MCPcopy Create free account
hub / github.com/FinancialComputingUCL/LOBFrame / cache_dataset

Method cache_dataset

loaders/custom_dataset.py:251–264  ·  view source on GitHub ↗
(self, dataset_index)

Source from the content-addressed store, hash-verified

249 return self.cumulative_lengths[-1]
250
251 def cache_dataset(self, dataset_index):
252 if self.current_cache_index >= 0:
253 # Remove the least recently used cache entry
254 self.cache_data[self.current_cache_index] = None
255 self.cache_indices[self.current_cache_index] = None
256
257 # Select a random cache slot for the new dataset
258 self.current_cache_index = random.randint(0, self.cache_size - 1)
259
260 # Cache the data from the CSV file
261 df = pl.read_csv(self.csv_files[dataset_index]).to_pandas()
262
263 self.cache_data[self.current_cache_index] = df.values[:, 1:].astype(np.float32)
264 self.cache_indices[self.current_cache_index] = dataset_index
265
266 def __getitem__(self, index):
267 try:

Callers 1

__getitem__Method · 0.95

Calls

no outgoing calls

Tested by

no test coverage detected