def _prepare_specs(self):
with self.yaml_path.open("r") as file:
specs = yaml.safe_load(file)
ParameterValidator.assert_keys_present(specs.keys(), ["definitions", "instructions"], GalaxyTrainMLModel.__name__, "YAML specification")
ParameterValidator.assert_all_in_valid_list(specs.keys(), ["definitions", "instructions", "output"], GalaxyTrainMLModel.__name__,
"YAML specification")
ParameterValidator.assert_type_and_value(specs["instructions"], dict, GalaxyTrainMLModel.__name__, "instructions")
assert len(list(specs["instructions"].keys())) == 1, f"{GalaxyTrainMLModel.__name__}: one instruction has to be specified under " \
f"`instructions`, got the following instead: {list(specs['instructions'].keys())}."
self.instruction_name = list(specs["instructions"].keys())[0]
ParameterValidator.assert_type_and_value(specs['instructions'][self.instruction_name], dict, GalaxyTrainMLModel.__name__,
self.instruction_name)
ParameterValidator.assert_keys_present(specs['instructions'][self.instruction_name].keys(), ['type'], GalaxyTrainMLModel.__name__,
self.instruction_name)
assert specs['instructions'][self.instruction_name]['type'] == TrainMLModelInstruction.__name__[:-11], \
f"{GalaxyTrainMLModel.__name__}: instruction `type` under {self.instruction_name} has to be {TrainMLModelInstruction.__name__[:-11]} " \
f"for this tool."
assert len(
specs['instructions'][self.instruction_name]['labels']) == 1, f"{GalaxyTrainMLModel.__name__}: one label has to be specified under " \
f"`labels`, got the following instead: {specs['instructions'][self.instruction_name]['labels']}."
Util.check_paths(specs, GalaxyTrainMLModel.__name__)
Util.update_result_paths(specs, self.result_path, self.yaml_path)
def _prepare_parameters(motif_filepath: str, match_v_genes: bool, sum_counts: bool, name: str = None):
ParameterValidator.assert_type_and_value(match_v_genes, bool, "MatchedRegexEncoder", "match_v_genes")
ParameterValidator.assert_type_and_value(sum_counts, bool, "MatchedRegexEncoder", "sum_counts")
motif_filepath = Path(motif_filepath)
assert motif_filepath.is_file(), f"MatchedRegexEncoder: the file {motif_filepath} does not exist. " \
f"Specify the correct path under motif_filepath."
file_columns = list(pd.read_csv(motif_filepath, sep="\t", iterator=False, dtype=str, nrows=0).columns)
ParameterValidator.assert_all_in_valid_list(file_columns, ["id"] + [f"{c.value}V" for c in Chain] + [f"{c.value}_regex" for c in Chain], "MatchedRegexEncoder", "motif_filepath (column names)")
chains = [colname.split("_")[0] for colname in file_columns if colname.endswith("_regex")]
if match_v_genes:
for chain in chains:
assert f"{chain}V" in file_columns, f"MatchedRegexEncoder: expected column {chain}V to be present in the columns of motif_filepath. " \
f"Remove {chain}_regex from columns, or set match_v_genes to False."
return {
"motif_filepath": motif_filepath,
"match_v_genes": match_v_genes,
"sum_counts": sum_counts,
"chains": chains,
"name": name
}
def _prepare_report_config(self, instruction_key, instruction, split_key,
symbol_table):
if "reports" in instruction[split_key] and len(
instruction[split_key]["reports"]) > 0:
location = f"{instruction_key}/{split_key}/reports"
report_types = list(signature(ReportConfig).parameters.keys())
ParameterValidator.assert_all_in_valid_list(
instruction[split_key]["reports"].keys(), report_types,
location, "reports")
for report_type in instruction[split_key]["reports"]:
ParameterValidator.assert_type_and_value(
instruction[split_key]["reports"][report_type], list,
f"{location}/{report_type}", report_type)
report_config_input = {
report_type: {
report_id: symbol_table.get(report_id)
for report_id in instruction[split_key]["reports"]
[report_type]
}
for report_type in instruction[split_key]["reports"]
}
else:
report_config_input = {}
return report_config_input
def import_dataset(params, name: str) -> SequenceDataset:
"""
Returns randomly generated receptor dataset according to the parameters;
YAML specification:
result_path: path/where/to/store/results/
sequence_count: 100 # number of random sequences to generate
chain_1_length_probabilities:
14: 0.8 # 80% of all generated sequences for all sequences will have length 14
15: 0.2 # 20% of all generated sequences across all sequences will have length 15
labels:
epitope1: # label name
True: 0.5 # 50% of the sequences will have class True
False: 0.5 # 50% of the sequences will have class False
epitope2: # next label with classes that will be assigned to sequences independently of the previous label or other parameters
1: 0.3 # 30% of the generated sequences will have class 1
0: 0.7 # 70% of the generated sequences will have class 0
"""
valid_keys = [
"sequence_count", "length_probabilities", "labels", "result_path"
]
ParameterValidator.assert_all_in_valid_list(
list(params.keys()), valid_keys, "RandomSequenceDatasetImport",
"params")
return RandomDatasetGenerator.generate_sequence_dataset(
sequence_count=params["sequence_count"],
length_probabilities=params["length_probabilities"],
labels=params["labels"],
path=params["result_path"])
def build_object(cls, **kwargs):
location = "TrainingPerformance"
valid_metrics = [m.name for m in Metric]
name = kwargs["name"] if "name" in kwargs else None
metrics = kwargs["metrics"] if "metrics" in kwargs else valid_metrics
metrics = [m.upper() for m in metrics]
ParameterValidator.assert_all_in_valid_list(metrics, valid_metrics, location, 'metrics')
return TrainingPerformance(set(metrics), name=name)
def import_dataset(params: dict, dataset_name: str) -> RepertoireDataset:
valid_keys = [
"result_path", "repertoire_count", "sequence_count_probabilities",
"sequence_length_probabilities", "labels"
]
ParameterValidator.assert_all_in_valid_list(
list(params.keys()), valid_keys, "RandomRepertoireDatasetImport",
"params")
return RandomDatasetGenerator.generate_repertoire_dataset(
repertoire_count=params["repertoire_count"],
sequence_count_probabilities=params[
"sequence_count_probabilities"],
sequence_length_probabilities=params[
"sequence_length_probabilities"],
labels=params["labels"],
path=params["result_path"])
def _update_specs(self):
with self.yaml_path.open('r') as file:
specs = yaml.safe_load(file)
ParameterValidator.assert_keys_present(specs.keys(),
["definitions", "instructions"],
DatasetGenerationTool.__name__,
"YAML specification")
ParameterValidator.assert_all_in_valid_list(
specs.keys(), ["definitions", "instructions", "output"],
DatasetGenerationTool.__name__, "YAML specification")
self._check_dataset(specs)
self._check_instruction(specs)
Util.check_paths(specs, DatasetGenerationTool.__name__)
Util.update_result_paths(specs, self.result_path, self.yaml_path)
def _parse_ml_method(ml_method_id: str, ml_specification) -> tuple:
valid_class_values = ReflectionHandler.all_nonabstract_subclass_basic_names(
MLMethod, "", "ml_methods/")
if type(ml_specification) is str:
ml_specification = {ml_specification: {}}
ml_specification = {
**DefaultParamsLoader.load("ml_methods/", "MLMethod"),
**ml_specification
}
ml_specification_keys = list(ml_specification.keys())
ParameterValidator.assert_all_in_valid_list(
list(ml_specification_keys),
["model_selection_cv", "model_selection_n_folds"] +
valid_class_values, "MLParser", ml_method_id)
non_default_keys = [
key for key in ml_specification.keys()
if key not in ["model_selection_cv", "model_selection_n_folds"]
]
assert len(ml_specification_keys) == 3, f"MLParser: ML method {ml_method_id} was not correctly specified. Expected at least 1 key " \
f"(ML method name), got {len(ml_specification_keys) - 2} instead: " \
f"{str([key for key in non_default_keys])[1:-1]}."
ml_method_class_name = non_default_keys[0]
ml_method_class = ReflectionHandler.get_class_by_name(
ml_method_class_name, "ml_methods/")
ml_specification[ml_method_class_name] = {
**DefaultParamsLoader.load("ml_methods/",
ml_method_class_name,
log_if_missing=False),
**ml_specification[ml_method_class_name]
}
method, params = MLParser.create_method_instance(
ml_specification, ml_method_class, ml_method_id)
ml_specification[ml_method_class_name] = params
method.name = ml_method_id
return method, ml_specification
def parse(self,
key: str,
instruction: dict,
symbol_table: SymbolTable,
path: Path = None) -> SubsamplingInstruction:
valid_keys = [
"type", "dataset", "subsampled_dataset_sizes",
"dataset_export_formats"
]
ParameterValidator.assert_keys(instruction.keys(), valid_keys,
SubsamplingParser.__name__, key)
dataset_keys = symbol_table.get_keys_by_type(SymbolType.DATASET)
ParameterValidator.assert_in_valid_list(instruction['dataset'],
dataset_keys,
SubsamplingParser.__name__,
f'{key}/dataset')
dataset = symbol_table.get(instruction['dataset'])
ParameterValidator.assert_type_and_value(
instruction['subsampled_dataset_sizes'], list,
SubsamplingParser.__name__, f'{key}/subsampled_dataset_sizes')
ParameterValidator.assert_all_type_and_value(
instruction['subsampled_dataset_sizes'], int,
SubsamplingParser.__name__, f'{key}/subsampled_dataset_sizes', 1,
dataset.get_example_count())
valid_export_formats = ReflectionHandler.all_nonabstract_subclass_basic_names(
DataExporter, 'Exporter', "dataset_export/")
ParameterValidator.assert_type_and_value(
instruction['dataset_export_formats'], list,
SubsamplingParser.__name__, f"{key}/dataset_export_formats")
ParameterValidator.assert_all_in_valid_list(
instruction['dataset_export_formats'], valid_export_formats,
SubsamplingParser.__name__, f"{key}/dataset_export_formats")
return SubsamplingInstruction(
dataset=dataset,
subsampled_dataset_sizes=instruction['subsampled_dataset_sizes'],
dataset_export_formats=[
ReflectionHandler.get_class_by_name(export_format + "Exporter",
"dataset_export/")
for export_format in instruction['dataset_export_formats']
],
name=key)
def parse_exporters(self, instruction):
if instruction["export_formats"] is not None:
class_path = "dataset_export/"
ParameterValidator.assert_all_in_valid_list(
instruction["export_formats"],
ReflectionHandler.all_nonabstract_subclass_basic_names(
DataExporter, 'Exporter', class_path),
location="SimulationParser",
parameter_name="export_formats")
exporters = [
ReflectionHandler.get_class_by_name(f"{item}Exporter",
class_path)
for item in instruction["export_formats"]
]
else:
exporters = None
return exporters
def parse_encoder(key: str, specs: dict):
class_path = "encodings"
valid_encoders = ReflectionHandler.all_nonabstract_subclass_basic_names(
DatasetEncoder, "Encoder", class_path)
encoder = ObjectParser.get_class(specs, valid_encoders, "Encoder",
class_path, "EncodingParser", key)
params = ObjectParser.get_all_params(specs, class_path,
encoder.__name__[:-7], key)
required_params = [
p for p in list(
inspect.signature(encoder.__init__).parameters.keys())
if p != "self"
]
ParameterValidator.assert_all_in_valid_list(
params.keys(), required_params, "EncoderParser",
f"{key}/{encoder.__name__.replace('Encoder', '')}")
return encoder, params
def build_object(cls, **kwargs):
location = "Coefficients"
coefs_to_plot = [coef.upper() for coef in kwargs["coefs_to_plot"]]
name = kwargs["name"] if "name" in kwargs else None
ParameterValidator.assert_all_in_valid_list(
coefs_to_plot,
[item.name.upper() for item in CoefficientPlottingSetting],
location, "coefs_to_plot")
if CoefficientPlottingSetting.CUTOFF.name in coefs_to_plot:
cutoff = kwargs["cutoff"]
ParameterValidator.assert_type_and_value(cutoff, list, location,
"cutoff")
ParameterValidator.assert_all_type_and_value(cutoff,
Number,
location,
"cutoff",
min_inclusive=1e-15)
else:
cutoff = []
if CoefficientPlottingSetting.N_LARGEST.name in coefs_to_plot:
n_largest = kwargs["n_largest"]
ParameterValidator.assert_type_and_value(n_largest, list, location,
"n_largest")
ParameterValidator.assert_all_type_and_value(n_largest,
int,
location,
"n_largest",
min_inclusive=1)
else:
n_largest = []
coefs = CoefficientPlottingSettingList()
for keyword in coefs_to_plot:
coefs.append(CoefficientPlottingSetting[keyword.upper()])
return Coefficients(coefs_to_plot=coefs,
cutoff=cutoff,
n_largest=n_largest,
name=name)
def _extract_reports(self):
with self.specification_path.open("r") as file:
workflow_specification = yaml.safe_load(file)
report_keys = list(workflow_specification['instructions'].values()
)[0]['benchmark_reports']
ParameterValidator.assert_all_in_valid_list(
report_keys,
list(workflow_specification['definitions']['reports'].keys()),
MultiDatasetBenchmarkTool.__name__, "benchmark_reports")
reports = {
key: value
for key, value in workflow_specification['definitions']
['reports'].items() if key in report_keys
}
symbol_table, _ = ReportParser.parse_reports(reports, SymbolTable())
self.reports = [
entry.item for entry in symbol_table.get_by_type(SymbolType.REPORT)
]
def parse(self,
key: str,
instruction: dict,
symbol_table: SymbolTable,
path: Path = None) -> DatasetExportInstruction:
location = "DatasetExportParser"
ParameterValidator.assert_keys(
list(instruction.keys()), DatasetExportParser.REQUIRED_KEYS +
DatasetExportParser.OPTIONAL_KEYS, location, key, False)
ParameterValidator.assert_keys_present(
list(instruction.keys()), DatasetExportParser.REQUIRED_KEYS,
location, key)
valid_formats = ReflectionHandler.all_nonabstract_subclass_basic_names(
DataExporter, "Exporter", 'dataset_export/')
ParameterValidator.assert_all_in_valid_list(
instruction["export_formats"], valid_formats, location,
"export_formats")
ParameterValidator.assert_all_in_valid_list(
instruction["datasets"],
symbol_table.get_keys_by_type(SymbolType.DATASET), location,
"datasets")
return DatasetExportInstruction(
datasets=[
symbol_table.get(dataset_key)
for dataset_key in instruction["datasets"]
],
exporters=[
ReflectionHandler.get_class_by_name(f"{key}Exporter",
"dataset_export/")
for key in instruction["export_formats"]
],
preprocessing_sequence=symbol_table.get(
instruction["preprocessing_sequence"])
if "preprocessing_sequence" in instruction else None,
name=key)
