Export Predictions

An example script to export model predictions using asap has been defined in tutorials/export_predictions.py.

asap.export_predictions(experiment_name, model, eval_dataset, logs_dir, out_dir, batch_size=64, use_map=False)

Export a pre-trained model's predictions as bigwig. One file will be generated for each chromosome in the eval_dataset.

Args:

• experiment_name (str): The name of the experiment. This will be used to load model checkpoints.
• model (str): The model name to evaluate. Choose from [cnn, lstm, dcnn, convnext_cnn, convnext_lstm, convnext_dcnn, convnext_transformer].
• eval_dataset (asap.dataloader.WGDataset): The whole-genome dataset corresponding to which predictions will be generated.
• logs_dir (str): The directory to load model checkpoints from.
• out_dir (str): The directory to save bigwigs.
• batch_size (int): The batch size for prediction.
• use_map (bool): If mappability information was used during training.

Returns:

• None