Speed up wave.resource module (MHKiT-Software#352)

* fix assignment in type_handling

* temporary testing file

* initial conversion of energy_period and frequency_moment to DataArray

* energy_period working with variety of types and converting to dataArray base

* extend xr.dataarray basis to all wave.resource functions

* remove testing script

* black formatting

* fix most test formatting

* use dataarrays instead of datasets in wave.performance

* revert surface_elevation function back to datasets

* Revert "revert surface_elevation function back to datasets"

This reverts commit a2d5f61.

* allow datasets, 2d dataframes. Update test formatting appropriately

* simplify and improve robustness of convert_to_dataarray for 1-var dataframes and 2+ var datasets

* update test formatting

* clean up frequency_bin and method checks in elevation_surface

* update and annotate type_handling

* black formatting

* minor type fix

* update type references in loads

* update type references in loads - v2

* black formatting

* fix call to fillna() in MLER example

* fix references to k in MLER example

* add variable names to Hm0, Te, and Tp for DataFrame creation

* update pd.Series naming in examples

* fix typo in pacwave notebook

* add variable names after data conversion

* update pacwave example

* add type check to all wave.resource variable naming

* update wave example with new data types

* pull buoy name from metadata in cdip example

* tighten up example timing

* address minor review comments, add some type checking tests

* complicated dataset and dataframe handling in surface_elevation and elevation_spectrum

* restore and simplify dataset input to elevation_spectrum, surface_elevation

* black formatting

* update missed docstring

* use np.nan_to_num correctly in test_loads

* Speed up wave (#6)

* matplotlib >=3.8

* remove debug

---------

Co-authored-by: ssolson <[email protected]>

Author: akeeste

.github/workflows/generate_notebook_matrix.py

@@ -14,14 +14,14 @@
     "cdip_example.ipynb": 420,
     "Delft3D_example.ipynb": 180,
     "directional_waves.ipynb": 180,
-    "environmental_contours_example.ipynb": 360,
-    "extreme_response_contour_example.ipynb": 360,
-    "extreme_response_full_sea_state_example.ipynb": 360,
-    "extreme_response_MLER_example.ipynb": 360,
+    "environmental_contours_example.ipynb": 240,
+    "extreme_response_contour_example.ipynb": 240,
+    "extreme_response_full_sea_state_example.ipynb": 240,
+    "extreme_response_MLER_example.ipynb": 240,
     "loads_example.ipynb": 180,
     "metocean_example.ipynb": 180,
     "mooring_example.ipynb": 240,
-    "PacWave_resource_characterization_example.ipynb": 780,
+    "PacWave_resource_characterization_example.ipynb": 240,
     "power_example.ipynb": 180,
     "qc_example.ipynb": 180,
     "river_example.ipynb": 180,

examples/PacWave_resource_characterization_example.ipynb

@@ -1119,12 +1119,19 @@
     "\n",
     "    Tz_list.append(resource.average_zero_crossing_period(year_data.T))\n",
     "\n",
-    "# Concatenate list of Series into a single DataFrame\n",
+    "# Concatenate each list of Series into a single Series\n",
     "Te = pd.concat(Te_list, axis=0)\n",
     "Tp = pd.concat(Tp_list, axis=0)\n",
     "Hm0 = pd.concat(Hm0_list, axis=0)\n",
     "J = pd.concat(J_list, axis=0)\n",
     "Tz = pd.concat(Tz_list, axis=0)\n",
+    "\n",
+    "# Name each Series and concat into a dataFrame\n",
+    "Te.name = 'Te'\n",
+    "Tp.name = 'Tp'\n",
+    "Hm0.name = 'Hm0'\n",
+    "J.name = 'J'\n",
+    "Tz.name = 'Tz'\n",
     "data = pd.concat([Hm0, Te, Tp, J, Tz], axis=1)\n",
     "\n",
     "# Calculate wave steepness\n",
@@ -1590,7 +1597,7 @@
     "    J = []\n",
     "    for i in range(len(result)):\n",
     "        b = resource.jonswap_spectrum(f, result.Tp[i], result.Hm0[i])\n",
-    "        J.extend([resource.energy_flux(b, h=399.0).values[0][0]])\n",
+    "        J.extend([resource.energy_flux(b, h=399.0).item()])\n",
     "\n",
     "    result[\"J\"] = J\n",
     "    results[N] = result\n",
@@ -1622,7 +1629,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
+   "version": "3.12.7"
   }
  },
  "nbformat": 4,

examples/cdip_example.ipynb

@@ -576,7 +576,7 @@
     "Hs = buoy_data[\"data\"][\"wave\"][\"waveHs\"]\n",
     "Tp = buoy_data[\"data\"][\"wave\"][\"waveTp\"]\n",
     "Dp = buoy_data[\"data\"][\"wave\"][\"waveDp\"]\n",
-    "buoy_name = buoy_data[\"data\"][\"wave\"].name\n",
+    "buoy_name = buoy_data[\"metadata\"][\"name\"]\n",
     "ax = graphics.plot_compendium(Hs, Tp, Dp, buoy_name)"
    ]
   },
@@ -590,7 +590,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -604,7 +604,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
+   "version": "3.12.7"
   }
  },
  "nbformat": 4,

examples/environmental_contours_example.ipynb

@@ -647,9 +647,13 @@
     "    Hm0_list.append(resource.significant_wave_height(year_data.T))\n",
     "    Te_list.append(resource.energy_period(year_data.T))\n",
     "\n",
-    "# Concatenate list of Series into a single DataFrame\n",
+    "# Concatenate each list of Series into a single Series\n",
     "Te = pd.concat(Te_list, axis=0)\n",
     "Hm0 = pd.concat(Hm0_list, axis=0)\n",
+    "\n",
+    "# Name each Series and concat into a dataFrame\n",
+    "Te.name = 'Te'\n",
+    "Hm0.name = 'Hm0'\n",
     "Hm0_Te = pd.concat([Hm0, Te], axis=1)\n",
     "\n",
     "# Drop any NaNs created from the calculation of Hm0 or Te\n",
@@ -800,7 +804,7 @@
    "source": [
     "## Resource Clusters\n",
     "\n",
-    "Often in resource characterization we want to pick a few representative sea state to run an alaysis. To do this with the resource data in python we reccomend using a Gaussian Mixture Model (a more generalized k-means clustering method). Using sckitlearn this is very straigth forward. We combine our Hm0 and Te data into an N x 2 numpy array. We specify our number of components (number of representative sea states) and then call the fit method on the data. Fianlly, using the methods `means_` and `weights` we can organize the results into an easily digestable table."
+    "Often in resource characterization we want to pick a few representative sea state to run an alaysis. To do this with the resource data in python we reccomend using a Gaussian Mixture Model (a more generalized k-means clustering method). Using sckitlearn this is very straight forward. We combine our Hm0 and Te data into an N x 2 numpy array. We specify our number of components (number of representative sea states) and then call the fit method on the data. Fianlly, using the methods `means_` and `weights` we can organize the results into an easily digestable table."
    ]
   },
   {
@@ -933,9 +937,13 @@
     "    Hm0_list.append(resource.significant_wave_height(year_data.T))\n",
     "    Tp_list.append(resource.peak_period(year_data.T))\n",
     "\n",
-    "# Concatenate list of Series into a single DataFrame\n",
+    "# Concatenate each list of Series into a single Series\n",
     "Tp = pd.concat(Tp_list, axis=0)\n",
     "Hm0 = pd.concat(Hm0_list, axis=0)\n",
+    "\n",
+    "# Name each Series and concat into a dataFrame\n",
+    "Tp.name = 'Tp'\n",
+    "Hm0.name = 'Hm0'\n",
     "Hm0_Tp = pd.concat([Hm0, Tp], axis=1)\n",
     "\n",
     "# Drop any NaNs created from the calculation of Hm0 or Te\n",
@@ -1116,7 +1124,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3.9.13 ('.venv': venv)",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -1130,7 +1138,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
+   "version": "3.12.4"
   },
   "vscode": {
    "interpreter": {

examples/extreme_response_MLER_example.ipynb

@@ -197,11 +197,11 @@
     "\n",
     "# generate wave number k\n",
     "k = resource.wave_number(wave_freq, 70)\n",
-    "k = k.fillna(0)\n",
+    "np.nan_to_num(k, 0)\n",
     "\n",
     "peakHeightDesired = Hs / 2 * 1.9\n",
     "mler_norm = extreme.mler_wave_amp_normalize(\n",
-    "    peakHeightDesired, mler_data, sim, k.k.values\n",
+    "    peakHeightDesired, mler_data, sim, k\n",
     ")"
    ]
   },
@@ -239,7 +239,7 @@
     }
    ],
    "source": [
-    "mler_ts = extreme.mler_export_time_series(RAO.values, mler_norm, sim, k.k.values)\n",
+    "mler_ts = extreme.mler_export_time_series(RAO.values, mler_norm, sim, k)\n",
     "mler_ts.plot(xlabel=\"Time (s)\", ylabel=\"[m] / [*]\", xlim=[-100, 100], grid=True)"
    ]
   },
@@ -256,7 +256,7 @@
    "hash": "6acc4428af86beefd6565514d05fe9ce8e024621fafadd3627cdac7b7bd68bc4"
   },
   "kernelspec": {
-   "display_name": "Python 3.8.10 64-bit ('MHKdev': conda)",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -270,10 +270,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
-  },
-  "orig_nbformat": 4
+   "version": "3.12.4"
+  }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }

examples/extreme_response_contour_example.ipynb

@@ -72,9 +72,13 @@
     "    Hm0_list.append(resource.significant_wave_height(year_data.T))\n",
     "    Te_list.append(resource.energy_period(year_data.T))\n",
     "\n",
-    "# Concatenate list of Series into a single DataFrame\n",
+    "# Concatenate each list of Series into a single Series\n",
     "Te = pd.concat(Te_list, axis=0)\n",
     "Hm0 = pd.concat(Hm0_list, axis=0)\n",
+    "\n",
+    "# Name each Series and concat into a dataFrame\n",
+    "Te.name = 'Te'\n",
+    "Hm0.name = 'Hm0'\n",
     "Hm0_Te = pd.concat([Hm0, Te], axis=1)\n",
     "\n",
     "# Drop any NaNs created from the calculation of Hm0 or Te\n",
@@ -323,7 +327,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
+   "version": "3.12.4"
   }
  },
  "nbformat": 4,

examples/extreme_response_full_sea_state_example.ipynb

@@ -75,9 +75,13 @@
     "    Hm0_list.append(resource.significant_wave_height(year_data.T))\n",
     "    Te_list.append(resource.energy_period(year_data.T))\n",
     "\n",
-    "# Concatenate list of Series into a single DataFrame\n",
+    "# Concatenate each list of Series into a single Series\n",
     "Te = pd.concat(Te_list, axis=0)\n",
     "Hm0 = pd.concat(Hm0_list, axis=0)\n",
+    "\n",
+    "# Name each Series and concat into a dataFrame\n",
+    "Te.name = 'Te'\n",
+    "Hm0.name = 'Hm0'\n",
     "Hm0_Te = pd.concat([Hm0, Te], axis=1)\n",
     "\n",
     "# Drop any NaNs created from the calculation of Hm0 or Te\n",
@@ -573,7 +577,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
+   "version": "3.12.4"
   }
  },
  "nbformat": 4,