removed sampler since it is not used anymore

Author: Federico Errica

docs/source/mlwiz.data.rst

@@ -19,15 +19,6 @@ data.provider
    :undoc-members:
    :show-inheritance:
 
-data.sampler
--------------------------
-
-.. automodule:: mlwiz.data.sampler
-   :members:
-   :private-members:
-   :undoc-members:
-   :show-inheritance:
-
 data.splitter
 --------------------------
 

mlwiz/data/provider.py

@@ -17,7 +17,6 @@
 
 import mlwiz.data.dataset
 from mlwiz.data.dataset import DatasetInterface
-from mlwiz.data.sampler import RandomSampler
 from mlwiz.data.splitter import Splitter, SingleGraphSplitter
 from mlwiz.data.util import load_dataset, single_graph_collate
 
@@ -133,10 +132,6 @@ class DataProvider:
     special, but here is where the i-th element of a dataset could be
     pre-processed before constructing the mini-batches.
 
-    IMPORTANT: if the dataset is to be shuffled, you MUST use a
-    :class:`mlwiz.data.sampler.RandomSampler` object to determine the
-    permutation.
-
     Args:
         storage_folder (str): the path of the root folder in which data is stored
         splits_filepath (str): the filepath of the splits. with additional
@@ -369,9 +364,8 @@ def _get_loader(
                 dataset, sampler=sampler, batch_size=batch_size, **kwargs
             )
         elif shuffle is True:
-            sampler = RandomSampler(dataset)
             dataloader = self.data_loader_class(
-                dataset, sampler=sampler, batch_size=batch_size, **kwargs
+                dataset, shuffle=True, batch_size=batch_size, **kwargs
             )
         else:
             dataloader = self.data_loader_class(
@@ -725,10 +719,9 @@ def _get_loader(
         kwargs.update(self.data_loader_args)
 
         if shuffle is True:
-            sampler = RandomSampler(dataset)
             dataloader = self.data_loader_class(
                 dataset,
-                sampler=sampler,
+                shuffle=True,
                 batch_size=batch_size,
                 collate_fn=single_graph_collate,
                 **kwargs,

mlwiz/data/sampler.py

@@ -11,7 +11,7 @@
 
 import pytest
 import torch
-from torch.utils.data import DataLoader, SequentialSampler
+from torch.utils.data import DataLoader, RandomSampler, SequentialSampler
 
 import mlwiz.data.provider as provider_mod
 from mlwiz.data.provider import (
@@ -20,7 +20,6 @@
     SubsetTrainEval,
     _iterable_worker_init_fn,
 )
-from mlwiz.data.sampler import RandomSampler
 from mlwiz.data.splitter import Splitter
 
 
@@ -191,7 +190,7 @@ def test_data_provider_requires_seed_and_fold_ids():
 
 
 def test_data_provider_get_loader_uses_random_sampler_when_shuffling():
-    """Ensure shuffle=True uses MLWiz ``RandomSampler`` (with stored permutation)."""
+    """Ensure shuffle=True uses a random sampler."""
     provider = DataProvider(
         storage_folder="unused",
         splits_filepath="unused",

tests/data/test_data_provider_additional.py

@@ -109,7 +109,7 @@ def _raise_termination():
         engine._check_termination()
 
 
-def test_infer_sequential_sampler_requires_no_permutation_and_sets_main_keys(
+def test_infer_sets_main_keys(
     tmp_path,
 ):
     """infer() should set MAIN_* keys for downstream consumers."""