torch-molecule 0.1.0.dev2__tar.gz → 0.1.2__tar.gz

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: torch_molecule
-Version: 0.1.0.dev2
+Version: 0.1.2
 Summary: Deep learning packages for molecular discovery with a simple sklearn-style interface
 Author-email: Gang Liu <gliu7@nd.edu>
 License: MIT
@@ -16,14 +16,11 @@ License-File: LICENSE
 Requires-Dist: torch>=2.2.0
 Requires-Dist: torch-geometric>=2.6.1
 Requires-Dist: numpy
-Requires-Dist: pandas>=2.2.3
-Requires-Dist: click
 Requires-Dist: huggingface_hub>=0.22.2
 Requires-Dist: joblib>=1.3.2
 Requires-Dist: networkx>=3.2.1
-Requires-Dist: PyYAML>=6.0.2
 Requires-Dist: rdkit>=2023.9.5
-Requires-Dist: scikit_learn>=1.4.1.post1
+Requires-Dist: scikit_learn>=1.2.2
 Requires-Dist: scipy>=1.14.1
 Requires-Dist: tqdm>=4.66.2
 Requires-Dist: optuna>=4.0.0
@@ -35,7 +32,7 @@ Requires-Dist: flake8; extra == "dev"
 Dynamic: license-file
 
 <p align="center">
-  <img src="assets/logo.png" alt="torch-molecule logo" width="600"/>
+  <img src="https://github.com/liugangcode/torch-molecule/blob/main/assets/logo.png?raw=true" alt="torch-molecule logo" width="600"/>
 </p>
 
 <p align="center">
@@ -53,9 +50,16 @@ Dynamic: license-file
 
 ---
 
-`torch-molecule` is a package under active development that facilitates molecular discovery through deep learning, featuring a user-friendly, `sklearn`-style interface. It includes model checkpoints for efficient deployment and benchmarking across a range of molecular tasks. Currently, the package focuses on three main components: **Predictive Models**, **Generative Models**, and **Representation Models**. See the [List of Supported Models](#list-of-supported-models) section for all available models.
+`torch-molecule` is a package that facilitates molecular discovery through deep learning, featuring a user-friendly, `sklearn`-style interface. It includes model checkpoints for efficient deployment and benchmarking across a range of molecular tasks. Currently, the package focuses on three main components: **Predictive Models**, **Generative Models**, and **Representation Models**. See the [List of Supported Models](#list-of-supported-models) section for all available models.
+
+### API Comparison
+
+| Functionality | scikit-learn | torch-molecule |
+|---------------|-------------|----------------|
+| Property Prediction | `predictor.fit/predict(...)` | `predictor.fit/autofit/predict(...)` |
+| Representation Learning | Not supported | `encoder.fit/encode(...)` |
+| Molecular Generation | Not supported | `generator.fit/generate(...)` |
 
-> **Note**: This project is in active development, and features may change.
 
 ## Installation
 
@@ -65,26 +69,25 @@ Dynamic: license-file
    conda activate torch_molecule
    ```
 
-2. **Install `torch_molecule` from GitHub**:
+2. **Install using pip (0.1.2)**:
 
-   Clone the repository:
    ```bash
-   git clone https://github.com/liugangcode/torch-molecule
+   pip install torch-molecule
    ```
 
-   Install the requirements:
-   ```bash
-   pip install -r requirements.txt
-   ```
+3. **Install from source for the latest version**:
+
+   Clone the repository:
 
-   Editable install:
    ```bash
-   pip install -e .
+   git clone https://github.com/liugangcode/torch-molecule
+   cd torch-molecule
    ```
 
-3. **Install `torch_molecule` from PyPI** (Legacy):
+   Install:
+
    ```bash
-   pip install -i https://test.pypi.org/simple/ torch-molecule
+   pip install .
    ```
 
 ### Additional Packages
@@ -95,6 +98,12 @@ Dynamic: license-file
 | BFGNNMolecularPredictor | torch-scatter |
 | GRINMolecularPredictor | torch-scatter |
 
+**For models that require `torch-scatter`: Install using the following command: `pip install torch-scatter -f https://data.pyg.org/whl/torch-${TORCH}+${CUDA}.html`, e.g.,
+
+> `pip install torch-scatter -f https://data.pyg.org/whl/torch-2.7.1+cu128.html`
+
+**For models that require `transformers`:** `pip install transformers`
+
 ## Usage
 
 Refer to the `tests` folder for more use cases.
@@ -236,8 +245,8 @@ See the structure of `torch_molecule` with the command `tree -L 2 torch_molecule
 2. **Generative Models**: Done: Graph DiT, GraphGA, DiGress, GDS, MolGPT TODO: more
 3. **Representation Models**: Done: MoAMa, AttrMasking, ContextPred, EdgePred. Many pretrained models from HF. TODO: checkpoints, more 
 
-## Acknowledgements
+> **Note**: This project is in active development, and features may change.
 
-This project is under active development, and some features may change over time.
+## Acknowledgements
 
 The project template was adapted from [https://github.com/lwaekfjlk/python-project-template](https://github.com/lwaekfjlk/python-project-template). We thank the authors for their contribution to the open-source community.

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/README.md

@@ -1,5 +1,5 @@
 <p align="center">
-  <img src="assets/logo.png" alt="torch-molecule logo" width="600"/>
+  <img src="https://github.com/liugangcode/torch-molecule/blob/main/assets/logo.png?raw=true" alt="torch-molecule logo" width="600"/>
 </p>
 
 <p align="center">
@@ -17,9 +17,16 @@
 
 ---
 
-`torch-molecule` is a package under active development that facilitates molecular discovery through deep learning, featuring a user-friendly, `sklearn`-style interface. It includes model checkpoints for efficient deployment and benchmarking across a range of molecular tasks. Currently, the package focuses on three main components: **Predictive Models**, **Generative Models**, and **Representation Models**. See the [List of Supported Models](#list-of-supported-models) section for all available models.
+`torch-molecule` is a package that facilitates molecular discovery through deep learning, featuring a user-friendly, `sklearn`-style interface. It includes model checkpoints for efficient deployment and benchmarking across a range of molecular tasks. Currently, the package focuses on three main components: **Predictive Models**, **Generative Models**, and **Representation Models**. See the [List of Supported Models](#list-of-supported-models) section for all available models.
+
+### API Comparison
+
+| Functionality | scikit-learn | torch-molecule |
+|---------------|-------------|----------------|
+| Property Prediction | `predictor.fit/predict(...)` | `predictor.fit/autofit/predict(...)` |
+| Representation Learning | Not supported | `encoder.fit/encode(...)` |
+| Molecular Generation | Not supported | `generator.fit/generate(...)` |
 
-> **Note**: This project is in active development, and features may change.
 
 ## Installation
 
@@ -29,26 +36,25 @@
    conda activate torch_molecule
    ```
 
-2. **Install `torch_molecule` from GitHub**:
+2. **Install using pip (0.1.2)**:
 
-   Clone the repository:
    ```bash
-   git clone https://github.com/liugangcode/torch-molecule
+   pip install torch-molecule
    ```
 
-   Install the requirements:
-   ```bash
-   pip install -r requirements.txt
-   ```
+3. **Install from source for the latest version**:
+
+   Clone the repository:
 
-   Editable install:
    ```bash
-   pip install -e .
+   git clone https://github.com/liugangcode/torch-molecule
+   cd torch-molecule
    ```
 
-3. **Install `torch_molecule` from PyPI** (Legacy):
+   Install:
+
    ```bash
-   pip install -i https://test.pypi.org/simple/ torch-molecule
+   pip install .
    ```
 
 ### Additional Packages
@@ -59,6 +65,12 @@
 | BFGNNMolecularPredictor | torch-scatter |
 | GRINMolecularPredictor | torch-scatter |
 
+**For models that require `torch-scatter`: Install using the following command: `pip install torch-scatter -f https://data.pyg.org/whl/torch-${TORCH}+${CUDA}.html`, e.g.,
+
+> `pip install torch-scatter -f https://data.pyg.org/whl/torch-2.7.1+cu128.html`
+
+**For models that require `transformers`:** `pip install transformers`
+
 ## Usage
 
 Refer to the `tests` folder for more use cases.
@@ -200,8 +212,8 @@ See the structure of `torch_molecule` with the command `tree -L 2 torch_molecule
 2. **Generative Models**: Done: Graph DiT, GraphGA, DiGress, GDS, MolGPT TODO: more
 3. **Representation Models**: Done: MoAMa, AttrMasking, ContextPred, EdgePred. Many pretrained models from HF. TODO: checkpoints, more 
 
-## Acknowledgements
+> **Note**: This project is in active development, and features may change.
 
-This project is under active development, and some features may change over time.
+## Acknowledgements
 
 The project template was adapted from [https://github.com/lwaekfjlk/python-project-template](https://github.com/lwaekfjlk/python-project-template). We thank the authors for their contribution to the open-source community.

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "torch_molecule"
-version = "0.1.0.dev2"
+version = "0.1.2"
 description = "Deep learning packages for molecular discovery with a simple sklearn-style interface"
 authors = [{name = "Gang Liu", email = "gliu7@nd.edu"}]
 readme = "README.md"
@@ -19,14 +19,11 @@ dependencies = [
     "torch>=2.2.0",
     "torch-geometric>=2.6.1",
     "numpy",
-    "pandas>=2.2.3",
-    "click",
     "huggingface_hub>=0.22.2",
     "joblib>=1.3.2",
     "networkx>=3.2.1",
-    "PyYAML>=6.0.2",
     "rdkit>=2023.9.5",
-    "scikit_learn>=1.4.1.post1",
+    "scikit_learn>=1.2.2",
     "scipy>=1.14.1",
     "tqdm>=4.66.2",
     "optuna>=4.0.0",

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.1.0"
+__version__ = "0.1.2"
 
 """
 predictor module

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule/predictor/gnn/modeling_gnn.py

@@ -317,7 +317,7 @@ class GNNMolecularPredictor(BaseMolecularPredictor):
         y_val: Optional[Union[List, np.ndarray]] = None,
         X_unlbl: Optional[List[str]] = None,
         search_parameters: Optional[Dict[str, ParameterSpec]] = None,
-        n_trials: int = 10,
+        n_trials: int = 10, 
     ) -> "GNNMolecularPredictor":
         """Automatically find the best hyperparameters using Optuna optimization."""
         import optuna
@@ -565,19 +565,17 @@ class GNNMolecularPredictor(BaseMolecularPredictor):
                 best_eval = current_eval
                 best_state_dict = self.model.state_dict()
                 cnt_wait = 0
+                if self.verbose:
+                    print(
+                        f"Better result found at Epoch {epoch}: Loss = {np.mean(train_losses):.4f}, "
+                        f"{self.evaluate_name} = {best_eval:.4f}"
+                    )
             else:
                 cnt_wait += 1
                 if cnt_wait > self.patience:
                     if self.verbose:
                         print(f"Early stopping triggered after {epoch} epochs")
                     break
-            
-            if self.verbose and epoch % 10 == 0:
-                print(
-                    f"Epoch {epoch}: Loss = {np.mean(train_losses):.4f}, "
-                    f"{self.evaluate_name} = {current_eval:.4f}, "
-                    f"Best {self.evaluate_name} = {best_eval:.4f}"
-                )
 
         # Restore best model
         if best_state_dict is not None:
@@ -621,7 +619,11 @@ class GNNMolecularPredictor(BaseMolecularPredictor):
         self.model.eval()
         predictions = []
         with torch.no_grad():
-            for batch in tqdm(loader, disable=not self.verbose):
+            if self.verbose:
+                iterator = tqdm(loader, desc="Predicting")
+            else:
+                iterator = loader
+            for batch in iterator:
                 batch = batch.to(self.device)
                 out = self.model(batch)
                 predictions.append(out["prediction"].cpu().numpy())

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule/predictor/lstm/model.py

@@ -60,8 +60,7 @@ class LSTM(nn.Module):
         self.lstm2 = nn.LSTM(input_size=LSTMunits * 2, hidden_size=LSTMunits, bidirectional=True, batch_first=True)
         self.timedist_dense = nn.Linear(LSTMunits * 2, hidden_dim)
         self.relu = nn.ReLU()
-        self.fc = nn.Linear(hidden_dim * max_input_len, 1)
-
+        self.fc = nn.Linear(hidden_dim * max_input_len, num_task)
 
     def initialize_parameters(self, seed=None):
         """Initialize model parameters randomly.

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule/predictor/lstm/modeling_lstm.py

@@ -575,7 +575,6 @@ class LSTMMolecularPredictor(BaseMolecularPredictor):
                 y_true_list.append(batched_label.cpu().numpy())
         
         y_pred = np.concatenate(y_pred_list, axis=0)
-        y_pred = y_pred.reshape(-1, 1)  
         y_true = np.concatenate(y_true_list, axis=0)
         
         # Compute metric
@@ -606,7 +605,7 @@ class LSTMMolecularPredictor(BaseMolecularPredictor):
         for batch in iterator:
             batched_input, batched_label = batch
             batched_input = batched_input.to(self.device)
-            batched_label = batched_label.to(self.device).view(-1, 1)
+            batched_label = batched_label.to(self.device)
             optimizer.zero_grad()
 
             # Forward pass and loss computation

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule/predictor/smiles_transformer/modeling_transformer.py

@@ -240,7 +240,6 @@ class SMILESTransformerMolecularPredictor(LSTMMolecularPredictor):
                 mode='min' if not self.evaluate_higher_better else 'max',
                 factor=self.scheduler_factor,
                 patience=self.scheduler_patience,
-                verbose=self.verbose
             )
         
         return optimizer, scheduler

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule/predictor/ssr/model.py

@@ -193,8 +193,6 @@ class SSR(nn.Module):
         
         return new_batch
 
-
-
     def initialize_parameters(self, seed=None):
         """
         Randomly initialize all model parameters using the init_weights function.

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: torch_molecule
-Version: 0.1.0.dev2
+Version: 0.1.2
 Summary: Deep learning packages for molecular discovery with a simple sklearn-style interface
 Author-email: Gang Liu <gliu7@nd.edu>
 License: MIT
@@ -16,14 +16,11 @@ License-File: LICENSE
 Requires-Dist: torch>=2.2.0
 Requires-Dist: torch-geometric>=2.6.1
 Requires-Dist: numpy
-Requires-Dist: pandas>=2.2.3
-Requires-Dist: click
 Requires-Dist: huggingface_hub>=0.22.2
 Requires-Dist: joblib>=1.3.2
 Requires-Dist: networkx>=3.2.1
-Requires-Dist: PyYAML>=6.0.2
 Requires-Dist: rdkit>=2023.9.5
-Requires-Dist: scikit_learn>=1.4.1.post1
+Requires-Dist: scikit_learn>=1.2.2
 Requires-Dist: scipy>=1.14.1
 Requires-Dist: tqdm>=4.66.2
 Requires-Dist: optuna>=4.0.0
@@ -35,7 +32,7 @@ Requires-Dist: flake8; extra == "dev"
 Dynamic: license-file
 
 <p align="center">
-  <img src="assets/logo.png" alt="torch-molecule logo" width="600"/>
+  <img src="https://github.com/liugangcode/torch-molecule/blob/main/assets/logo.png?raw=true" alt="torch-molecule logo" width="600"/>
 </p>
 
 <p align="center">
@@ -53,9 +50,16 @@ Dynamic: license-file
 
 ---
 
-`torch-molecule` is a package under active development that facilitates molecular discovery through deep learning, featuring a user-friendly, `sklearn`-style interface. It includes model checkpoints for efficient deployment and benchmarking across a range of molecular tasks. Currently, the package focuses on three main components: **Predictive Models**, **Generative Models**, and **Representation Models**. See the [List of Supported Models](#list-of-supported-models) section for all available models.
+`torch-molecule` is a package that facilitates molecular discovery through deep learning, featuring a user-friendly, `sklearn`-style interface. It includes model checkpoints for efficient deployment and benchmarking across a range of molecular tasks. Currently, the package focuses on three main components: **Predictive Models**, **Generative Models**, and **Representation Models**. See the [List of Supported Models](#list-of-supported-models) section for all available models.
+
+### API Comparison
+
+| Functionality | scikit-learn | torch-molecule |
+|---------------|-------------|----------------|
+| Property Prediction | `predictor.fit/predict(...)` | `predictor.fit/autofit/predict(...)` |
+| Representation Learning | Not supported | `encoder.fit/encode(...)` |
+| Molecular Generation | Not supported | `generator.fit/generate(...)` |
 
-> **Note**: This project is in active development, and features may change.
 
 ## Installation
 
@@ -65,26 +69,25 @@ Dynamic: license-file
    conda activate torch_molecule
    ```
 
-2. **Install `torch_molecule` from GitHub**:
+2. **Install using pip (0.1.2)**:
 
-   Clone the repository:
    ```bash
-   git clone https://github.com/liugangcode/torch-molecule
+   pip install torch-molecule
    ```
 
-   Install the requirements:
-   ```bash
-   pip install -r requirements.txt
-   ```
+3. **Install from source for the latest version**:
+
+   Clone the repository:
 
-   Editable install:
    ```bash
-   pip install -e .
+   git clone https://github.com/liugangcode/torch-molecule
+   cd torch-molecule
    ```
 
-3. **Install `torch_molecule` from PyPI** (Legacy):
+   Install:
+
    ```bash
-   pip install -i https://test.pypi.org/simple/ torch-molecule
+   pip install .
    ```
 
 ### Additional Packages
@@ -95,6 +98,12 @@ Dynamic: license-file
 | BFGNNMolecularPredictor | torch-scatter |
 | GRINMolecularPredictor | torch-scatter |
 
+**For models that require `torch-scatter`: Install using the following command: `pip install torch-scatter -f https://data.pyg.org/whl/torch-${TORCH}+${CUDA}.html`, e.g.,
+
+> `pip install torch-scatter -f https://data.pyg.org/whl/torch-2.7.1+cu128.html`
+
+**For models that require `transformers`:** `pip install transformers`
+
 ## Usage
 
 Refer to the `tests` folder for more use cases.
@@ -236,8 +245,8 @@ See the structure of `torch_molecule` with the command `tree -L 2 torch_molecule
 2. **Generative Models**: Done: Graph DiT, GraphGA, DiGress, GDS, MolGPT TODO: more
 3. **Representation Models**: Done: MoAMa, AttrMasking, ContextPred, EdgePred. Many pretrained models from HF. TODO: checkpoints, more 
 
-## Acknowledgements
+> **Note**: This project is in active development, and features may change.
 
-This project is under active development, and some features may change over time.
+## Acknowledgements
 
 The project template was adapted from [https://github.com/lwaekfjlk/python-project-template](https://github.com/lwaekfjlk/python-project-template). We thank the authors for their contribution to the open-source community.

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule.egg-info/SOURCES.txt

@@ -1,7 +1,6 @@
 LICENSE
 README.md
 pyproject.toml
-tests/test_io.py
 torch_molecule/__init__.py
 torch_molecule.egg-info/PKG-INFO
 torch_molecule.egg-info/SOURCES.txt

{torch_molecule-0.1.0.dev2 → torch_molecule-0.1.2}/torch_molecule.egg-info/requires.txt

@@ -1,14 +1,11 @@
 torch>=2.2.0
 torch-geometric>=2.6.1
 numpy
-pandas>=2.2.3
-click
 huggingface_hub>=0.22.2
 joblib>=1.3.2
 networkx>=3.2.1
-PyYAML>=6.0.2
 rdkit>=2023.9.5
-scikit_learn>=1.4.1.post1
+scikit_learn>=1.2.2
 scipy>=1.14.1
 tqdm>=4.66.2
 optuna>=4.0.0