Update M3GNet potential training notebook for the demonstrating of ob…

…taining and using element offset (#288) * improve TensorNet model coverage * Update pyproject.toml Signed-off-by: Tsz Wai Ko <[email protected]> * Improve the unit test for SO(3) equivarance in TensorNet class * improve SO3Net model class coverage and simplify TensorNet implementations * improve the coverage in MLP_norm class * Improve the implementation of three-body interactions * fixed black * Optimize the speed of _compute_3body class * type checking is added for scheduler * update M3GNet Potential training notebook for the demonstration of obtaining and using element offsets * Downgrade sympy to avoid crash of SO3 operations --------- Signed-off-by: Tsz Wai Ko <[email protected]>
materialsvirtuallab · Jul 14, 2024 · 6006ab4 · 6006ab4
1 parent f3e2b49
commit 6006ab4
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 1 deletion.
diff --git a/examples/Training a M3GNet Potential with PyTorch Lightning.ipynb b/examples/Training a M3GNet Potential with PyTorch Lightning.ipynb
@@ -269,7 +269,10 @@
     "# download a pre-trained M3GNet\n",
     "m3gnet_nnp = matgl.load_model(\"M3GNet-MP-2021.2.8-PES\")\n",
     "model_pretrained = m3gnet_nnp.model\n",
-    "lit_module_finetune = PotentialLightningModule(model=model_pretrained, lr=1e-4, include_line_graph=True)"
+    "# obtain element energy offset\n",
+    "property_offset = m3gnet_nnp.element_refs.property_offset\n",
+    "# you should test whether including the original property_offset helps improve training and validation accuracy\n",
+    "lit_module_finetune = PotentialLightningModule(model=model_pretrained, element_refs=property_offset, lr=1e-4, include_line_graph=True)"
    ]
   },
   {

diff --git a/requirements.txt b/requirements.txt
@@ -6,3 +6,4 @@ ase==3.23.0
 pydantic==2.7.1
 boto3==1.34.101
 numpy==1.26.4
+sympy==1.12.1