8201 Fix DataFrame subsets indexing in CSVDataset() (#8351)

Fixes #8201 .

### Description

`convert_tables_to_dicts` was using `.loc` to index DataFrames which was
changed to `.iloc`. It was causing unexpected behavior in `CSVDataset`
as demonstrated in #8201 because `.loc` expects labels, but was instead
provided positions of the rows.
Unittest was added which fails before the change and passes afterwards.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Bartosz Grabowski <58475557+bartosz-grabowski@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>

Author: 3 people

monai/data/utils.py

@@ -1473,7 +1473,7 @@ def convert_tables_to_dicts(
     # parse row indices
     rows: list[int | str] = []
     if row_indices is None:
-        rows = slice(df.shape[0])  # type: ignore
+        rows = df.index.tolist()
     else:
         for i in row_indices:
             if isinstance(i, (tuple, list)):

tests/data/test_csv_dataset.py

@@ -179,6 +179,20 @@ def prepare_csv_file(data, filepath):
                 },
             )
 
+            # test pre-loaded DataFrame subset
+            df = pd.read_csv(filepath1)
+            df_subset = df.iloc[[1, 3, 4]]
+            dataset = CSVDataset(src=df_subset, col_groups={"ehr": [f"ehr_{i}" for i in range(3)]})
+            self.assertEqual(len(dataset), 3)
+            np.testing.assert_allclose([round(i, 4) for i in dataset[1]["ehr"]], [3.3333, 3.2353, 3.4000])
+
+            # test pre-loaded DataFrame subset with row_indices != None
+            df = pd.read_csv(filepath1)
+            df_subset = df.iloc[[1, 3, 4]]
+            dataset = CSVDataset(src=df_subset, row_indices=[1, 3], col_groups={"ehr": [f"ehr_{i}" for i in range(3)]})
+            self.assertEqual(len(dataset), 2)
+            np.testing.assert_allclose([round(i, 4) for i in dataset[1]["ehr"]], [3.3333, 3.2353, 3.4000])
+
             # test pre-loaded multiple DataFrames, join tables with kwargs
             dfs = [pd.read_csv(i) for i in filepaths]
             dataset = CSVDataset(src=dfs, on="subject_id")