Skip to content

Commit 2f5ffc4

Browse files
scottprahlscottprahl
committed
cleanup field.py
1 parent 36d5dfa commit 2f5ffc4

2 files changed

Lines changed: 224 additions & 277 deletions

File tree

‎docs/15_2D_fields.ipynb‎

Lines changed: 3 additions & 102 deletions
Original file line numberDiff line numberDiff line change
@@ -4,22 +4,7 @@
44
"cell_type": "markdown",
55
"id": "f49966c6",
66
"metadata": {},
7-
"source": [
8-
"# scattnlay vs miepython Near-Field Validation\n",
9-
"\n",
10-
"This notebook validates `miepython` near-field electric and magnetic fields against\n",
11-
"precomputed references from **scattnlay**:\n",
12-
"\n",
13-
"- Project: [scattnlay](https://github.com/ovidiopr/scattnlay)\n",
14-
"- Implementation: C++ (with Python bindings / CLI)\n",
15-
"- Scope: electromagnetic scattering and near fields for multilayered spheres\n",
16-
"\n",
17-
"Reference arrays are generated once by:\n",
18-
"\n",
19-
"`python generate_scattnlay_reference_fields.py`\n",
20-
"\n",
21-
"and stored in `docs/data/` as `.npy` files.\n"
22-
]
7+
"source": "# scattnlay vs miepython Near-Field Validation\n\nThis notebook validates `miepython` near-field electric and magnetic fields against\nprecomputed references from **scattnlay**:\n\n- Project: [scattnlay](https://github.com/ovidiopr/scattnlay)\n- Implementation: C++ (with Python bindings / CLI)\n- Scope: electromagnetic scattering and near fields for multilayered spheres\n\nReference arrays are generated by:\n\n`python docs/generate_scattnlay_reference_fields.py`\n\nand stored in `docs/data/` as `.npy` files.\n"
238
},
249
{
2510
"cell_type": "markdown",
@@ -51,91 +36,7 @@
5136
"id": "6fc2d052",
5237
"metadata": {},
5338
"outputs": [],
54-
"source": [
55-
"from pathlib import Path\n",
56-
"\n",
57-
"import matplotlib.pyplot as plt\n",
58-
"import numpy as np\n",
59-
"\n",
60-
"from miepython.field import eh_near_cartesian\n",
61-
"\n",
62-
"\n",
63-
"def find_data_dir(default=Path(\"data\")):\n",
64-
" candidates = [default, Path(\"docs\") / default]\n",
65-
" for candidate in candidates:\n",
66-
" if candidate.exists():\n",
67-
" return candidate.resolve()\n",
68-
" raise FileNotFoundError(\n",
69-
" \"Could not find reference data directory. \"\n",
70-
" \"Run: python docs/generate_scattnlay_reference_fields.py\"\n",
71-
" )\n",
72-
"\n",
73-
"\n",
74-
"def load_scattnlay_case_from_npy(case_key, data_dir=None):\n",
75-
" base = find_data_dir() if data_dir is None else Path(data_dir).resolve()\n",
76-
" paths = {\n",
77-
" \"X\": base / f\"scattnlay_{case_key}_X.npy\",\n",
78-
" \"Z\": base / f\"scattnlay_{case_key}_Z.npy\",\n",
79-
" \"E\": base / f\"scattnlay_{case_key}_E.npy\",\n",
80-
" \"H\": base / f\"scattnlay_{case_key}_H.npy\",\n",
81-
" }\n",
82-
"\n",
83-
" missing = [str(path) for path in paths.values() if not path.exists()]\n",
84-
" if missing:\n",
85-
" msg = \"Missing precomputed scattnlay files:\\n\" + \"\\n\".join(missing)\n",
86-
" msg += \"\\nRun: python docs/generate_scattnlay_reference_fields.py\"\n",
87-
" raise FileNotFoundError(msg)\n",
88-
"\n",
89-
" X = np.load(paths[\"X\"])\n",
90-
" Z = np.load(paths[\"Z\"])\n",
91-
" E = np.load(paths[\"E\"])\n",
92-
" H = np.load(paths[\"H\"])\n",
93-
" return X, Z, E, H\n",
94-
"\n",
95-
"\n",
96-
"def summarize_errors(label, Eabs_s, Eabs_m, Habs_s, Habs_m):\n",
97-
" E_rel_err = np.abs(Eabs_m - Eabs_s) / np.maximum(Eabs_s, 1e-12)\n",
98-
" H_rel_err = np.abs(Habs_m - Habs_s) / np.maximum(Habs_s, 1e-12)\n",
99-
"\n",
100-
" print(f\"[{label}] scattnlay max |E| = {np.max(Eabs_s):.6g}\")\n",
101-
" print(f\"[{label}] miepython max |E| = {np.max(Eabs_m):.6g}\")\n",
102-
" print(f\"[{label}] scattnlay max |H| (eta-scaled) = {np.max(Habs_s):.6g}\")\n",
103-
" print(f\"[{label}] miepython max |H| = {np.max(Habs_m):.6g}\")\n",
104-
" print(f\"[{label}] E relative error median/max = {np.median(E_rel_err):.3e} / {np.max(E_rel_err):.3e}\")\n",
105-
" print(f\"[{label}] H relative error median/max = {np.median(H_rel_err):.3e} / {np.max(H_rel_err):.3e}\")\n",
106-
"\n",
107-
" return E_rel_err, H_rel_err\n",
108-
"\n",
109-
"\n",
110-
"def plot_comparison(title, Xg, Zg, radius, Eabs_s, Eabs_m, E_rel_err, Habs_s, Habs_m, H_rel_err):\n",
111-
" fig, axes = plt.subplots(2, 3, figsize=(13, 8), constrained_layout=True)\n",
112-
"\n",
113-
" plots = [\n",
114-
" (Eabs_s, \"scattnlay |E|\", \"turbo\"),\n",
115-
" (Eabs_m, \"miepython |E|\", \"turbo\"),\n",
116-
" (E_rel_err, \"|E| relative error\", \"viridis\"),\n",
117-
" (Habs_s, \"scattnlay |H| (eta-scaled)\", \"magma\"),\n",
118-
" (Habs_m, \"miepython |H|\", \"magma\"),\n",
119-
" (H_rel_err, \"|H| relative error\", \"viridis\"),\n",
120-
" ]\n",
121-
"\n",
122-
" for ax, (arr, ttl, cmap) in zip(axes.flat, plots):\n",
123-
" im = ax.imshow(\n",
124-
" arr.T,\n",
125-
" origin=\"lower\",\n",
126-
" extent=(Xg.min(), Xg.max(), Zg.min(), Zg.max()),\n",
127-
" cmap=cmap,\n",
128-
" aspect=\"equal\",\n",
129-
" )\n",
130-
" circle_color = \"w\" if cmap in (\"magma\", \"viridis\") else \"k\"\n",
131-
" ax.add_patch(plt.Circle((0.0, 0.0), radius, fill=False, color=circle_color, lw=1.1))\n",
132-
" ax.set_title(ttl)\n",
133-
" ax.set_xlabel(\"x\")\n",
134-
" ax.set_ylabel(\"z\")\n",
135-
" fig.colorbar(im, ax=ax, fraction=0.046, pad=0.02)\n",
136-
"\n",
137-
" fig.suptitle(title, y=1.02)\n"
138-
]
39+
"source": "from pathlib import Path\n\nimport matplotlib.pyplot as plt\nimport numpy as np\n\nfrom miepython.field import eh_near_cartesian\n\n\ndef find_data_dir(default=Path(\"data\")):\n candidates = [default, Path(\"docs\") / default]\n for candidate in candidates:\n if candidate.exists():\n return candidate.resolve()\n raise FileNotFoundError(\n \"Could not find reference data directory. \"\n \"Run: python docs/generate_scattnlay_reference_fields.py\"\n )\n\n\ndef load_scattnlay_case_from_npy(case_key, data_dir=None):\n base = find_data_dir() if data_dir is None else Path(data_dir).resolve()\n paths = {\n \"X\": base / f\"scattnlay_{case_key}_X.npy\",\n \"Z\": base / f\"scattnlay_{case_key}_Z.npy\",\n \"E\": base / f\"scattnlay_{case_key}_E.npy\",\n \"H\": base / f\"scattnlay_{case_key}_H.npy\",\n }\n\n missing = [str(path) for path in paths.values() if not path.exists()]\n if missing:\n msg = \"Missing precomputed scattnlay files:\\n\" + \"\\n\".join(missing)\n msg += \"\\nRun: python docs/generate_scattnlay_reference_fields.py\"\n raise FileNotFoundError(msg)\n\n X = np.load(paths[\"X\"])\n Z = np.load(paths[\"Z\"])\n E = np.load(paths[\"E\"])\n H = np.load(paths[\"H\"])\n return X, Z, E, H\n\n\ndef summarize_errors(label, Eabs_s, Eabs_m, Habs_s, Habs_m):\n E_rel_err = np.abs(Eabs_m - Eabs_s) / np.maximum(Eabs_s, 1e-12)\n H_rel_err = np.abs(Habs_m - Habs_s) / np.maximum(Habs_s, 1e-12)\n\n print(f\"[{label}] scattnlay max |E| = {np.max(Eabs_s):.6g}\")\n print(f\"[{label}] miepython max |E| = {np.max(Eabs_m):.6g}\")\n print(f\"[{label}] scattnlay max |H| (eta-scaled) = {np.max(Habs_s):.6g}\")\n print(f\"[{label}] miepython max |H| = {np.max(Habs_m):.6g}\")\n print(f\"[{label}] E relative error median/max = {np.median(E_rel_err):.3e} / {np.max(E_rel_err):.3e}\")\n print(f\"[{label}] H relative error median/max = {np.median(H_rel_err):.3e} / {np.max(H_rel_err):.3e}\")\n\n return E_rel_err, H_rel_err\n\n\ndef plot_comparison(title, Xg, Zg, radius, Eabs_s, Eabs_m, E_rel_err, Habs_s, Habs_m, H_rel_err):\n fig, axes = plt.subplots(2, 3, figsize=(13, 8), constrained_layout=True)\n\n plots = [\n (Eabs_s, \"scattnlay |E|\", \"turbo\"),\n (Eabs_m, \"miepython |E|\", \"turbo\"),\n (E_rel_err, \"|E| relative error\", \"viridis\"),\n (Habs_s, \"scattnlay |H| (eta-scaled)\", \"magma\"),\n (Habs_m, \"miepython |H|\", \"magma\"),\n (H_rel_err, \"|H| relative error\", \"viridis\"),\n ]\n\n for ax, (arr, ttl, cmap) in zip(axes.flat, plots):\n im = ax.imshow(\n arr.T,\n origin=\"lower\",\n extent=(Xg.min(), Xg.max(), Zg.min(), Zg.max()),\n cmap=cmap,\n aspect=\"equal\",\n )\n circle_color = \"w\" if cmap in (\"magma\", \"viridis\") else \"k\"\n ax.add_patch(plt.Circle((0.0, 0.0), radius, fill=False, color=circle_color, lw=1.1))\n ax.set_title(ttl)\n ax.set_xlabel(\"x\")\n ax.set_ylabel(\"z\")\n fig.colorbar(im, ax=ax, fraction=0.046, pad=0.02)\n\n fig.suptitle(title, y=1.02)\n"
13940
},
14041
{
14142
"cell_type": "markdown",
@@ -333,4 +234,4 @@
333234
},
334235
"nbformat": 4,
335236
"nbformat_minor": 5
336-
}
237+
}

0 commit comments

Comments
 (0)