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

add testing to m3gnet potential training example #179

Merged
merged 1 commit into from
Oct 12, 2023
Merged

add testing to m3gnet potential training example #179

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 21 additions & 1 deletion docs/tutorials/Training a M3GNet Potential with PyTorch Lightning.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,7 +100,7 @@ Finally, we will initialize the Pytorch Lightning trainer and run the fitting. H
# If you wish to disable GPU or MPS (M1 mac) training, use the accelerator="cpu" kwarg.
logger = CSVLogger("logs", name="M3GNet_training")
trainer = pl.Trainer(max_epochs=1, accelerator="cpu", logger=logger)
trainer.fit(model=lit_module, train_dataloaders=train_loader, val_dataloaders=val_loader)
trainer.fit(model=lit_module, train_dataloaders=train_loader, val_dataloaders=val_loader, inference_mode=False)
```

GPU available: False, used: False
Expand All @@ -127,7 +127,27 @@ trainer.fit(model=lit_module, train_dataloaders=train_loader, val_dataloaders=va

Epoch 0: 100%|██████████| 163/163 [00:40<00:00, 3.99it/s, v_num=1, val_Total_Loss=0.815, val_Energy_MAE=0.674, val_Force_MAE=0.289, val_Stress_MAE=0.000, val_Site_Wise_MAE=0.000, val_Energy_RMSE=0.730, val_Force_RMSE=0.419, val_Stress_RMSE=0.000, val_Site_Wise_RMSE=0.000, train_Total_Loss=21.60, train_Energy_MAE=2.500, train_Force_MAE=0.349, train_Stress_MAE=0.000, train_Site_Wise_MAE=0.000, train_Energy_RMSE=2.660, train_Force_RMSE=0.487, train_Stress_RMSE=0.000, train_Site_Wise_RMSE=0.000]

```python
# test the model, remember to set inference_mode=False in trainer (see above)
trainer.test(dataloaders=test_loader)
```

Restoring states from the checkpoint path at logs/M3GNet_training/version_0/checkpoints/epoch=0-step=163.ckpt
Loaded model weights from the checkpoint at logs/M3GNet_training/version_0/checkpoints/epoch=0-step=163.ckpt
Testing DataLoader 0: 100%|██████████| 21/21 [00:04<00:00, 4.25it/s]
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
┃ Test metric ┃ DataLoader 0 ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
│ test_Energy_MAE │ 0.39233624935150146 │
│ test_Energy_RMSE │ 0.47101688385009766 │
│ test_Force_MAE │ 0.2673814296722412 │
│ test_Force_RMSE │ 0.3861512243747711 │
│ test_Site_Wise_MAE │ 0.0 │
│ test_Site_Wise_RMSE │ 0.0 │
│ test_Stress_MAE │ 0.0 │
│ test_Stress_RMSE │ 0.0 │
│ test_Total_Loss │ 0.754666268825531 │
└───────────────────────────┴───────────────────────────┘

```python
# save trained model
Expand Down
136 changes: 26 additions & 110 deletions examples/Training a M3GNet Potential with PyTorch Lightning.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,8 @@
"execution_count": null,
"id": "6355190a",
"metadata": {
"id": "6355190a"
"id": "6355190a",
"tags": []
},
"outputs": [],
"source": [
Expand Down Expand Up @@ -76,24 +77,10 @@
]
},
"id": "bd0ce8a2-ec68-4160-9457-823fb9e6a35d",
"outputId": "2252a59c-9a70-4673-926f-9ed8fc69ed0d"
"outputId": "2252a59c-9a70-4673-926f-9ed8fc69ed0d",
"tags": []
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Retrieving ThermoDoc documents: 100%|██████████| 407/407 [00:00<00:00, 4962446.88it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"407 downloaded from MP.\n"
]
}
],
"outputs": [],
"source": [
"# Obtain your API key here: https://next-gen.materialsproject.org/api\n",
"mpr = MPRester(api_key=\"YOUR_API_KEY\")\n",
Expand Down Expand Up @@ -131,7 +118,8 @@
"base_uri": "https://localhost:8080/"
},
"id": "e9dc84cb",
"outputId": "b9f93f24-0fd6-4737-a8e4-e87804cd3ad2"
"outputId": "b9f93f24-0fd6-4737-a8e4-e87804cd3ad2",
"tags": []
},
"outputs": [],
"source": [
Expand Down Expand Up @@ -219,52 +207,10 @@
]
},
"id": "7472d071",
"outputId": "9d10c152-752f-4afc-8759-c5174ea446b9"
"outputId": "9d10c152-752f-4afc-8759-c5174ea446b9",
"tags": []
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: False, used: False\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"\n",
" | Name | Type | Params\n",
"--------------------------------------------\n",
"0 | mae | MeanAbsoluteError | 0 \n",
"1 | rmse | MeanSquaredError | 0 \n",
"2 | model | Potential | 282 K \n",
"--------------------------------------------\n",
"282 K Trainable params\n",
"0 Non-trainable params\n",
"282 K Total params\n",
"1.130 Total estimated model params size (MB)\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 0: 100%|██████████| 163/163 [00:39<00:00, 4.16it/s, v_num=1, val_Total_Loss=0.815, val_Energy_MAE=0.674, val_Force_MAE=0.289, val_Stress_MAE=0.000, val_Site_Wise_MAE=0.000, val_Energy_RMSE=0.730, val_Force_RMSE=0.419, val_Stress_RMSE=0.000, val_Site_Wise_RMSE=0.000, train_Total_Loss=21.60, train_Energy_MAE=2.500, train_Force_MAE=0.349, train_Stress_MAE=0.000, train_Site_Wise_MAE=0.000, train_Energy_RMSE=2.660, train_Force_RMSE=0.487, train_Stress_RMSE=0.000, train_Site_Wise_RMSE=0.000]"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"`Trainer.fit` stopped: `max_epochs=1` reached.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 0: 100%|██████████| 163/163 [00:40<00:00, 3.99it/s, v_num=1, val_Total_Loss=0.815, val_Energy_MAE=0.674, val_Force_MAE=0.289, val_Stress_MAE=0.000, val_Site_Wise_MAE=0.000, val_Energy_RMSE=0.730, val_Force_RMSE=0.419, val_Stress_RMSE=0.000, val_Site_Wise_RMSE=0.000, train_Total_Loss=21.60, train_Energy_MAE=2.500, train_Force_MAE=0.349, train_Stress_MAE=0.000, train_Site_Wise_MAE=0.000, train_Energy_RMSE=2.660, train_Force_RMSE=0.487, train_Stress_RMSE=0.000, train_Site_Wise_RMSE=0.000]\n"
]
}
],
"outputs": [],
"source": [
"# If you wish to disable GPU or MPS (M1 mac) training, use the accelerator=\"cpu\" kwarg.\n",
"logger = CSVLogger(\"logs\", name=\"M3GNet_training\")\n",
Expand All @@ -276,7 +222,20 @@
{
"cell_type": "code",
"execution_count": null,
"id": "e9f0202f",
"id": "a3d971e0-1065-4b7f-a03f-06a105ce0939",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"# test the model, remember to set inference_mode=False in trainer (see above)\n",
"trainer.test(dataloaders=test_loader)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8d118b9f-a3fe-4d50-b51e-5a6d1c6dccb7",
"metadata": {},
"outputs": [],
"source": [
Expand Down Expand Up @@ -359,50 +318,7 @@
"id": "4133225a-5990-4b97-9d73-88195df87a1a",
"outputId": "f149a68a-eef7-4726-b3a1-723525dc908f"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: False, used: False\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"\n",
" | Name | Type | Params\n",
"--------------------------------------------\n",
"0 | mae | MeanAbsoluteError | 0 \n",
"1 | rmse | MeanSquaredError | 0 \n",
"2 | model | Potential | 288 K \n",
"--------------------------------------------\n",
"288 K Trainable params\n",
"0 Non-trainable params\n",
"288 K Total params\n",
"1.153 Total estimated model params size (MB)\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 0: 100%|██████████| 163/163 [00:37<00:00, 4.34it/s, v_num=1, val_Total_Loss=5.420, val_Energy_MAE=1.070, val_Force_MAE=0.407, val_Stress_MAE=0.000, val_Site_Wise_MAE=0.000, val_Energy_RMSE=1.330, val_Force_RMSE=0.615, val_Stress_RMSE=0.000, val_Site_Wise_RMSE=0.000, train_Total_Loss=21.80, train_Energy_MAE=3.270, train_Force_MAE=0.572, train_Stress_MAE=0.000, train_Site_Wise_MAE=0.000, train_Energy_RMSE=3.430, train_Force_RMSE=0.870, train_Stress_RMSE=0.000, train_Site_Wise_RMSE=0.000]"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"`Trainer.fit` stopped: `max_epochs=1` reached.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 0: 100%|██████████| 163/163 [00:39<00:00, 4.17it/s, v_num=1, val_Total_Loss=5.420, val_Energy_MAE=1.070, val_Force_MAE=0.407, val_Stress_MAE=0.000, val_Site_Wise_MAE=0.000, val_Energy_RMSE=1.330, val_Force_RMSE=0.615, val_Stress_RMSE=0.000, val_Site_Wise_RMSE=0.000, train_Total_Loss=21.80, train_Energy_MAE=3.270, train_Force_MAE=0.572, train_Stress_MAE=0.000, train_Site_Wise_MAE=0.000, train_Energy_RMSE=3.430, train_Force_RMSE=0.870, train_Stress_RMSE=0.000, train_Site_Wise_RMSE=0.000]\n"
]
}
],
"outputs": [],
"source": [
"# If you wish to disable GPU or MPS (M1 mac) training, use the accelerator=\"cpu\" kwarg.\n",
"logger = CSVLogger(\"logs\", name=\"M3GNet_finetuning\")\n",
Expand Down Expand Up @@ -468,7 +384,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.9"
"version": "3.10.12"
}
},
"nbformat": 4,
Expand Down