diff --git a/.github/workflows/ci-linux.yml b/.github/workflows/ci-linux.yml
index cff3d78e4..36ebefe31 100644
--- a/.github/workflows/ci-linux.yml
+++ b/.github/workflows/ci-linux.yml
@@ -22,6 +22,7 @@ jobs:
         run: |
           nox -s test-data-download-source
           nox -s test-data-download-generated-ci
-          nox -s coverage --verbose
+          pip install .[test]
+          pytest --cov=./ --cov-report=xml
       - name: upload code coverage
         uses: codecov/codecov-action@v3
diff --git a/src/vak/learncurve/frame_classification.py b/src/vak/learncurve/frame_classification.py
index 0eff85775..4c02a74ba 100644
--- a/src/vak/learncurve/frame_classification.py
+++ b/src/vak/learncurve/frame_classification.py
@@ -171,11 +171,12 @@ def learning_curve_for_frame_classification_model(
     dataset_df = dataset_df[
         (dataset_df.train_dur.notna()) & (dataset_df.replicate_num.notna())
     ]
-    train_durs = sorted(dataset_df["train_dur"].unique())
-    replicate_nums = [
-        int(replicate_num)
-        for replicate_num in sorted(dataset_df["replicate_num"].unique())
-    ]
+    train_durs = sorted(
+        [int(train_dur) for train_dur in dataset_df["train_dur"].unique()]
+    )
+    replicate_nums = sorted(
+        [int(replicate_num) for replicate_num in dataset_df["replicate_num"].unique()]
+    )
     to_do = []
     for train_dur in train_durs:
         for replicate_num in replicate_nums:
diff --git a/src/vak/prep/frame_classification/source_files.py b/src/vak/prep/frame_classification/source_files.py
index e5888db9f..bc2223550 100644
--- a/src/vak/prep/frame_classification/source_files.py
+++ b/src/vak/prep/frame_classification/source_files.py
@@ -1,3 +1,9 @@
+"""Get source files for a dataset, or make them.
+
+Gets either audio or spectrogram files from ``data dir``,
+possibly paired with annotation files."""
+from __future__ import annotations
+
 import logging
 import pathlib