1_2_1_Linux_3_10_12_22_12_1_v1'

======================================================================
ERROR: test_load (tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 311, in
test_load
load_module("gmm", {"CHANNEL_COUNT": 2, "MIXTURE_COUNT": 2,
"MIXTURE_SIZE": 3}, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_2_2_3_Linux_3_10_12_22_12_1'

======================================================================
ERROR: test_spacing_35 (tests.test_spacing.TestSpacingCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_spacing.py", line 289, in
test_spacing
    res: MetaTensor = tr(**call_param)  # type: ignore
File "/home/szabolcslorincz/MONAI/monai/transforms/spatial/array.py",
line 525, in __call__
    data_array = self.sp_resample(
File "/home/szabolcslorincz/MONAI/monai/transforms/spatial/array.py",
line 223, in __call__
    return spatial_resample(
File
"/home/szabolcslorincz/MONAI/monai/transforms/spatial/functional.py",
line 178, in spatial_resample
img = affine_xform(img.unsqueeze(0), theta=xform.to(img),
spatial_size=spatial_size).squeeze(0) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/nn/modules/module.py",
line 1511, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/nn/modules/module.py",
line 1520, in _call_impl
    return forward_call(*args, **kwargs)
File
"/home/szabolcslorincz/MONAI/monai/networks/layers/spatial_transforms.py",
line 584, in forward
grid = nn.functional.affine_grid(theta=theta[:, :sr],
size=list(dst_size), align_corners=self.align_corners)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/nn/functional.py",
line 4418, in affine_grid
    return torch.affine_grid_generator(theta, size, align_corners)
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate
840.00 MiB. GPU 0 has a total capacity of 8.00 GiB of which 0 bytes is
free. Including non-PyTorch memory, this process has 17179869184.00 GiB
memory in use. Of the allocated memory 6.47 GiB is allocated by PyTorch,
and 156.63 MiB is reserved by PyTorch but unallocated. If reserved but
unallocated memory is large try setting
PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation.
See documentation for Memory Management
(https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

======================================================================
FAIL: test_seg_res_net_1_cuda
(tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
108, in test_seg_res_net
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 233543 / 1451520 (16.1%)
Greatest absolute difference: 0.0014852285385131836 at index (0, 19, 12,
13, 22) (up to 0.0001 allowed)
Greatest relative difference: 589.0405883789062 at index (0, 32, 21, 16,
15) (up to 0.001 allowed)

======================================================================
FAIL: test_unet_4_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6705 / 49152 (13.6%)
Greatest absolute difference: 0.0015408992767333984 at index (11, 0, 21,
27) (up to 0.0001 allowed)
Greatest relative difference: 51.80112838745117 at index (14, 0, 25, 19)
(up to 0.001 allowed)

======================================================================
FAIL: test_unet_5_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6218 / 49152 (12.7%)
Greatest absolute difference: 0.0015670061111450195 at index (2, 0, 23,
14) (up to 0.0001 allowed)
Greatest relative difference: 7.987473964691162 at index (8, 0, 27, 8)
(up to 0.001 allowed)

======================================================================
FAIL: test_unet_6_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6743 / 49152 (13.7%)
Greatest absolute difference: 0.0015552043914794922 at index (1, 1, 9,
11) (up to 0.0001 allowed)
Greatest relative difference: 2.0317020416259766 at index (11, 0, 19,
21) (up to 0.001 allowed)

======================================================================
FAIL: test_unet_7_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6823 / 49152 (13.9%)
Greatest absolute difference: 0.0018431544303894043 at index (10, 0, 9,
19) (up to 0.0001 allowed)
Greatest relative difference: 4.297887325286865 at index (13, 0, 12, 13)
(up to 0.001 allowed)

======================================================================
FAIL: test_shape_2 (tests.test_multi_scale.TestMultiScale)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_multi_scale.py", line 59,
in test_shape
np.testing.assert_allclose(result.detach().cpu().numpy(), expected_val,
rtol=1e-5)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/numpy/testing/_private/utils.py",
line 1504, in assert_allclose
    assert_array_compare(compare, actual, desired, err_msg=str(err_msg),
  File "/usr/lib/python3.10/contextlib.py", line 79, in inner
    return func(*args, **kwds)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/numpy/testing/_private/utils.py",
line 797, in assert_array_compare
    raise AssertionError(msg)
AssertionError: 
Not equal to tolerance rtol=1e-05, atol=0

Mismatched elements: 1 / 1 (100%)
Max absolute difference: 0.
Max relative difference: 0.
 x: array(0.715212, dtype=float32)
 y: array(0.715228)

----------------------------------------------------------------------
Ran 14392 tests in 2252.916s

FAILED (failures=6, errors=6, skipped=450)

---------

Signed-off-by: Szabolcs Botond Lorincz Molnar <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@KumoLiu @Nic-Ma
Update base image to 2403 (#7600)
763347d 
### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: Nic Ma <[email protected]>
@Lucas-rbnt @KumoLiu
simplification of the sincos positional encoding in patchembedding.py (…
195d7dd 
…#7605)

Fixes #7564 .

### Description

As discussed, a small simplification for the creation of sincos
positional encoding where we don't need to use the `torch.no_grad()`
context or copy the tensor with `copy_` from torch which doesn't
preserve the `requires_grad` attribute here.

The changes are simple and are linked to the corresponding comment
#7564, the output is already in float32 so it doesn't seem particularly
necessary to apply the conversion previously done.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Lucas Robinet <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@Lucas-rbnt @KumoLiu @pre-commit-ci
harmonization and clarification of dice losses variants docs and asso…
625967c 
…ciated tests (#7587)

### Description

This PR aims to clarify and harmonise the code for the DiceLoss variants
in the `monai/losses/dice.py` file. With the `to_onehot_y` `softmax` and
`sigmoid` arguments, I didn't necessarily understand the ValueError that
occurred when I passed a target of size NH[WD]. I had a bit of trouble
reading the documentation and understanding it. I thought that they had
to be the same shape as they are displayed, unlike the number of
dimensions in the input, so I added that.
Besides, in the documentation is written:
```python
"""
raises:
      ValueError: When number of channels for target is neither 1 nor the same as input.

"""
```
Trying to reproduce this, we give an input with a number of channels $N$
and target a number of channels of $M$, with $M \neq N$ and $M > 1$.
```python
loss = DiceCELoss()
input = torch.rand(1, 4, 3, 3)
target = torch.randn(1, 2, 3, 3)
loss(input, target)
>: AssertionError: ground truth has different shape (torch.Size([1, 2, 3, 3])) from input (torch.Size([1, 4, 3, 3]))
```
This error in the Dice is an `AssertionError` and not a `ValueError` as
expected and the explanation can be confusing and doesn't give a clear
idea of the error here. The classes concerned and harmonised are
`DiceFocalLoss`, `DiceCELoss` and `GeneralizedDiceFocalLoss` with the
addition of tests that behave correctly and handle this harmonisation.

Also, feel free to modify or make suggestions regarding the changes made
in the docstring to make them more understandable (in my opinion, but
other readers and users will probably have a different view).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Lucas Robinet <[email protected]>
Signed-off-by: Lucas Robinet <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
KumoLiu and others added 29 commits July 23, 2024 17:58
@KumoLiu @pre-commit-ci
Make ViT and Unetr to be torchscript comaptible (#7937)
37917e0 
Fixes #7936

### Description
- Pre-define `self.causal_mask = torch.Tensor()` before register buffer
- Move norm_cross_attn and cross_attn out of if block

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@mingxin-zheng @pre-commit-ci @KumoLiu
Add checks for monai bundles after download and warn if incompatible (#…
316934a 
…7938)

Fixes #7930 .

### Description

Check the monai version in metadata JSON and warn if the version is
newer than the package being used.


### Demonstration

Warning when the version is hardcoded to 1.2 from monaihosting
```
root@MS-7D31:/workspace/MONAI# python -m monai.bundle download spleen_ct_segmentation
2024-07-23 11:00:31,286 - INFO - --- input summary of monai.bundle.scripts.download ---
2024-07-23 11:00:31,286 - INFO - > name: 'spleen_ct_segmentation'
2024-07-23 11:00:31,286 - INFO - > source: 'monaihosting'
2024-07-23 11:00:31,286 - INFO - > remove_prefix: 'monai_'
2024-07-23 11:00:31,286 - INFO - > progress: True
2024-07-23 11:00:31,286 - INFO - ---


2024-07-23 11:00:31,985 - INFO - Expected md5 is None, skip md5 check for file /root/.cache/torch/hub/bundle/spleen_ct_segmentation_v0.5.8.zip.
2024-07-23 11:00:31,986 - INFO - File exists: /root/.cache/torch/hub/bundle/spleen_ct_segmentation_v0.5.8.zip, skipped downloading.
2024-07-23 11:00:31,986 - INFO - Writing into directory: /root/.cache/torch/hub/bundle.
2024-07-23 11:00:32,176 - WARNING - Your MONAI version is 1.2, but the bundle is built on MONAI version 1.3.2.
```

Auto select version if the download src is from NGC
```
root@MS-7D31:/workspace/MONAI# BUNDLE_DOWNLOAD_SRC=ngc python -m monai.bundle download spleen_ct_segmentation
2024-07-23 11:02:12,277 - INFO - --- input summary of monai.bundle.scripts.download ---
2024-07-23 11:02:12,277 - INFO - > name: 'spleen_ct_segmentation'
2024-07-23 11:02:12,277 - INFO - > source: 'ngc'
2024-07-23 11:02:12,277 - INFO - > remove_prefix: 'monai_'
2024-07-23 11:02:12,277 - INFO - > progress: True
2024-07-23 11:02:12,277 - INFO - ---


monai_spleen_ct_segmentation_v0.3.7.zip: 34.0MB [00:01, 24.1MB/s]                                                                                                                                                       
2024-07-23 11:02:17,953 - INFO - Downloaded: /root/.cache/torch/hub/bundle/monai_spleen_ct_segmentation_v0.3.7.zip
2024-07-23 11:02:17,954 - INFO - Expected md5 is None, skip md5 check for file /root/.cache/torch/hub/bundle/monai_spleen_ct_segmentation_v0.3.7.zip.
2024-07-23 11:02:17,954 - INFO - Writing into directory: /root/.cache/torch/hub/bundle/spleen_ct_segmentation.
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Mingxin Zheng <[email protected]>
Signed-off-by: Mingxin Zheng <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <[email protected]>
@hjmjohnson @ericspod
Cleanup warnings when collecting tests (#7914)
12d00ce 
Minimize warning information so that important warnings and errors are
more easily accessible

warnings identified with
```bash
python3  -W ignore::DeprecationWarning -m pytest -k dummy
```

Fixing warnings like:
cannot collect test class 'TestTimeAugmentation' because it has a
__init__ constructor

Use `__test__ = False` to suppress attempted collection in those non
test-classes

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.

Signed-off-by: Hans Johnson <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
@KumoLiu @pre-commit-ci
Fix Incompatible types in assignment issue (#7950)
3f0c768 
Fixes #7947

### Description
Add type ignore in several redefinitions cases

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@ericspod @pre-commit-ci @yiheng-wang-nv
Adding metadata schema to the code base itself (#7409)
2e53df7 
Fixes #7303 #6959.

### Description

This adds the schema file into the code base (but this maybe should be
elsewhere). The changes implement a number of new things:

* Moved definitions into a `$defs` section per the JSON schema standard
* Permits multiple input arguments and return results from networks with
arbitrary names using the `patternProperties` mechanism
* Allows the types of inputs and outputs to be, additional to just
tensors, numbers, booleans, or strings
* Outputs after post processing can be specified with the
`post_processed_outputs` section if they are significantly changed with
the post-process transforms defined in scripts
* Multiple network IO formats can be specified in addition to
`network_data_format`, these must follow the pattern
`<name>_data_format`
* `required_packages_version` added in addition to
`optional_packages_version`

#7253 depends on this schema change.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Eric Kerfoot <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Yiheng Wang <[email protected]>
@virginiafdez
Addition of norm_eps (#7962)
54019e4 
Fixes # .

### Description

Addition of norm_eps to spade_autoencoderkl.py as per
#7946.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Virginia Fernandez <[email protected]>
Signed-off-by: Virginia Fernandez <[email protected]>
Co-authored-by: Virginia Fernandez <[email protected]>
Co-authored-by: Virginia Fernandez <[email protected]>
@KumoLiu @ericspod @mingxin-zheng
Rename optional_packages_version to required_packages_version (#7253
f1ef3e8 
)

Fixes #6959.

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: KumoLiu <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
Co-authored-by: Mingxin Zheng <[email protected]>
@KumoLiu
Fix outdated link in the docs (#7971)
139b62c 
Fixes #7968

### Description

Fix outdated link in the docs

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <[email protected]>
@yiheng-wang-nv @KumoLiu
7982-fix-ci-issue (#7983)
1ece8a5 
Fixes #7982 

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@yiheng-wang-nv @KumoLiu
7973-add-ngc-prefix (#7974)
ae5a04d 
Fixes #7973 .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Yiheng Wang <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@dcfidalgo @KumoLiu
Fix: Small logic mistake in the AsDiscrete.__call__ method (#7984)
56ee32e 
Hi MONAI Team! 
Thank you very much for this super nice framework, really appreciate it!
Just found a small logic mistake in one of the transform classes. To
reproduce:
```python
import torch
from monai.transforms.post.array import AsDiscrete

transform = AsDiscrete(argmax=True)
prediction = torch.rand(2, 3, 3)

transform(prediction, argmax=False)
# will still apply argmax
```

### Description

Proposed fix: `argmax` is explicitly checked for `None` in the `__cal__`
method.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: David Carreto Fidalgo <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@virginiafdez @KumoLiu @yiheng-wang-nv
Flash attention (#7977)
6c23fd0 
Fixes #7944.

### Description

In response to Issue #7944,
I added the new functionality scaled_dot_product_attention from PyTorch
to re-enable flash attention, present in the original MONAI Generative
Models repository. This is allowed for torch >= 2.0 and when argument
save_attn = False. Errors are raised otherwise. I ran quick tests and
added some checks on test_selfattention and test_crossattention scripts
to make sure the outputs are the same as not using flash attention.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Virginia Fernandez <[email protected]>
Co-authored-by: Virginia Fernandez <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: Yiheng Wang <[email protected]>
@yiheng-wang-nv @KumoLiu
7994-enhance-mlpblock (#7995)
49a1e34 
Fixes #7994  .

### Description
The current implementation does not support tuple input of "GEGLU" since
it only change the out features of the first linear layer when the input
is a string of "GEGLU".

This PR enhances it, and also enable "vista3d" mode to support #7987 
Tests are added to cover the changes.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@Bala93 @pre-commit-ci
@ericspod @KumoLiu
Initial commit -- Adding calibration loss specific to segmentation (#…
660891f 
…7819)

### Description

Model calibration has helped in developing reliable deep learning
models. In this pull request, I have added a new loss function NACL
(https://arxiv.org/abs/2303.06268, https://arxiv.org/abs/2401.14487)
which has shown promising results for both discriminative and
calibration in segmentation.

**Future Plans:** Currently, MONAI has some of the alternative loss
functions (Label Smoothing, and Focal Loss), but it doesn't have the
calibration specific loss functions (https://arxiv.org/abs/2111.15430,
https://arxiv.org/abs/2209.09641). Besides, these methods are better
evaluated with calibration metrics, Expected Calibration Error
(https://lightning.ai/docs/torchmetrics/stable/classification/calibration_error.html).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Balamurali <[email protected]>
Signed-off-by: bala93 <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@KumoLiu
Ensure location as tuple in wsireader (#8007)
4a3117f 
Fixes #8006

### Description

Ensure location as tuple in wsireader

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <[email protected]>
@ytl0623 @KumoLiu
Add label smoothing param in DiceCELoss (#8000)
0bb05d7 
Fixes #7957 

### Description
In this modified version I made the following changes:
1. Added `label_smoothing: float = 0.0` parameter in `__init__` method,
default value is 0.0.
2. When creating the `self.cross_entropy` instance, pass the
`label_smoothing` parameter to `nn.CrossEntropyLoss`.
3. Added `self.label_smoothing = label_smoothing` in the `__init__`
method to save this parameter for access when needed.

For example:
```
from monai.losses import DiceCELoss

# Before
criterion = DiceCELoss()
criterion.cross_entropy.label_smoothing = 0.1

# Now
criterion = DiceCELoss(label_smoothing=0.1)
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: ytl0623 <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@KumoLiu @pre-commit-ci @ericspod
Add include_fc and use_combined_linear argument in the SABlock (#…
069519d 
…7996)

Fixes #7991
Fixes #7992

### Description
Add `include_fc` and `use_combined_linear` argument in the `SABlock`.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <[email protected]>
@K-Rilla @Saelra
@pre-commit-ci @dedeepyasai @KumoLiu @ken-ni @DuretiShemsi
Replaced package "pkg_resources" with "packaging" (#7953)
6be7b13 
Fixes #7559 .

### Description  
Replaced "pkg_resources" references with "packaging" in
MONAI/monai/utils/module.py & setup.py
Changes were made in functions "pytorch_after", "version_leq",
"version_geq".

### Types of changes
- Non-breaking change (fix or new feature that would not break existing
functionality).

---------

Signed-off-by: dedeepyasai <[email protected]>
Signed-off-by: saelra <[email protected]>
Signed-off-by: Kelvin R <[email protected]>
Signed-off-by: ken-ni <[email protected]>
Signed-off-by: Dureti <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Co-authored-by: saelra <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: dedeepyasai <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: Ratanachat Saelee <[email protected]>
Co-authored-by: ken-ni <[email protected]>
Co-authored-by: Dureti <[email protected]>
@yiheng-wang-nv @KumoLiu @ericspod
Add utils for vista3d (#7999)
f848002 
This PR is a part of #7987 

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Signed-off-by: Yiheng Wang <[email protected]>
Co-authored-by: YunLiu <[email protected]>
Co-authored-by: Eric Kerfoot <[email protected]>
@myron @pre-commit-ci @KumoLiu
Adding a network CellSamWrapper (#7981)
6243031 
Adding a network CellSamWrapper, a thin wrapper around SAM, which can be
used for 2D segmentation tasks.



### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: am <am>
Signed-off-by: myron <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Co-authored-by: am <am>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <[email protected]>
@guopengf @pre-commit-ci
Refactor DiffusionModelUNetMaisi (#7989)
250c18d 
Fixes #7988 .

### Description

Refactor DiffusionModelUNetMaisi to use monai core components. 

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@KumoLiu @pre-commit-ci
Remove segment-anything in setup.cfg (#8010)
7a6f680 
Fixes #8009

### Description

Remove segment-anything in setup.cfg

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@guopengf @pre-commit-ci @KumoLiu
Refactor AutoencoderKlMaisi (#7993)
6858114 
Fixes #7988 .

### Description

Refactor AutoencoderKlMaisi to use monai core components.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <[email protected]>
Signed-off-by: YunLiu <[email protected]>
Signed-off-by: Pengfei Guo <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <[email protected]>
@guopengf @pre-commit-ci @KumoLiu
Refactor ControlNetMaisi (#8005)
9dbfe16 
Fixes #7988 .

### Description

Refactor ControlNetMaisi to use monai core components.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <[email protected]>
@KumoLiu
Fix ci issue in test_vit (#8013)
34ce94d 
Fixes #8012


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
@KumoLiu @pre-commit-ci
Fix module can not import correctly issue (#8015)
4877767 
Fixes #8014


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
@KumoLiu
Fix 'torch.device' object has no attribute 'gpu_id' issue in trt expo…
e85580a 
…rt (#8019)

Part of #8017


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <[email protected]>
@heyufan1995 @pre-commit-ci
@yiheng-wang-nv @KumoLiu
Add vista network (#7987)
77304dd 
Fixes # .

### Description

Add VISTA3D model architecture to MONAI core

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: heyufan1995 <[email protected]>
Signed-off-by: Yufan He <[email protected]>
Signed-off-by: Yiheng Wang <[email protected]>
Signed-off-by: Yiheng Wang <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Yiheng Wang <[email protected]>
Co-authored-by: Yiheng Wang <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@Bala93 @pre-commit-ci
@ericspod @KumoLiu
NACLLoss memory management (#8020)
7b9a523 
Fixes # .

### Description

Calling contiguous after applying the permute option to work with view
operation in apply_filter
(https://github.com/Project-MONAI/MONAI/blob/59a7211070538586369afd4a01eca0a7fe2e742e/monai/networks/layers/simplelayers.py#L293).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Balamurali <[email protected]>
Signed-off-by: bala93 <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <[email protected]>
Co-authored-by: YunLiu <[email protected]>
@KumoLiu KumoLiu closed this Aug 19, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.