Merge pull request #237 from tobac-project/hotfix

Version 1.4.1 release

.zenodo.json

@@ -55,6 +55,11 @@
             "affiliation": "Leibniz Institute for Tropospheric Research, Leipzig (Germany)",
             "orcid": "0000-0002-5350-1445"
         },
+        {
+            "name": "Lettl, Kolya",
+            "affiliation": "Leibniz Institute for Tropospheric Research, Leipzig (Germany)",
+            "orcid": "0000-0002-4524-8152"
+        },
         {
             "name": "Raut, Bhupendra A.",
             "affiliation": "Northwestern-Argonne Institute of Science and Engineering, Argonne National Laboratory",

CHANGELOG.md

@@ -1,5 +1,16 @@
 ### Tobac Changelog
 
+_**Version 1.4.1:**_
+
+**Bug fixes**
+
+- Fixed a bug with predictive tracking that would duplicate column names if the input dataset has coordinates x and/or y [#217](https://github.com/tobac-project/tobac/pull/217)
+- Set extrapolate parameter to 0 in example notebooks to prevent not implemented error [#217](https://github.com/tobac-project/tobac/pull/217)
+
+**Documentation**
+
+- Regenerated example notebooks so that they are up to date for the present version [#233](https://github.com/tobac-project/tobac/pull/233)
+
 _**Version 1.4.0:**_
 
 **Enhancements**

tobac/__init__.py

@@ -77,4 +77,4 @@
 from . import merge_split
 
 # Set version number
-__version__ = "1.4.0"
+__version__ = "1.4.1"

tobac/tests/test_feature_detection.py

@@ -194,3 +194,36 @@ def test_feature_detection_threshold_sort(test_threshs, target):
             target=target,
         )
         assert_frame_equal(fd_output_first, fd_output_test)
+
+
+def test_feature_detection_coords():
+    """Tests that the output features dataframe contains all the coords of the input iris cube"""
+    test_dset_size = (50, 50)
+    test_hdim_1_pt = 20.0
+    test_hdim_2_pt = 20.0
+    test_hdim_1_sz = 5
+    test_hdim_2_sz = 5
+    test_amp = 2
+    test_min_num = 2
+
+    test_data = np.zeros(test_dset_size)
+    test_data = tbtest.make_feature_blob(
+        test_data,
+        test_hdim_1_pt,
+        test_hdim_2_pt,
+        h1_size=test_hdim_1_sz,
+        h2_size=test_hdim_2_sz,
+        amplitude=test_amp,
+    )
+    test_data_iris = tbtest.make_dataset_from_arr(test_data, data_type="iris")
+    fd_output_first = feat_detect.feature_detection_multithreshold_timestep(
+        test_data_iris,
+        0,
+        threshold=[1, 2, 3],
+        n_min_threshold=test_min_num,
+        dxy=1,
+        target="maximum",
+    )
+
+    for coord in test_data_iris.coords():
+        assert coord.name() in fd_output_first

tobac/tests/test_tracking.py

@@ -49,6 +49,14 @@ def test_linking_trackpy():
         expected_out_feature.sort_index(axis=1), actual_out_feature.sort_index(axis=1)
     )
 
+    # Check that we only add two columns, and all the other columns are the same as the input features
+    assert len(actual_out_feature.columns.tolist()) == len(
+        set(actual_out_feature.columns.tolist())
+    )
+    assert set(actual_out_feature.columns.tolist()) - set(
+        test_feature.columns.tolist()
+    ) == {"cell"}
+
 
 @pytest.mark.parametrize(
     "max_trackpy, max_tobac, adaptive_step, adaptive_stop",
@@ -142,6 +150,10 @@ def test_trackpy_predict():
 
     assert_frame_equal(expected_output.sort_index(), output.sort_index())
 
+    # Check that we only add two columns, and all the other columns are the same as the input features
+    assert len(output.columns.tolist()) == len(set(output.columns.tolist()))
+    assert set(output.columns.tolist()) - set(features.columns.tolist()) == {"cell"}
+
 
 def test_tracking_extrapolation():
     """Tests the extrapolation capabilities of tracking.

tobac/tracking.py

@@ -3,18 +3,18 @@
 The individual features and associated area/volumes identified in
 each timestep have to be linked into trajectories to analyse
 the time evolution of their properties for a better understanding of
-the underlying physical processes. 
+the underlying physical processes.
 The implementations are structured in a way that allows for the future
 addition of more complex tracking methods recording a more complex
 network of relationships between features at different points in
-time. 
+time.
 
 References
 ----------
 .. Heikenfeld, M., Marinescu, P. J., Christensen, M.,
    Watson-Parris, D., Senf, F., van den Heever, S. C.
-   & Stier, P. (2019). tobac 1.2: towards a flexible 
-   framework for tracking and analysis of clouds in 
+   & Stier, P. (2019). tobac 1.2: towards a flexible
+   framework for tracking and analysis of clouds in
    diverse datasets. Geoscientific Model Development,
    12(11), 4551-4570.
 """
@@ -233,12 +233,12 @@ def linking_trackpy(
             tp.linking.Linker.MAX_SUB_NET_SIZE_ADAPTIVE = subnetwork_size
 
     # deep copy to preserve features field:
-    features_linking = deepcopy(features)
+    features = deepcopy(features)
 
     if method_linking == "random":
         #     link features into trajectories:
         trajectories_unfiltered = tp.link(
-            features_linking,
+            features,
             search_range=search_range,
             memory=memory,
             t_column="frame",
@@ -251,6 +251,9 @@ def linking_trackpy(
     elif method_linking == "predict":
 
         # avoid setting pos_columns by renaimng to default values to avoid trackpy bug
+        features.rename(
+            columns={"y": "__temp_y_coord", "x": "__temp_x_coord"}, inplace=True
+        )
         features.rename(columns={"hdim_1": "y", "hdim_2": "x"}, inplace=True)
 
         # generate list of features as input for df_link_iter to avoid bug in df_link
@@ -280,7 +283,9 @@ def linking_trackpy(
         trajectories_unfiltered.rename(
             columns={"y": "hdim_1", "x": "hdim_2"}, inplace=True
         )
-        features.rename(columns={"y": "hdim_1", "x": "hdim_2"}, inplace=True)
+        trajectories_unfiltered.rename(
+            columns={"__temp_y_coord": "y", "__temp_x_coord": "x"}, inplace=True
+        )
     else:
         raise ValueError("method_linking unknown")