Correct crown-curve-indices definition in trait pipeline (#83)

* add mermaid diagram notebook

* display mermaid diagrams and save them as pngs

* add tests for json serialization

* check attributes of labeled frame instead of labeledframe which does not assert as equal

* updated documentation for loading series from slp paths

* update DicotPipeline notebook with `Series` changes

* modify curve index function so that warnings are not output when lengths are not valid

* fix definition of `crown_curve_indices`

* fix `crown_curve_indices` in `OlderMonocotPipeline`

* assert curve indices are between 0 and 1 in tests

sleap_roots/lengths.py

@@ -125,7 +125,7 @@ def get_curve_index(
         & (~np.isnan(base_tip_dists))
         & (lengths > 0)
         & (lengths >= base_tip_dists),
-        (lengths - base_tip_dists) / lengths,
+        (lengths - base_tip_dists) / np.where(lengths != 0, lengths, np.nan),
         np.nan,
     )
 

sleap_roots/series.py

@@ -473,11 +473,10 @@ def load_series_from_slps(
 ) -> List[Series]:
     """Load a list of Series from a list of .slp paths.
 
-    To load the `Series`, the files must be named with the following convention:
+    To load the `Series`, the files must be named with the following convention.
+    The `slp_paths` are expeted to have the `series_name` in the filename and "primary",
+    "lateral", or "crown" in the filename to differentiate the predictions.
     h5_path: '/path/to/scan/series_name.h5'
-    primary_path: '/path/to/scan/series_name.model{model_id}.rootprimary.slp'
-    lateral_path: '/path/to/scan/series_name.model{model_id}.rootlateral.slp'
-    crown_path: '/path/to/scan/series_name.model{model_id}.rootcrown.slp'
     Note that everything is expected to be in the same folder.
 
     Our pipeline outputs prediction files with this format:
@@ -500,7 +499,7 @@ def load_series_from_slps(
         if h5s:
             # Get directory of the h5s
             h5_dir = Path(slp_paths[0]).parent
-            # Generate the path to the .h5 file
+            # Create path to the .h5 file
             h5_path = h5_dir / f"{series_name}.h5"
         else:
             h5_path = None

sleap_roots/trait_pipelines.py

@@ -1597,8 +1597,8 @@ def define_traits(self) -> List[TraitDef]:
             ),
             TraitDef(
                 name="crown_curve_indices",
-                fn=get_base_tip_dist,
-                input_traits=["crown_base_pts", "crown_tip_pts"],
+                fn=get_curve_index,
+                input_traits=["crown_lengths", "crown_base_tip_dists"],
                 scalar=False,
                 include_in_csv=True,
                 kwargs={},
@@ -1974,8 +1974,8 @@ def define_traits(self) -> List[TraitDef]:
             ),
             TraitDef(
                 name="crown_curve_indices",
-                fn=get_base_tip_dist,
-                input_traits=["crown_base_pts", "crown_tip_pts"],
+                fn=get_curve_index,
+                input_traits=["crown_lengths", "crown_base_tip_dists"],
                 scalar=False,
                 include_in_csv=True,
                 kwargs={},

tests/test_lengths.py

@@ -288,7 +288,6 @@ def test_invalid_scalar_values():
     assert np.isnan(get_curve_index(0, 8))
 
 
-# tests for `get_root_lengths`
 def test_curve_index_float():
     assert get_curve_index(10.0, 5.0) == 0.5
 

tests/test_series.py

@@ -280,7 +280,12 @@ def test_get_frame_rice_10do(
     # Get the crown labeled frame
     crown_lf = frames.get("crown")
 
-    assert crown_lf == expected_labeled_frame
+    # Compare the attributes of the labeled frames
+    assert crown_lf.frame_idx == expected_labeled_frame.frame_idx
+    assert crown_lf.instances == expected_labeled_frame.instances
+    assert crown_lf.video.filename == expected_labeled_frame.video.filename
+    assert crown_lf.video.shape == expected_labeled_frame.video.shape
+    assert crown_lf.video.backend == expected_labeled_frame.video.backend
     assert series.series_name == "0K9E8BI"
 
 
@@ -302,7 +307,12 @@ def test_get_frame_rice_10do_no_video(
     # Get the crown labeled frame
     crown_lf = frames.get("crown")
 
-    assert crown_lf == expected_labeled_frame
+    # Compare the attributes of the labeled frames
+    assert crown_lf.frame_idx == expected_labeled_frame.frame_idx
+    assert crown_lf.instances == expected_labeled_frame.instances
+    assert crown_lf.video.filename == expected_labeled_frame.video.filename
+    assert crown_lf.video.shape == expected_labeled_frame.video.shape
+    assert crown_lf.video.backend == expected_labeled_frame.video.backend
     assert series.series_name == "0K9E8BI"
 
 

tests/test_trait_pipelines.py

@@ -1,10 +1,14 @@
 import numpy as np
 import pandas as pd
+import json
+import pytest
+
 from sleap_roots.trait_pipelines import (
     DicotPipeline,
     YoungerMonocotPipeline,
     OlderMonocotPipeline,
     MultipleDicotPipeline,
+    NumpyArrayEncoder,
 )
 from sleap_roots.series import (
     Series,
@@ -15,6 +19,47 @@
 )
 
 
+def test_numpy_array_serialization():
+    array = np.array([1, 2, 3])
+    expected = [1, 2, 3]
+    json_str = json.dumps(array, cls=NumpyArrayEncoder)
+    assert json.loads(json_str) == expected
+
+
+def test_numpy_int64_serialization():
+    int64_value = np.int64(42)
+    expected = 42
+    json_str = json.dumps(int64_value, cls=NumpyArrayEncoder)
+    assert json.loads(json_str) == expected
+
+
+def test_unsupported_type_serialization():
+    class UnsupportedType:
+        pass
+
+    with pytest.raises(TypeError):
+        json.dumps(UnsupportedType(), cls=NumpyArrayEncoder)
+
+
+def test_mixed_data_serialization():
+    data = {
+        "array": np.array([1, 2, 3]),
+        "int64": np.int64(42),
+        "regular_int": 99,
+        "list": [4, 5, 6],
+        "dict": {"key": "value"},
+    }
+    expected = {
+        "array": [1, 2, 3],
+        "int64": 42,
+        "regular_int": 99,
+        "list": [4, 5, 6],
+        "dict": {"key": "value"},
+    }
+    json_str = json.dumps(data, cls=NumpyArrayEncoder)
+    assert json.loads(json_str) == expected
+
+
 def test_dicot_pipeline(
     canola_h5,
     soy_h5,
@@ -107,12 +152,17 @@ def test_younger_monocot_pipeline(rice_pipeline_output_folder):
     assert (
         rice_traits["curve_index"].fillna(0) >= 0
     ).all(), "curve_index in rice_traits contains negative values"
+    assert rice_traits["curve_index"].fillna(0).max() <= 1, "curve_index in rice_traits contains values greater than 1"
     assert (
         all_traits["curve_index_median"] >= 0
     ).all(), "curve_index in all_traits contains negative values"
+    assert all_traits["curve_index_median"].max() <= 1, "curve_index in all_traits contains values greater than 1"
     assert (
         all_traits["crown_curve_indices_mean_median"] >= 0
     ).all(), "crown_curve_indices_mean_median in all_traits contains negative values"
+    assert (
+        all_traits["crown_curve_indices_mean_median"] <= 1
+    ).all(), "crown_curve_indices_mean_median in all_traits contains values greater than 1"
     assert (
         (0 <= rice_traits["crown_angles_proximal_p95"])
         & (rice_traits["crown_angles_proximal_p95"] <= 180)
@@ -169,6 +219,9 @@ def test_older_monocot_pipeline(rice_10do_pipeline_output_folder):
     assert (
         all_traits["crown_curve_indices_mean_median"] >= 0
     ).all(), "crown_curve_indices_mean_median in all_traits contains negative values"
+    assert (
+        all_traits["crown_curve_indices_mean_median"] <= 1
+    ).all(), "crown_curve_indices_mean_median in all_traits contains values greater than 1"
     assert (
         (0 <= rice_traits["crown_angles_proximal_p95"])
         & (rice_traits["crown_angles_proximal_p95"] <= 180)