Merge pull request #54 from ggmarshall/main

FFT factory funcs fix and bugfix for interpolated time point thresh

src/dspeed/processing_chain.py

@@ -1406,20 +1406,11 @@ def __init__(
                 )
 
                 # reshape just in case there are some missing dimensions
-                arshape = list(param.shape)
+                arshape = list(param.buffer.shape)
                 for idim in range(-1, -1 - len(shape), -1):
-                    if (
-                        len(arshape) < len(shape) + 1 + idim
-                        or arshape[idim] != shape[idim]
-                    ):
+                    if len(arshape) < -idim or arshape[idim] != shape[idim]:
                         arshape.insert(len(arshape) + idim + 1, 1)
-
-                if param.is_const:
-                    param = param.get_buffer(grid).reshape(arshape)
-                else:
-                    param = param.get_buffer(grid).reshape(
-                        tuple([self.proc_chain._block_width] + arshape)
-                    )
+                param = param.get_buffer(grid).reshape(arshape)
 
             elif isinstance(param, str):
                 # Convert string into integer buffer if appropriate

src/dspeed/processors/time_point_thresh.py

@@ -157,19 +157,19 @@ def interpolated_time_point_thresh(
         for i in range(int(t_start), len(w_in) - 1, 1):
             if w_in[i] <= a_threshold < w_in[i + 1]:
                 i_cross = i
-                return
+                break
     else:
         for i in range(int(t_start), 1, -1):
             if w_in[i - 1] < a_threshold <= w_in[i]:
                 i_cross = i - 1
-                return
+                break
 
     if i_cross == -1:
         return
 
     if mode_in == ord("i"):  # return index before crossing
         t_out[0] = i_cross
-    elif mode_in == ord("f"):  # return index before crossing
+    elif mode_in == ord("f"):  # return index after crossing
         t_out[0] = i_cross + 1
     elif mode_in == ord("c"):  # return index before crossing
         t_out[0] = i_cross

tests/configs/icpc-dsp-config.json

@@ -33,7 +33,8 @@
     "cuspEmax",
     "zacEmax",
     "zacEftp",
-    "cuspEftp"
+    "cuspEftp",
+    "wf_psd"
   ],
   "processors": {
     "tp_min, tp_max, wf_min, wf_max": {
@@ -357,6 +358,12 @@
       "module": "numpy",
       "args": ["tp_0_est", "tp_aoe_max/16", "tp_aoe_samp"],
       "unit": "ns"
+    },
+    "wf_psd": {
+      "function": "psd",
+      "module": "dspeed.processors",
+      "args": ["wf_blsub", "wf_psd"],
+      "init_args": ["wf_blsub", "wf_psd"]
     }
   }
 }

tests/processors/test_time_point_thresh.py

@@ -0,0 +1,184 @@
+import numpy as np
+import pytest
+
+from dspeed.errors import DSPFatal
+from dspeed.processors import interpolated_time_point_thresh, time_point_thresh
+
+
+def test_time_point_thresh(compare_numba_vs_python):
+    """Testing function for the time_point_thresh processor."""
+
+    # test for nan if w_in has a nan
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    w_in[4] = np.nan
+    assert np.isnan(
+        compare_numba_vs_python(
+            time_point_thresh,
+            w_in,
+            1,
+            11,
+            0,
+        )
+    )
+
+    # test for nan if nan is passed to a_threshold
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert np.isnan(
+        compare_numba_vs_python(
+            time_point_thresh,
+            w_in,
+            np.nan,
+            11,
+            0,
+        )
+    )
+
+    # test for nan if nan is passed to t_start
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert np.isnan(
+        compare_numba_vs_python(
+            time_point_thresh,
+            w_in,
+            1,
+            np.nan,
+            0,
+        )
+    )
+
+    # test for nan if nan is passed to walk_forward
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert np.isnan(
+        compare_numba_vs_python(
+            time_point_thresh,
+            w_in,
+            1,
+            11,
+            np.nan,
+        )
+    )
+
+    # test for error if t_start non integer
+    with pytest.raises(DSPFatal):
+        w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+        time_point_thresh(w_in, 1, 10.5, 0, np.array([0.0]))
+
+    # test for error if walk_forward non integer
+    with pytest.raises(DSPFatal):
+        w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+        time_point_thresh(w_in, 1, 11, 0.5, np.array([0.0]))
+
+    # test for error if t_start out of range
+    with pytest.raises(DSPFatal):
+        w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+        time_point_thresh(w_in, 1, 12, 0, np.array([0.0]))
+
+    # test walk backward
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert compare_numba_vs_python(time_point_thresh, w_in, 1, 11, 0) == 8.0
+
+    # test walk forward
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert compare_numba_vs_python(time_point_thresh, w_in, 3, 0, 1) == 4.0
+
+
+def test_interpolated_time_point_thresh(compare_numba_vs_python):
+    """Testing function for the interpolated_time_point_thresh processor."""
+
+    # test for nan if w_in has a nan
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    w_in[4] = np.nan
+    assert np.isnan(
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1.0, 11.0, 0, 105)
+    )
+
+    # test for nan if nan is passed to a_threshold
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert np.isnan(
+        compare_numba_vs_python(
+            interpolated_time_point_thresh, w_in, np.nan, 11.0, 0, 105
+        )
+    )
+
+    # test for nan if nan is passed to t_start
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert np.isnan(
+        compare_numba_vs_python(
+            interpolated_time_point_thresh, w_in, 1.0, np.nan, 0, 105
+        )
+    )
+
+    # test for nan if t_start out of range
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert np.isnan(
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1.0, 12, 0, 105)
+    )
+
+    # test walk backward mode 'i'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1, 11, 0, 105)
+        == 7.0
+    )
+
+    # test walk forward mode 'i'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 3, 0, 1, 105)
+        == 4.0
+    )
+
+    # test walk backward mode 'f'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1, 11, 0, 102)
+        == 8.0
+    )
+
+    # test walk forward mode 'f'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 3, 0, 1, 102)
+        == 5.0
+    )
+
+    # test walk backward mode 'f'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1, 11, 0, 99)
+        == 7.0
+    )
+
+    # test walk forward mode 'f'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 3, 0, 1, 99)
+        == 4.0
+    )
+
+    # test walk backward mode 'n'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1, 11, 0, 110)
+        == 7.5
+    )
+
+    # test walk forward mode 'n'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 3, 0, 1, 110)
+        == 4.5
+    )
+
+    # test walk backward mode 'l'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 1.5, 11, 0, 108)
+        == 8.5
+    )
+
+    # test walk forward mode 'l'
+    w_in = np.concatenate([np.arange(-1, 5, 1), np.arange(-1, 5, 1)], dtype="float")
+    assert (
+        compare_numba_vs_python(interpolated_time_point_thresh, w_in, 3.5, 0, 1, 108)
+        == 4.5
+    )