---
tags:
- torch_molecule
- molecular-property-prediction
library_name: torch_molecule
---

# ContextPredMolecularEncoder Model

## Model Description
- **Model Type**: ContextPredMolecularEncoder
- **Framework**: torch_molecule
- **Last Updated**: 2025-04-07

## Task Summary
| Task | Version | Last Updated | Parameters | Metrics |
|------|---------|--------------|------------|----------|
| default | 0.0.2 | 2025-04-07 | 3,034,505 |  |


## Usage

```python
from torch_molecule import ContextPredMolecularEncoder

# Load model for specific task
model = ContextPredMolecularEncoder()
model.load(
    "local_model_dir/ContextPredMolecularEncoder.pt",
    repo="Einae/ContextPred"
)

# For predictor: Make predictions
# predictions = model.predict(smiles_list)
# For generator: Make generations
# generations = model.generate(n_samples)
# For encoder: Make encodings
# encodings = model.encode(smiles_list)
```

## Tasks Details


### default Task
- **Current Version**: 0.0.2
- **Last Updated**: 2025-04-07
- **Parameters**: 3,034,505
- **Configuration**:
```python
{
  "mode": "cbow",
  "context_size": 2,
  "neg_samples": 1,
  "encoder_type": "gin-virtual",
  "readout": "sum",
  "num_layer": 3,
  "hidden_size": 300,
  "drop_ratio": 0.5,
  "norm_layer": "batch_norm",
  "batch_size": 128,
  "epochs": 100,
  "learning_rate": 0.001,
  "weight_decay": 0.0,
  "grad_clip_value": null,
  "use_lr_scheduler": false,
  "scheduler_factor": 0.5,
  "scheduler_patience": 5,
  "fitting_epoch": 99,
  "device": {
    "_type": "unknown",
    "repr": "cuda:0"
  },
  "verbose": false,
  "model_name": "ContextPredMolecularEncoder"
}
```