- 
@@ -154,7 +154,7 @@ To install additional packages required for development, tests, docs, or tutoria
Getting started with PINA
- +
@@ -169,19 +169,19 @@ Solving a differential problem in PINA follows a clean four-step pipeline
- Define the problem and constraints using the - Problem API. + Problem API.
- Design your model using PyTorch, PyTorch Geometric, or import from the - Model API. + Model API.
- Select or build a Solver using the - Solver API. + Solver API.
- Train with the - Trainer API, + Trainer API, powered by PyTorch Lightning.
-
@@ -381,6 +381,76 @@ If you would like to contribute, please read the
+ +
+
+
+
+
+
+
+
+ Fundings
+
+
+ + A significant part of PINA has been developed either as a + by-product of funded research projects or through contributions + from researchers holding university-funded positions. + We gratefully acknowledge this support, particularly from: +
+ +-
+
- + + H2020 ERC CoG 2015 AROMA-CFD project (Grant No. 681447) + , + led by Principal Investigator + + Prof. Gianluigi Rozza + + at + + SISSA MathLab + . + +
- + + Next Generation EU + , + supporting Italy's environmental and digital transitions. + +
+
+ 
+
+
+
+
+
+ + We also acknowledge the contribution of Maria Strazzullo to the early development of the package. + Special thanks go to all the students and researchers from different universities who have contributed to its development. + Finally, we warmly thank all the contributors + who are the true heart of PINA! +
+ +
+
+ 
+
+
+
+
+
+ +
Citation
diff --git a/SECURITY.md b/SECURITY.md
index a425740c2..794df5667 100644
--- a/SECURITY.md
+++ b/SECURITY.md
@@ -16,4 +16,4 @@ Security fixes are given priority and might be enough to cause a new version to
## Reporting a Vulnerability
-To ensure vulnerability reports reach the maintainers as quickly as possible, the preferred way is to use the ["Report a vulnerability"](https://github.com/mathLab/PINA/security/advisories/new) button under the "Security" tab of the associated GitHub project. This creates a private communication channel between the reporter and the maintainers.
+To ensure vulnerability reports reach the maintainers as quickly as possible, the preferred way is to use the ["Report a vulnerability"](https://github.com/PINA-org/PINA/security/advisories/new) button under the "Security" tab of the associated GitHub project. This creates a private communication channel between the reporter and the maintainers.
diff --git a/STEERING-COMMITTEE.md b/STEERING-COMMITTEE.md
index b97c3a0b9..a44326596 100644
--- a/STEERING-COMMITTEE.md
+++ b/STEERING-COMMITTEE.md
@@ -1,13 +1,12 @@
# Steering Committee
-This document lists the members of the Organization's Steering Committee (in alphabetical order). Voting members may be added once approved by the Steering Committee as described in the [charter](github.com/mathLab/PINA/blob/master/CHARTER.md). By adding your name to this list you are agreeing to abide by all Organization polices, including the [charter](github.com/mathLab/PINA/blob/master/CHARTER.md), the [code of conduct](https://github.com/mathLab/PINA/blob/master/CODE_OF_CONDUCT.md), the [trademark policy](https://github.com/mathLab/PINA/blob/master/TRADEMARKS.md), and the [antitrust policy](https://github.com/mathLab/PINA/blob/master/ANTITRUST.md). If you are serving on the Steering Committee because of your affiliation with another organization (designated below), you represent that you have authority to bind that organization to these policies.
+This document lists the members of the Organization's Steering Committee (in alphabetical order). Voting members may be added once approved by the Steering Committee as described in the [charter](https://github.com/PINA-org/PINA/blob/master/CHARTER.md). By adding your name to this list you are agreeing to abide by all Organization polices, including the [charter](https://github.com/PINA-org/PINA/blob/master/CHARTER.md), the [code of conduct](https://github.com/PINA-org/PINA/blob/master/CODE_OF_CONDUCT.md), the [trademark policy](https://github.com/PINA-org/PINA/blob/master/TRADEMARKS.md), and the [antitrust policy](https://github.com/PINA-org/PINA/blob/master/ANTITRUST.md). If you are serving on the Steering Committee because of your affiliation with another organization (designated below), you represent that you have authority to bind that organization to these policies.
| **NAME** | **Handle** | **Affiliated Organization** |
| ------------ | ------------ | --------------------------- |
| Giovanni Canali | @GiovanniCanali | SISSA |
-| Dario Coscia | @dario-coscia | SISSA |
+| Dario Coscia | @dario-coscia | CuspAI |
| Nicola Demo | @ndem0 | SISSA - FAST COMPUTING SRL |
-| Filippo Olivo | @FilippoOlivo | SISSA - FAST COMPUTING SRL |
---
## Attribution
diff --git a/docs/source/_contributing.rst b/docs/source/_contributing.rst
index b981f6ba7..ea2073534 100644
--- a/docs/source/_contributing.rst
+++ b/docs/source/_contributing.rst
@@ -22,7 +22,7 @@ We encourage all contributions, big or small!
Reporting Bugs
--------------
-If you find a bug, please open an `issue `_ and include:
+If you find a bug, please open an `issue `_ and include:
- A clear and descriptive title
- Steps to reproduce the problem
@@ -36,7 +36,7 @@ Suggesting Enhancements
We welcome new ideas! If you have an idea to improve PINA:
-1. Check the `issue tracker `_ or the `discussions `_ to see if someone has already suggested it.
+1. Check the `issue tracker `_ or the `discussions `_ to see if someone has already suggested it.
2. If not, open a new issue describing:
- The enhancement you would like
- Why it would be useful
diff --git a/docs/source/_installation.rst b/docs/source/_installation.rst
index a0bca26f5..1e4eb3541 100644
--- a/docs/source/_installation.rst
+++ b/docs/source/_installation.rst
@@ -29,7 +29,7 @@ The official distribution is on GitHub. Clone the repository:
.. code-block:: bash
- git clone https://github.com/mathLab/PINA
+ git clone https://github.com/PINA-org/PINA
Then install in editable mode:
diff --git a/docs/source/_team.rst b/docs/source/_team.rst
index f1fc47422..fbd48acd0 100644
--- a/docs/source/_team.rst
+++ b/docs/source/_team.rst
@@ -5,27 +5,16 @@
PINA Team
==========
-**PINA** is part of the `PyTorch ecosystem `_ and currently developed at the `SISSA MathLab `_ . A significant part of **PINA** has been written either as a by-product of funded projects or by people on university-funded positions. We are very grateful for this support. In particular, we acknowledge:
* `H2020 ERC CoG 2015 AROMA-CFD project 681447 `_ at `SISSA MathLab `_ who are the real heart of **PINA**!
+We also acknowledge the contribution of `Maria Strazzullo `_ who are the real heart of **PINA**!
.. figure:: index_files/university_dev_pina.png
:align: center
diff --git a/docs/source/_templates/layout.html b/docs/source/_templates/layout.html
index c1bc42107..7c0e1c383 100644
--- a/docs/source/_templates/layout.html
+++ b/docs/source/_templates/layout.html
@@ -6,7 +6,7 @@
{% trans copyright=copyright|e %}© Copyright {{ copyright }}, PINA Developers.{% endtrans %}
+ href="https://github.com/PINA-org/PINA/">PINA Developers.{% endtrans %}
Created using Sphinx and the PyData Theme.
diff --git a/docs/source/conf.py b/docs/source/conf.py
index 7f5f4f8c7..9e09f6600 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -127,7 +127,7 @@
"icon_links": [
{
"name": "GitHub",
- "url": "https://github.com/mathLab/PINA",
+ "url": "https://github.com/PINA-org/PINA",
"icon": "fab fa-github",
"type": "fontawesome",
},
diff --git a/docs/source/index.rst b/docs/source/index.rst
index 924a7fdbf..39d6c4773 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -16,7 +16,7 @@ Welcome to PINA's documentation!
Get Started
Quickstart
- View on GitHub
+ View on GitHub
@@ -185,7 +185,7 @@ Physics-Informed Neural Networks (PINNs), Neural Operators, and more.
API <_rst/_code>
Tutorials <_tutorial>
Contributing <_contributing>
- Team & Foundings <_team.rst>
+ Team & Fundings <_team.rst>
Installing <_installation>
Cite PINA <_cite.rst>
License <_LICENSE.rst>
diff --git a/docs/source/index_files/foudings.png b/docs/source/index_files/fundings.png
similarity index 100%
rename from docs/source/index_files/foudings.png
rename to docs/source/index_files/fundings.png
diff --git a/joss/paper.md b/joss/paper.md
index b63d46d43..133266e85 100644
--- a/joss/paper.md
+++ b/joss/paper.md
@@ -62,8 +62,8 @@ Among all these frameworks, PINA wants to emerge for its easiness of usage, allo
Built over PyTorch --- in order to inherit the \verb+autograd+ module and all the other features already implemented --- PINA provides indeed documented API to explain usage and capabilities of the different classes. We have built several abstract interfaces not only for better structure of the source code but especially to give the final user an easy entry point to implement their own extensions, like new loss functions, new training procedures, and so on. This aspect, together with the capability to use all the PyTorch models, makes it possible to incorporate almost any existing architecture into the PINA framework.
We have decided to build it on top of PyTorch in order to exploit the \verb+autograd+ module, as well as all the other features implemented in this framework. The final outcome is then a library with incremental complexity, capable of being used by the new users to perform the first investigation using PINNs, but also as a core framework to actively develop new features to improve the discussed methodology.
-The high-level structure of the package is depicted in our [API](https://github.com/mathLab/PINA/tree/master/readme/API_color.png); the approximated solution of a differential equation can be implemented using PINA in a few lines of code thanks to the intuitive and user-friendly interface.
-Besides the user-friendly interface, PINA also offers several examples and tutorials, aiming to guide new users toward an easy exploration of the software features. The online documentation is released at \url{https://mathlab.github.io/PINA/}, while the robustness of the package is continuously monitored by unit tests.
+The high-level structure of the package is depicted in our [API](https://github.com/PINA-org/PINA/tree/master/readme/PINA_API.png); the approximated solution of a differential equation can be implemented using PINA in a few lines of code thanks to the intuitive and user-friendly interface.
+Besides the user-friendly interface, PINA also offers several examples and tutorials, aiming to guide new users toward an easy exploration of the software features. The online documentation is released at \url{https://PINA-org.github.io/PINA/}, while the robustness of the package is continuously monitored by unit tests.
PINA workflow is characterized by 3 main steps: the problem formulation, the model definition, i.e., the structure of the neural network used, and the training, eventually followed by the data visualization.
diff --git a/pina/_src/core/label_tensor.py b/pina/_src/core/label_tensor.py
index d97ef077d..4aebb7206 100644
--- a/pina/_src/core/label_tensor.py
+++ b/pina/_src/core/label_tensor.py
@@ -374,24 +374,52 @@ def cat(tensors, dim=0):
return cat_tensor
@staticmethod
- def stack(tensors):
+ def stack(tensors, dim=0):
"""
Stacks a list of tensors along a new dimension. For more details, see
:meth:`torch.stack`.
:param list[LabelTensor] tensors: A list of tensors to stack.
All tensors must have the same shape.
+ :param int dim: Dimension along which to stack the tensors. Default is 0.
:return: A new :class:`~pina.label_tensor.LabelTensor` instance obtained
by stacking the input tensors.
:rtype: LabelTensor
"""
+ if not tensors:
+ return []
+ if len(tensors) == 1:
+ return tensors[0]
+
+ # In PyTorch, stacking adds a new dimension, so valid dims are - (ndim + 1) <= dim <= ndim
+ ndim = tensors[0].ndim
+ if dim < 0:
+ dim = ndim + 1 + dim
+ if dim < 0 or dim > ndim:
+ raise IndexError(
+ f"Dimension out of range (expected to be in range of [{-(ndim + 1)}, {ndim}], but got {dim})"
+ )
+
+ # Check that all tensors have consistent labels
+ tensors_labels = [tensor.stored_labels for tensor in tensors]
+ for key in tensors_labels[0]:
+ if any(
+ tensors_labels[i].get(key) != tensors_labels[0][key]
+ for i in range(len(tensors_labels))
+ ):
+ raise RuntimeError("All tensors must have the same labels.")
# Perform stacking in torch
- new_tensor = torch.stack(tensors)
+ new_tensor = torch.stack(tensors, dim=dim)
+
+ # Shift label dimension indices at or after `dim` by +1
+ labels = {}
+ for key, value in tensors[0]._labels.items():
+ if key >= dim:
+ labels[key + 1] = value
+ else:
+ labels[key] = value
- # Increase labels keys by 1
- labels = tensors[0]._labels
- labels = {key + 1: value for key, value in labels.items()}
new_tensor._labels = labels
return new_tensor
diff --git a/pina/_src/problem/zoo/inverse_poisson_problem.py b/pina/_src/problem/zoo/inverse_poisson_problem.py
index 9a5a8d908..b30b5d596 100644
--- a/pina/_src/problem/zoo/inverse_poisson_problem.py
+++ b/pina/_src/problem/zoo/inverse_poisson_problem.py
@@ -154,11 +154,11 @@ def __init__(self, load=True, data_size=1.0):
# Define URLs for input and output data
input_url = (
- "https://github.com/mathLab/PINA/raw/refs/heads/master"
+ "https://github.com/PINA-org/PINA/raw/refs/heads/master"
"/tutorials/tutorial7/data/pts_0.5_0.5"
)
output_url = (
- "https://github.com/mathLab/PINA/raw/refs/heads/master"
+ "https://github.com/PINA-org/PINA/raw/refs/heads/master"
"/tutorials/tutorial7/data/pinn_solution_0.5_0.5"
)
diff --git a/pyproject.toml b/pyproject.toml
index c15169843..0b0c276e0 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -47,8 +47,8 @@ tutorial = [
]
[project.urls]
-Homepage = "https://mathlab.github.io/PINA/"
-Repository = "https://github.com/mathLab/PINA"
+Homepage = "https://pina-org.github.io/PINA/"
+Repository = "https://github.com/PINA-org/PINA"
[build-system]
requires = [ "setuptools>=41", "wheel", "setuptools-git-versioning>=2.0,<3", ]
diff --git a/tests/test_label_tensor/test_label_tensor.py b/tests/test_label_tensor/test_label_tensor.py
index ca4ae2f1a..e770fe3a8 100644
--- a/tests/test_label_tensor/test_label_tensor.py
+++ b/tests/test_label_tensor/test_label_tensor.py
@@ -229,6 +229,166 @@ def test_vstack_2D():
assert lt_stacked.full_labels[1]["name"] == "second"
+def test_stack_2D():
+ data_1 = torch.rand(20, 3)
+ labels = {1: {"dof": ["x", "y", "z"], "name": "vars"}}
+ lt1 = LabelTensor(data_1, labels)
+ lt2 = LabelTensor(data_1, labels)
+
+ # Test default dim=0
+ lt_stacked_0 = LabelTensor.stack([lt1, lt2], dim=0)
+ assert lt_stacked_0.shape == (2, 20, 3)
+ assert lt_stacked_0.full_labels[0]["dof"] == range(2)
+ assert lt_stacked_0.full_labels[0]["name"] == 0
+ assert lt_stacked_0.full_labels[1]["dof"] == range(20)
+ assert lt_stacked_0.full_labels[1]["name"] == 1
+ assert lt_stacked_0.full_labels[2]["dof"] == ["x", "y", "z"]
+ assert lt_stacked_0.full_labels[2]["name"] == "vars"
+ assert lt_stacked_0.stored_labels == {
+ 2: {"dof": ["x", "y", "z"], "name": "vars"}
+ }
+ assert lt_stacked_0.labels == ["x", "y", "z"]
+
+ # Test dim=1
+ lt_stacked_1 = LabelTensor.stack([lt1, lt2], dim=1)
+ assert lt_stacked_1.shape == (20, 2, 3)
+ assert lt_stacked_1.full_labels[0]["dof"] == range(20)
+ assert lt_stacked_1.full_labels[0]["name"] == 0
+ assert lt_stacked_1.full_labels[1]["dof"] == range(2)
+ assert lt_stacked_1.full_labels[1]["name"] == 1
+ assert lt_stacked_1.full_labels[2]["dof"] == ["x", "y", "z"]
+ assert lt_stacked_1.full_labels[2]["name"] == "vars"
+ assert lt_stacked_1.stored_labels == {
+ 2: {"dof": ["x", "y", "z"], "name": "vars"}
+ }
+ assert lt_stacked_1.labels == ["x", "y", "z"]
+
+ # Test dim=2 and negative dim=-1
+ lt_stacked_2 = LabelTensor.stack([lt1, lt2], dim=2)
+ lt_stacked_neg = LabelTensor.stack([lt1, lt2], dim=-1)
+ for lt_s in [lt_stacked_2, lt_stacked_neg]:
+ assert lt_s.shape == (20, 3, 2)
+ assert lt_s.full_labels[0]["dof"] == range(20)
+ assert lt_s.full_labels[0]["name"] == 0
+ assert lt_s.full_labels[1]["dof"] == ["x", "y", "z"]
+ assert lt_s.full_labels[1]["name"] == "vars"
+ assert lt_s.full_labels[2]["dof"] == range(2)
+ assert lt_s.full_labels[2]["name"] == 2
+ assert lt_s.stored_labels == {
+ 1: {"dof": ["x", "y", "z"], "name": "vars"}
+ }
+ assert lt_s.labels is None
+
+
+def test_stack_3D():
+ data_1 = torch.rand(20, 3, 2)
+ labels_1 = {
+ 1: {"dof": ["a", "b", "c"], "name": "first"},
+ 2: {"dof": ["x", "y"], "name": "second"},
+ }
+ lt1 = LabelTensor(data_1, labels_1)
+ data_2 = torch.rand(20, 3, 2)
+ lt2 = LabelTensor(data_2, labels_1)
+
+ # Test dim=0 (and dim=-4)
+ lt_stacked_0 = LabelTensor.stack([lt1, lt2], dim=0)
+ lt_stacked_neg4 = LabelTensor.stack([lt1, lt2], dim=-4)
+ for lt_s in [lt_stacked_0, lt_stacked_neg4]:
+ assert lt_s.shape == (2, 20, 3, 2)
+ assert lt_s.full_labels[0]["dof"] == range(2)
+ assert lt_s.full_labels[0]["name"] == 0
+ assert lt_s.full_labels[1]["dof"] == range(20)
+ assert lt_s.full_labels[1]["name"] == 1
+ assert lt_s.full_labels[2]["dof"] == ["a", "b", "c"]
+ assert lt_s.full_labels[2]["name"] == "first"
+ assert lt_s.full_labels[3]["dof"] == ["x", "y"]
+ assert lt_s.full_labels[3]["name"] == "second"
+ assert lt_s.stored_labels == {
+ 2: {"dof": ["a", "b", "c"], "name": "first"},
+ 3: {"dof": ["x", "y"], "name": "second"},
+ }
+ assert lt_s.labels == ["x", "y"]
+
+ # Test dim=1 (and dim=-3)
+ lt_stacked_1 = LabelTensor.stack([lt1, lt2], dim=1)
+ lt_stacked_neg3 = LabelTensor.stack([lt1, lt2], dim=-3)
+ for lt_s in [lt_stacked_1, lt_stacked_neg3]:
+ assert lt_s.shape == (20, 2, 3, 2)
+ assert lt_s.full_labels[0]["dof"] == range(20)
+ assert lt_s.full_labels[0]["name"] == 0
+ assert lt_s.full_labels[1]["dof"] == range(2)
+ assert lt_s.full_labels[1]["name"] == 1
+ assert lt_s.full_labels[2]["dof"] == ["a", "b", "c"]
+ assert lt_s.full_labels[2]["name"] == "first"
+ assert lt_s.full_labels[3]["dof"] == ["x", "y"]
+ assert lt_s.full_labels[3]["name"] == "second"
+ assert lt_s.stored_labels == {
+ 2: {"dof": ["a", "b", "c"], "name": "first"},
+ 3: {"dof": ["x", "y"], "name": "second"},
+ }
+ assert lt_s.labels == ["x", "y"]
+
+ # Test dim=2 (and dim=-2)
+ lt_stacked_2 = LabelTensor.stack([lt1, lt2], dim=2)
+ lt_stacked_neg2 = LabelTensor.stack([lt1, lt2], dim=-2)
+ for lt_s in [lt_stacked_2, lt_stacked_neg2]:
+ assert lt_s.shape == (20, 3, 2, 2)
+ assert lt_s.full_labels[0]["dof"] == range(20)
+ assert lt_s.full_labels[0]["name"] == 0
+ assert lt_s.full_labels[1]["dof"] == ["a", "b", "c"]
+ assert lt_s.full_labels[1]["name"] == "first"
+ assert lt_s.full_labels[2]["dof"] == range(2)
+ assert lt_s.full_labels[2]["name"] == 2
+ assert lt_s.full_labels[3]["dof"] == ["x", "y"]
+ assert lt_s.full_labels[3]["name"] == "second"
+ assert lt_s.stored_labels == {
+ 1: {"dof": ["a", "b", "c"], "name": "first"},
+ 3: {"dof": ["x", "y"], "name": "second"},
+ }
+ assert lt_s.labels == ["x", "y"]
+
+ # Test dim=3 (and dim=-1)
+ lt_stacked_3 = LabelTensor.stack([lt1, lt2], dim=3)
+ lt_stacked_neg1 = LabelTensor.stack([lt1, lt2], dim=-1)
+ for lt_s in [lt_stacked_3, lt_stacked_neg1]:
+ assert lt_s.shape == (20, 3, 2, 2)
+ assert lt_s.full_labels[0]["dof"] == range(20)
+ assert lt_s.full_labels[0]["name"] == 0
+ assert lt_s.full_labels[1]["dof"] == ["a", "b", "c"]
+ assert lt_s.full_labels[1]["name"] == "first"
+ assert lt_s.full_labels[2]["dof"] == ["x", "y"]
+ assert lt_s.full_labels[2]["name"] == "second"
+ assert lt_s.full_labels[3]["dof"] == range(2)
+ assert lt_s.full_labels[3]["name"] == 3
+ assert lt_s.stored_labels == {
+ 1: {"dof": ["a", "b", "c"], "name": "first"},
+ 2: {"dof": ["x", "y"], "name": "second"},
+ }
+ assert lt_s.labels is None
+
+
+def test_stack_errors():
+ data_1 = torch.rand(20, 3)
+ lt1 = LabelTensor(data_1, {1: {"dof": ["x", "y", "z"], "name": "vars"}})
+ lt2 = LabelTensor(data_1, {1: {"dof": ["a", "b", "c"], "name": "vars"}})
+
+ # Empty list
+ assert LabelTensor.stack([]) == []
+
+ # Single tensor
+ assert LabelTensor.stack([lt1]) is lt1
+
+ # Out of range dim
+ with pytest.raises(IndexError):
+ LabelTensor.stack([lt1, lt1], dim=3)
+ with pytest.raises(IndexError):
+ LabelTensor.stack([lt1, lt1], dim=-4)
+
+ # Inconsistent labels
+ with pytest.raises(RuntimeError):
+ LabelTensor.stack([lt1, lt2])
+
+
def test_sorting():
data = torch.ones(20, 5)
data[:, 0] = data[:, 0] * 4
diff --git a/utils/mathlab_versioning.py b/utils/pina_versioning.py
similarity index 100%
rename from utils/mathlab_versioning.py
rename to utils/pina_versioning.py
{% trans copyright=copyright|e %}© Copyright {{ copyright }}, PINA Developers.{% endtrans %}
+ href="https://github.com/PINA-org/PINA/">PINA Developers.{% endtrans %}
Created using Sphinx and the PyData Theme.