seqgra.seqgrae module¶
MIT - CSAIL - Gifford Lab - seqgra
seqgra ensemble
@author: Konstantin Krismer
- change_ds_size(data_definition: seqgra.model.data.datadefinition.DataDefinition, ds_size: int, training_set_fraction: float = 0.7, validation_set_fraction: float = 0.1, test_set_fraction: float = 0.2) → seqgra.model.data.datadefinition.DataDefinition[source]¶
- create_parser(default_ds_sizes: List[int] = [10000, 20000, 40000, 80000, 160000, 320000, 640000, 1280000])[source]¶
- parse_data_definition_file(file_name: str) → seqgra.model.data.datadefinition.DataDefinition[source]¶
- parse_model_definition_file(file_name: str) → seqgra.model.model.modeldefinition.ModelDefinition[source]¶
- run_seqgra_ensemble(analysis_id: str, data_def_file: str, data_folder: str, model_def_files: List[str], output_dir: str, ds_sizes: List[float], d_seeds: List[int], m_seeds: List[int], seed_grid: bool, gpu_id: int) → None[source]¶
- subsample_data_set(data_folder: str, output_dir: str, subsampling_rate: float, seed: int) → None[source]¶
- subsample_experimental_data(data_folder: str, output_dir: str, subsampling_rates: List[float], d_seeds: List[int]) → None[source]¶
- write_analysis_script(analysis_id: str, data_file_names: Optional[List[List[str]]], data_folders: Optional[List[List[str]]], model_file_names: List[List[str]], output_dir: str, analyses_dir: str, seed_grid: bool, gpu_id: int) → None[source]¶
- write_data_definition_file(data_definition: seqgra.model.data.datadefinition.DataDefinition, output_dir: str, ds_size: int, seed: int) → None[source]¶
- write_data_definition_files(data_def_file: str, output_dir: str, ds_sizes: List[float], d_seeds: List[int]) → None[source]¶
- write_model_definition_file(model_definition: seqgra.model.model.modeldefinition.ModelDefinition, output_dir: str, seed: int) → None[source]¶