MCPcopy Create free account
hub / github.com/Open-Bee/DataStudio / create_single_dataloader

Function create_single_dataloader

datastudio/datasets/builder.py:102–113  ·  view source on GitHub ↗

Create a single dataloader for the given dataset path.

(dataset_path)

Source from the content-addressed store, hash-verified

100 return []
101
102 def create_single_dataloader(dataset_path):
103 """Create a single dataloader for the given dataset path."""
104 current_config = config.copy()
105 current_config["dataset"] = dataset_path
106 return build_from_cfg(
107 current_config,
108 DATALOADER,
109 logger=logger,
110 batch_size=batch_size,
111 num_workers=num_workers,
112 checkpoint_manager=checkpoint_manager,
113 )
114
115 # Parallel or sequential loading
116 if parallel_load and len(pending_datasets) > 1:

Callers 1

build_dataloadersFunction · 0.85

Calls 1

build_from_cfgFunction · 0.90

Tested by

no test coverage detected