Add external pages (#54)

* add mbd

* bump st version

* minor adjust

* restore loadscope to avoid missing ocp models

* clean workflows

* clean workflows

.github/README.md

@@ -125,14 +125,10 @@ streamlit run serve/app.py
 ### Add new benchmark tasks (WIP)
 
 > [!NOTE]
-> Please reuse or extend the general tasks defined as Prefect / [Atomate2](https://github.com/materialsproject/atomate2) / [Quacc](https://github.com/Quantum-Accelerators/quacc) workflow. 
-> The following are some tasks implemented:
-> - [Prefect structure optimization (OPT)](../mlip_arena/tasks/optimize.py)
-> - [Prefect molecular dynamics (MD)](../mlip_arena/tasks/md.py)
-> - [Prefect equation of states (EOS)](../mlip_arena/tasks/eos.py)
+> Please reuse, extend, or chain the general tasks defined [above](#list-of-implemented-tasks)
 
 <!-- 1. Follow the task template to implement the task class and upload the script along with metadata to the MLIP Arena [here](../mlip_arena/tasks/README.md).
-2. Code a benchmark script to evaluate the performance of your model on the task. The script should be able to load the model and the dataset, and output the evaluation metrics. -->
+1. Code a benchmark script to evaluate the performance of your model on the task. The script should be able to load the model and the dataset, and output the evaluation metrics. -->
 
 ### Add new MLIP models 
 

.github/workflows/release.yaml

@@ -1,10 +1,6 @@
 name: Publish Release
 
 on:
-  # workflow_run:
-  #   workflows: [Python Test]
-  #   branches: [main]
-  #   types: [completed]
   workflow_dispatch:
 
 permissions:
@@ -14,7 +10,6 @@ jobs:
   pypi:
     name: Publish to PyPI
     runs-on: ubuntu-latest
-    # needs: release  # This job runs after the release job
 
     steps:
     # Step 1: Checkout the code

.github/workflows/test.yaml

@@ -61,4 +61,4 @@ jobs:
         PREFECT_API_KEY: ${{ secrets.PREFECT_API_KEY }}
         PREFECT_API_URL: ${{ secrets.PREFECT_API_URL }}
       run: |
-        pytest -vra -n 5 tests 
+        pytest -vra -n 5 --dist=loadscope tests 

README.md

@@ -2,7 +2,7 @@
 title: MLIP Arena
 emoji: ⚛
 sdk: streamlit
-sdk_version: 1.36.0 # The latest supported version
+sdk_version: 1.43.2 # The latest supported version
 app_file: serve/app.py
 ---
 

mlip_arena/models/registry.yaml

@@ -114,7 +114,7 @@ ORBv2:
     - combustion
     - stability
   github: https://github.com/orbital-materials/orb-models
-  doi: 
+  doi: https://arxiv.org/abs/2410.22570
   date: 2024-10-15
   prediction: EFS
   nvt: true
@@ -185,7 +185,7 @@ MACE-MPA:
     - homonuclear-diatomics
     - stability
   github: https://github.com/ACEsuit/mace
-  doi: 
+  doi: https://arxiv.org/abs/2401.00096
   date: 2024-12-09
   prediction: EFS
   nvt: true

mlip_arena/tasks/registry.yaml

@@ -4,18 +4,18 @@ Homonuclear diatomics:
   task-layout: wide
   rank-page: homonuclear-diatomics
   last-update: 2024-09-19
-Thermal conductivity:
-  category: Material Properties
-  task-page: thermal-conductivity
+Combustion:
+  category: Molecular Dynamics
+  task-page: combustion
   task-layout: centered
-  rank-page: thermal-conductivity
+  rank-page: combustion
 High pressure stability:
   category: Molecular Dynamics
   task-page: stability
   task-layout: centered
   rank-page: 
-Combustion:
-  category: Molecular Dynamics
-  task-page: combustion
+Lattice thermal conductivity:
+  category: Properties and Physical Behaviors
+  task-page: thermal-conductivity
   task-layout: centered
-  rank-page: combustion
+  rank-page: thermal-conductivity

pyproject.toml

@@ -42,7 +42,7 @@ dependencies=[
 
 [project.optional-dependencies]
 app = [
-    "streamlit==1.38.0",
+    "streamlit==1.43.2",
     "plotly",
 ]
 test = [
@@ -64,7 +64,7 @@ test = [
     "pytest-xdist",
     "prefect==3.1.11",
     "pymatgen>=2025.1.9",
-    "streamlit==1.38.0"
+    "streamlit==1.43.2"
 ]
 mace = [
     "mace-torch==0.3.9",

requirements.txt

@@ -1,4 +1,4 @@
-streamlit>=1.36.0
+streamlit>=1.43.0
 plotly
 numpy
 scipy

serve/app.py

@@ -39,10 +39,7 @@ for task in TASKS:
     else:
         centered_pages.append(page)
 
-# mbd = st.page_link(page="https://matbench-discovery.materialsproject.org/", label="Matbench Discovery", icon=":material/extension:")
-# nav["Other benchmarks"].append(mbd)
-
-pg = st.navigation(nav)
+pg = st.navigation(nav, expanded=True)
 
 if pg in centered_pages:
     st.set_page_config(
@@ -67,18 +64,22 @@ else:
         },
     )
 
+
 st.toast(
     "MLIP Arena is currently in **pre-alpha**. The results are not stable. Please interpret them with care. Contributions are welcome. For more information, visit https://github.com/atomind-ai/mlip-arena.",
     icon="🍞",
 )
 
-# st.sidebar.markdown(
-# """
-# Other benchmarks
-# """
-# )
-# st.sidebar.page_link(
-#     "https://matbench-discovery.materialsproject.org/", label="Matbench Discovery", icon=":material/extension:"
-# )
+st.sidebar.markdown(
+"""
+Complementary Benchmarks
+"""
+)
+st.sidebar.page_link(
+    "https://matbench-discovery.materialsproject.org/", label="Matbench Discovery", icon=":material/link:"
+)
+st.sidebar.page_link(
+    "https://openkim.org/", label="OpenKIM", icon=":material/link:"
+)
 
 pg.run()

serve/ranks/homonuclear-diatomics.py

@@ -159,17 +159,21 @@ def render():
     )
     with st.expander("Explanation", icon=":material/info:"):
         st.caption(
-            """
+            r"""
             - **Conservation deviation**: The average deviation of force from negative energy gradient along the diatomic curves. 
             
             $$
-            \\text{Conservation deviation} = \\left\\langle\\left| \\mathbf{F}(\\mathbf{r})\\cdot\\frac{\\mathbf{r}}{\\|\\mathbf{r}\\|} +  \\nabla_rE\\right|\\right\\rangle_{r = \\|\\mathbf{r}\\|}
+            \text{Conservation deviation} = \left\langle\left| \mathbf{F}(\mathbf{r})\cdot\frac{\mathbf{r}}{\|\mathbf{r}\|} +  \nabla_rE\right|\right\rangle_{r = \|\mathbf{r}\|}
             $$
 
-            - **Spearman's coeff. (E: repulsion)**: Spearman's correlation coefficient of energy prediction within equilibrium distance $r \\in (r_{min}, r_o = \\argmin_{r} E(r))$.
-            - **Spearman's coeff. (F: descending)**: Spearman's correlation coefficient of force prediction within equilibrium distance $r \\in (r_{min}, r_o = \\argmin_{r} E(r))$.
+            - **Spearman's coeff. (E: repulsion)**: Spearman's correlation coefficient of energy prediction within equilibrium distance $r \in (r_{min}, r_o = \argmin_{r} E(r))$.
+            - **Spearman's coeff. (F: descending)**: Spearman's correlation coefficient of force prediction before maximum attraction $r \in (r_{min}, r_a = \argmin_{r} F(r))$.
             - **Tortuosity**: The ratio between total variation in energy and sum of absolute energy differences between $r_{min}$, $r_o$, and $r_{max}$.
-            - **Energy jump**: The sum of energy discontinuity.
+            - **Energy jump**: The sum of energy discontinuity between sampled points. 
+
+            $$
+            \text{Energy jump} = \sum_{r_i \in [r_\text{min}, r_\text{max}]} \left| \text{sign}{\left[ E(r_{i+1}) - E(r_i)\right]} - \text{sign}{\left[E(r_i) - E(r_{i-1})\right]}\right| \times \\ \left( \left|E(r_{i+1}) - E(r_i)\right| + \left|E(r_i) - E(r_{i-1})\right|\right)
+            $$
             - **Force flips**: The number of force direction changes.
             """
         )