Update docs.

materialsvirtuallab · Aug 11, 2023 · 24eae9c · 24eae9c
1 parent 561ffd6
commit 24eae9c
Show file tree

Hide file tree

Showing 9 changed files with 1,168 additions and 2,972 deletions.
diff --git a/docs/changes.md b/docs/changes.md
@@ -6,6 +6,11 @@ nav_order: 3
 
 # Change Log
 
+## 0.8.2
+- Add site-wise predictions for Potential. (@lbluque)
+- Enable CLI tool to be used for multi-fidelity models. (@kenko911)
+- Minor fix for model version for DIRECT model.
+
 ## 0.8.1
 - Fixed bug with loading of models trained with GPUs.
 - Updated default model for relaxations to be the `M3GNet-MP-2021.2.8-DIRECT-PES model`.

diff --git a/docs/matgl.apps.md b/docs/matgl.apps.md
@@ -14,65 +14,29 @@ potentials parameterizing the potential energy surface (PES).
 
 Implementation of Interatomic Potentials.
 
+### *class* matgl.apps.pes.Potential(model: nn.Module, data_mean: torch.Tensor | None = None, data_std: torch.Tensor | None = None, element_refs: np.ndarray | None = None, calc_forces: bool = True, calc_stresses: bool = True, calc_hessian: bool = False, calc_site_wise: bool = False)
 
-### _class_ matgl.apps.pes.Potential(model: nn.Module, data_mean: torch.Tensor | None = None, data_std: torch.Tensor | None = None, element_refs: np.ndarray | None = None, calc_forces: bool = True, calc_stresses: bool = True, calc_hessian: bool = False)
 Bases: `Module`, [`IOMixIn`](matgl.utils.md#matgl.utils.io.IOMixIn)
 
 A class representing an interatomic potential.
 
 Initialize Potential from a model and elemental references.
 
-
-* **Parameters**
-
-
-    * **model** – Model for predicting energies.
-
-
-    * **data_mean** – Mean of target.
-
-
-    * **data_std** – Std dev of target.
-
-
-    * **element_refs** – Element reference values for each element.
-
-
-    * **calc_forces** – Enable force calculations.
-
-
-    * **calc_stresses** – Enable stress calculations.
-
-
-    * **calc_hessian** – Enable hessian calculations.
-
-
+* **Parameters:**
+  * **model** – Model for predicting energies.
+  * **data_mean** – Mean of target.
+  * **data_std** – Std dev of target.
+  * **element_refs** – Element reference values for each element.
+  * **calc_forces** – Enable force calculations.
+  * **calc_stresses** – Enable stress calculations.
+  * **calc_hessian** – Enable hessian calculations.
+  * **calc_site_wise** – Enable site-wise property calculation.
 
 #### forward(g: dgl.DGLGraph, state_attr: torch.Tensor | None = None, l_g: dgl.DGLGraph | None = None)
 
-* **Parameters**
-
-
-    * **g** – DGL graph
-
-
-    * **state_attr** – State attrs
-
-
-    * **l_g** – Line graph.
-
-
-
-* **Returns**
-
-    torch.Tensor
-
-
-
-* **Return type**
-
-    energies, forces, stresses, hessian
-
-
-
-#### training(_: boo_ )
+* **Parameters:**
+  * **g** – DGL graph
+  * **state_attr** – State attrs
+  * **l_g** – Line graph.
+* **Returns:**
+  (energies, forces, stresses, hessian) or (energies, forces, stresses, hessian, site-wise properties)
diff --git a/docs/matgl.data.md b/docs/matgl.data.md
@@ -13,141 +13,87 @@ This package implements data manipulation tools.
 
 Module implementing various data transformers for PyTorch.
 
+### *class* matgl.data.transformer.LogTransformer
 
-### _class_ matgl.data.transformer.LogTransformer()
-Bases: `Transformer`
+Bases: [`Transformer`](#matgl.data.transformer.Transformer)
 
 Performs a natural log of the data.
 
-
 #### inverse_transform(data)
-Invert the log (exp).
-
-
-* **Parameters**
-
-    **data** – Input data
-
-
-
-* **Returns**
-
-    exp(data)
 
+Invert the log (exp).
 
+* **Parameters:**
+  **data** – Input data
+* **Returns:**
+  exp(data)
 
 #### transform(data)
-Take the log of the data.
-
-
-* **Parameters**
-
-    **data** – Input data
-
-
 
-* **Returns**
-
-    Scaled data
+Take the log of the data.
 
+* **Parameters:**
+  **data** – Input data
+* **Returns:**
+  Scaled data
 
+### *class* matgl.data.transformer.Normalizer(mean: float, std: float)
 
-### _class_ matgl.data.transformer.Normalizer(mean: float, std: float)
-Bases: `Transformer`
+Bases: [`Transformer`](#matgl.data.transformer.Transformer)
 
 Performs a scaling of the data by centering to the mean and dividing by the standard deviation.
 
+* **Parameters:**
+  * **mean** – Mean of the data
+  * **std** – Standard deviation of the data.
 
-* **Parameters**
-
-
-    * **mean** – Mean of the data
-
-
-    * **std** – Standard deviation of the data.
-
-
+#### *classmethod* from_data(data)
 
-#### _classmethod_ from_data(data)
 Create Normalizer from data.
 
-
-* **Parameters**
-
-    **data** – Input data.
-
-
-
-* **Returns**
-
-    Normalizer
-
-
+* **Parameters:**
+  **data** – Input data.
+* **Returns:**
+  Normalizer
 
 #### inverse_transform(data)
-Invert the scaling.
-
-
-* **Parameters**
-
-    **data** – Scaled data
-
-
-
-* **Returns**
-
-    Unscaled data
 
+Invert the scaling.
 
+* **Parameters:**
+  **data** – Scaled data
+* **Returns:**
+  Unscaled data
 
 #### transform(data)
-z-score the data by subtracting the mean and dividing by the standard deviation.
-
-
-* **Parameters**
-
-    **data** – Input data
-
-
 
-* **Returns**
-
-    Scaled data
+z-score the data by subtracting the mean and dividing by the standard deviation.
 
+* **Parameters:**
+  **data** – Input data
+* **Returns:**
+  Scaled data
 
+### *class* matgl.data.transformer.Transformer
 
-### _class_ matgl.data.transformer.Transformer()
 Bases: `object`
 
 Abstract base class defining a data transformer.
 
+#### *abstract* inverse_transform(data: Tensor)
 
-#### _abstract_ inverse_transform(data: Tensor)
 Inverse transformation to be performed on data.
 
+* **Parameters:**
+  **data** – Input data
+* **Returns:**
+  Inverse-transformed data.
 
-* **Parameters**
-
-    **data** – Input data
-
-
-
-* **Returns**
-
-    Inverse-transformed data.
-
+#### *abstract* transform(data: Tensor)
 
-
-#### _abstract_ transform(data: Tensor)
 Transformation to be performed on data.
 
-
-* **Parameters**
-
-    **data** – Input data
-
-
-
-* **Returns**
-
-    Transformed data.
+* **Parameters:**
+  **data** – Input data
+* **Returns:**
+  Transformed data.