setupEM 0.4.0__tar.gz → 0.5.0__tar.gz

{setupem-0.4.0/src/setupEM.egg-info → setupem-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: setupEM
-Version: 0.4.0
+Version: 0.5.0
 Summary: Python tool for configuration of gds2palace workflow with GUI.
 Author-email: Volker Muehlhaus <volker@muehlhaus.com>
 License-Expression: GPL-3.0-or-later
@@ -21,10 +21,14 @@ Dynamic: license-file
 
 ## What's New
 
-- A built-in **Result Viewer** (Create Model tab > View Results...) to browse and plot Touchstone S-parameter results without leaving setupEM
-- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
+setupEM includes two built-in tools for working with simulation results directly, without external scripts:
+
+- **Result Viewer** (Create Model tab > View Results...) plots Touchstone S-parameter results - dB/phase, Smith chart, zoomed Smith chart - without leaving setupEM. See section "[Result Viewer](#result-viewer)".
+- **Model Fit** (Create Model tab > Model Fit...) launches [snp2le](https://github.com/iic-jku/snp2le), an external open-source tool that extracts a lumped-element SPICE/Spectre netlist from S-parameter results - offering to install it via pip if it isn't already present. See section "[Model Fit](#model-fit)".
+
 - A graphical **Stackup XML Editor** (Tools > Edit Stackup XML...), including Variables, Reference-relative positioning, Derived Layers, and Thermal Tables
 - Input Files tab can now **override stackup Variables** (e.g. `total_thickness`) directly, without hand-editing the XML or the generated model
+- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
 
 See [CHANGES.md](https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/CHANGES.md) for details.
 
@@ -195,6 +199,8 @@ The buttons are used top down: You can first preview the resulting model geometr
 
 To start simulation, use the "Run palace" button. If you are on Linux, this will start Palace using script "run_sim". If you are on Windows, this will start the Linux Subsystem for Windows (WSL) and open a command prompt in the simulation directory.
 
+If the output directory already has results from a previous run, you are asked whether to delete or keep them before starting - default is to delete, so a rerun with different settings doesn't leave stale results mixed in with the new ones.
+
 <img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/createmodel3.png" alt="create" width="700">
 
 To start simulation on Linux, it is required that you have configured a script "run_sim" as described in the [gds2palace documentation](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/blob/main/doc/gds2palace_workflow_userguide.pdf). You can find a template [here](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/tree/main/scripts) in the gds2palace repository.
@@ -230,6 +236,24 @@ The matplotlib toolbar above the plot (pan/zoom/save as PNG) works as usual. A f
 
 Result Viewer can also be run standalone, without the full setupEM GUI, either directly (`python result_viewer.py [target_dir]`) or via the `resultViewer` console script installed with the package.
 
+## Model Fit
+
+Click **Model Fit...** on the Create Model tab (next to **View Results...**) to extract a lumped-element netlist from the current run's S-parameter result, using [snp2le](https://github.com/iic-jku/snp2le) - an external, open-source tool, not part of setupEM.
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/resultviewer_button.png" alt="model fit button" width="700">
+
+If snp2le isn't installed, setupEM offers to install it for you via pip:
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/modelfit1.png" alt="snp2le not installed" width="350">
+
+Choosing **Install** runs `pip install snp2le` in the Log panel and, once it succeeds, continues automatically - no need to click Model Fit a second time. Choosing **Cancel** logs the manual install command and the project link instead.
+
+Once snp2le is available, Model Fit locates the raw (not `_dc`, not `_deembedded`) Touchstone result file for the current run and launches the snp2le GUI in that file's directory. snp2le's GUI has no command-line option to preload a file, so the exact path is printed to the Log panel - load it via snp2le's own file picker:
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/modelfit2.png" alt="snp2le starting" width="700">
+
+If no raw result file exists yet (no simulation has been run), Model Fit shows a warning instead of starting snp2le - run a simulation first.
+
 ## Code 
 Behind the scenes, the setupEM user interface created Python model code for gds2palace, and you can check the resulting code on the "Code" tab.
 

{setupem-0.4.0 → setupem-0.5.0}/README.md

@@ -2,10 +2,14 @@
 
 ## What's New
 
-- A built-in **Result Viewer** (Create Model tab > View Results...) to browse and plot Touchstone S-parameter results without leaving setupEM
-- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
+setupEM includes two built-in tools for working with simulation results directly, without external scripts:
+
+- **Result Viewer** (Create Model tab > View Results...) plots Touchstone S-parameter results - dB/phase, Smith chart, zoomed Smith chart - without leaving setupEM. See section "[Result Viewer](#result-viewer)".
+- **Model Fit** (Create Model tab > Model Fit...) launches [snp2le](https://github.com/iic-jku/snp2le), an external open-source tool that extracts a lumped-element SPICE/Spectre netlist from S-parameter results - offering to install it via pip if it isn't already present. See section "[Model Fit](#model-fit)".
+
 - A graphical **Stackup XML Editor** (Tools > Edit Stackup XML...), including Variables, Reference-relative positioning, Derived Layers, and Thermal Tables
 - Input Files tab can now **override stackup Variables** (e.g. `total_thickness`) directly, without hand-editing the XML or the generated model
+- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
 
 See [CHANGES.md](doc/CHANGES.md) for details.
 
@@ -176,6 +180,8 @@ The buttons are used top down: You can first preview the resulting model geometr
 
 To start simulation, use the "Run palace" button. If you are on Linux, this will start Palace using script "run_sim". If you are on Windows, this will start the Linux Subsystem for Windows (WSL) and open a command prompt in the simulation directory.
 
+If the output directory already has results from a previous run, you are asked whether to delete or keep them before starting - default is to delete, so a rerun with different settings doesn't leave stale results mixed in with the new ones.
+
 <img src="./doc/png/createmodel3.png" alt="create" width="700">
 
 To start simulation on Linux, it is required that you have configured a script "run_sim" as described in the [gds2palace documentation](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/blob/main/doc/gds2palace_workflow_userguide.pdf). You can find a template [here](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/tree/main/scripts) in the gds2palace repository.
@@ -211,6 +217,24 @@ The matplotlib toolbar above the plot (pan/zoom/save as PNG) works as usual. A f
 
 Result Viewer can also be run standalone, without the full setupEM GUI, either directly (`python result_viewer.py [target_dir]`) or via the `resultViewer` console script installed with the package.
 
+## Model Fit
+
+Click **Model Fit...** on the Create Model tab (next to **View Results...**) to extract a lumped-element netlist from the current run's S-parameter result, using [snp2le](https://github.com/iic-jku/snp2le) - an external, open-source tool, not part of setupEM.
+
+<img src="./doc/png/resultviewer_button.png" alt="model fit button" width="700">
+
+If snp2le isn't installed, setupEM offers to install it for you via pip:
+
+<img src="./doc/png/modelfit1.png" alt="snp2le not installed" width="350">
+
+Choosing **Install** runs `pip install snp2le` in the Log panel and, once it succeeds, continues automatically - no need to click Model Fit a second time. Choosing **Cancel** logs the manual install command and the project link instead.
+
+Once snp2le is available, Model Fit locates the raw (not `_dc`, not `_deembedded`) Touchstone result file for the current run and launches the snp2le GUI in that file's directory. snp2le's GUI has no command-line option to preload a file, so the exact path is printed to the Log panel - load it via snp2le's own file picker:
+
+<img src="./doc/png/modelfit2.png" alt="snp2le starting" width="700">
+
+If no raw result file exists yet (no simulation has been run), Model Fit shows a warning instead of starting snp2le - run a simulation first.
+
 ## Code 
 Behind the scenes, the setupEM user interface created Python model code for gds2palace, and you can check the resulting code on the "Code" tab.
 

{setupem-0.4.0 → setupem-0.5.0}/README_pypi.md

@@ -2,10 +2,14 @@
 
 ## What's New
 
-- A built-in **Result Viewer** (Create Model tab > View Results...) to browse and plot Touchstone S-parameter results without leaving setupEM
-- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
+setupEM includes two built-in tools for working with simulation results directly, without external scripts:
+
+- **Result Viewer** (Create Model tab > View Results...) plots Touchstone S-parameter results - dB/phase, Smith chart, zoomed Smith chart - without leaving setupEM. See section "[Result Viewer](#result-viewer)".
+- **Model Fit** (Create Model tab > Model Fit...) launches [snp2le](https://github.com/iic-jku/snp2le), an external open-source tool that extracts a lumped-element SPICE/Spectre netlist from S-parameter results - offering to install it via pip if it isn't already present. See section "[Model Fit](#model-fit)".
+
 - A graphical **Stackup XML Editor** (Tools > Edit Stackup XML...), including Variables, Reference-relative positioning, Derived Layers, and Thermal Tables
 - Input Files tab can now **override stackup Variables** (e.g. `total_thickness`) directly, without hand-editing the XML or the generated model
+- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
 
 See [CHANGES.md](https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/CHANGES.md) for details.
 
@@ -176,6 +180,8 @@ The buttons are used top down: You can first preview the resulting model geometr
 
 To start simulation, use the "Run palace" button. If you are on Linux, this will start Palace using script "run_sim". If you are on Windows, this will start the Linux Subsystem for Windows (WSL) and open a command prompt in the simulation directory.
 
+If the output directory already has results from a previous run, you are asked whether to delete or keep them before starting - default is to delete, so a rerun with different settings doesn't leave stale results mixed in with the new ones.
+
 <img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/createmodel3.png" alt="create" width="700">
 
 To start simulation on Linux, it is required that you have configured a script "run_sim" as described in the [gds2palace documentation](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/blob/main/doc/gds2palace_workflow_userguide.pdf). You can find a template [here](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/tree/main/scripts) in the gds2palace repository.
@@ -211,6 +217,24 @@ The matplotlib toolbar above the plot (pan/zoom/save as PNG) works as usual. A f
 
 Result Viewer can also be run standalone, without the full setupEM GUI, either directly (`python result_viewer.py [target_dir]`) or via the `resultViewer` console script installed with the package.
 
+## Model Fit
+
+Click **Model Fit...** on the Create Model tab (next to **View Results...**) to extract a lumped-element netlist from the current run's S-parameter result, using [snp2le](https://github.com/iic-jku/snp2le) - an external, open-source tool, not part of setupEM.
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/resultviewer_button.png" alt="model fit button" width="700">
+
+If snp2le isn't installed, setupEM offers to install it for you via pip:
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/modelfit1.png" alt="snp2le not installed" width="350">
+
+Choosing **Install** runs `pip install snp2le` in the Log panel and, once it succeeds, continues automatically - no need to click Model Fit a second time. Choosing **Cancel** logs the manual install command and the project link instead.
+
+Once snp2le is available, Model Fit locates the raw (not `_dc`, not `_deembedded`) Touchstone result file for the current run and launches the snp2le GUI in that file's directory. snp2le's GUI has no command-line option to preload a file, so the exact path is printed to the Log panel - load it via snp2le's own file picker:
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/modelfit2.png" alt="snp2le starting" width="700">
+
+If no raw result file exists yet (no simulation has been run), Model Fit shows a warning instead of starting snp2le - run a simulation first.
+
 ## Code 
 Behind the scenes, the setupEM user interface created Python model code for gds2palace, and you can check the resulting code on the "Code" tab.
 

{setupem-0.4.0 → setupem-0.5.0}/src/setupEM/__init__.py

@@ -24,4 +24,4 @@ A Python tool for EM setup using gds2palace.
 from .setupEM import main
 
 __all__ = ["main"]
-__version__ = "0.4.0"
+__version__ = "0.5.0"

{setupem-0.4.0 → setupem-0.5.0}/src/setupEM/palace_results.py

@@ -1,199 +1,199 @@
-########################################################################
-#
-# Copyright 2025 Volker Muehlhaus and IHP PDK Authors
-#
-# Licensed under the GNU General Public License, Version 3.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#    https://www.gnu.org/licenses/gpl-3.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-########################################################################
-
-"""Parse AWS Palace solver output (palace.json / error-indicators.csv) into a
-human-readable summary for the setupEM "Create Model" log. No Qt dependency,
-so this can be exercised standalone against real Palace output directories.
-"""
-
-import os
-import json
-import csv
-import re
-
-_ITERATION_RE = re.compile(r'^iteration(\d+)$')
-
-
-def _find_output_dir(run_path, model_basename):
-    # config.json's Problem.Output is a path relative to config.json's own directory
-    config_path = os.path.join(run_path, 'config.json')
-    output_rel = None
-    if os.path.isfile(config_path):
-        try:
-            with open(config_path, 'r', encoding='utf-8') as f:
-                config = json.load(f)
-            output_rel = config.get('Problem', {}).get('Output')
-        except (OSError, json.JSONDecodeError):
-            output_rel = None
-    if not output_rel:
-        output_rel = 'output/' + model_basename
-    return os.path.normpath(os.path.join(run_path, output_rel))
-
-
-def _read_palace_json(dir_path):
-    path = os.path.join(dir_path, 'palace.json')
-    if not os.path.isfile(path):
-        return None
-    try:
-        with open(path, 'r', encoding='utf-8') as f:
-            data = json.load(f)
-    except (OSError, json.JSONDecodeError):
-        return None
-    return {
-        'duration_s': data.get('ElapsedTime', {}).get('Durations', {}).get('Total'),
-        'peak_ram_mb': data.get('PeakMemoryMegabytes', {}).get('Total'),
-        'dof': data.get('Problem', {}).get('DegreesOfFreedom'),
-        'mesh_elements': data.get('Problem', {}).get('MeshElements'),
-    }
-
-
-def _read_error_indicators(dir_path):
-    path = os.path.join(dir_path, 'error-indicators.csv')
-    if not os.path.isfile(path):
-        return None
-    try:
-        with open(path, 'r', encoding='utf-8', newline='') as f:
-            rows = [row for row in csv.reader(f, skipinitialspace=True) if row]
-    except OSError:
-        return None
-    if len(rows) < 2:
-        return None
-    headers = [h.strip() for h in rows[0]]
-    values = [v.strip() for v in rows[1]]
-    fields = dict(zip(headers, values))
-    try:
-        return {
-            'norm': float(fields['Norm']),
-            'minimum': float(fields['Minimum']),
-            'maximum': float(fields['Maximum']),
-            'mean': float(fields['Mean']),
-        }
-    except (KeyError, ValueError):
-        return None
-
-
-def _list_iteration_dirs(output_dir):
-    if not os.path.isdir(output_dir):
-        return []
-    found = []
-    for name in os.listdir(output_dir):
-        match = _ITERATION_RE.match(name)
-        full_path = os.path.join(output_dir, name)
-        if match and os.path.isdir(full_path):
-            found.append((int(match.group(1)), full_path))
-    found.sort(key=lambda item: item[0])
-    return [full_path for _, full_path in found]
-
-
-def _format_duration(seconds):
-    if seconds is None:
-        return 'n/a'
-    if seconds < 60:
-        return f"{seconds:.1f} s"
-    minutes, secs = divmod(int(round(seconds)), 60)
-    hours, minutes = divmod(minutes, 60)
-    if hours:
-        return f"{hours}h {minutes}m {secs}s"
-    return f"{minutes}m {secs}s"
-
-
-def _format_ram(mb):
-    if mb is None:
-        return 'n/a'
-    if mb >= 1024:
-        return f"{mb / 1024:.2f} GB"
-    return f"{mb:.1f} MB"
-
-
-def _format_int(n):
-    return 'n/a' if n is None else f"{n:,}"
-
-
-def _format_sci(x):
-    return 'n/a' if x is None else f"{x:.3e}"
-
-
-def build_results_summary(run_path, model_basename):
-    """Return a formatted multi-line summary of Palace results found under run_path,
-    or an explanatory message if nothing is there yet.
-    """
-    output_dir = _find_output_dir(run_path, model_basename)
-    if not os.path.isdir(output_dir):
-        return (f"No Palace results found yet (expected output directory: {output_dir}) "
-                 "-- has the simulation finished?")
-
-    iteration_dirs = _list_iteration_dirs(output_dir)
-
-    if not iteration_dirs:
-        summary = _read_palace_json(output_dir)
-        errors = _read_error_indicators(output_dir)
-        if summary is None:
-            return (f"No palace.json found yet in {output_dir} "
-                     "-- has the simulation finished?")
-        lines = [
-            f"=== Simulation results: {model_basename} ===",
-            f"Degrees of freedom : {_format_int(summary['dof'])}",
-            f"Mesh elements      : {_format_int(summary['mesh_elements'])}",
-            f"Simulation time    : {_format_duration(summary['duration_s'])}",
-            f"Peak RAM           : {_format_ram(summary['peak_ram_mb'])}",
-        ]
-        if errors:
-            lines.append(
-                f"Error indicator    : Norm={_format_sci(errors['norm'])}  "
-                f"Max={_format_sci(errors['maximum'])}  Mean={_format_sci(errors['mean'])}"
-            )
-        lines.append("=" * 40)
-        return "\n".join(lines)
-
-    # Adaptive mesh refinement: one row per iteration subfolder, plus the root as "Final"
-    rows = [(os.path.basename(it_dir), _read_palace_json(it_dir), _read_error_indicators(it_dir))
-            for it_dir in iteration_dirs]
-    rows.append(("Final", _read_palace_json(output_dir), _read_error_indicators(output_dir)))
-
-    headers = ["Iteration", "DOF", "Mesh elems", "Error Norm", "Error Max", "Error Mean", "Time", "Peak RAM"]
-    table_rows = []
-    for label, summary, errors in rows:
-        summary = summary or {}
-        errors = errors or {}
-        table_rows.append([
-            label,
-            _format_int(summary.get('dof')),
-            _format_int(summary.get('mesh_elements')),
-            _format_sci(errors.get('norm')),
-            _format_sci(errors.get('maximum')),
-            _format_sci(errors.get('mean')),
-            _format_duration(summary.get('duration_s')),
-            _format_ram(summary.get('peak_ram_mb')),
-        ])
-
-    widths = [max(len(headers[i]), *(len(r[i]) for r in table_rows)) for i in range(len(headers))]
-
-    def fmt_row(cells):
-        return " | ".join(
-            cell.ljust(widths[i]) if i == 0 else cell.rjust(widths[i])
-            for i, cell in enumerate(cells)
-        )
-
-    lines = [f"=== Simulation results: {model_basename} (adaptive mesh refinement) ==="]
-    header_line = fmt_row(headers)
-    lines.append(header_line)
-    lines.append("-+-".join("-" * w for w in widths))
-    for row in table_rows:
-        lines.append(fmt_row(row))
-    lines.append("=" * len(header_line))
-    return "\n".join(lines)
+########################################################################
+#
+# Copyright 2025 Volker Muehlhaus and IHP PDK Authors
+#
+# Licensed under the GNU General Public License, Version 3.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    https://www.gnu.org/licenses/gpl-3.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+########################################################################
+
+"""Parse AWS Palace solver output (palace.json / error-indicators.csv) into a
+human-readable summary for the setupEM "Create Model" log. No Qt dependency,
+so this can be exercised standalone against real Palace output directories.
+"""
+
+import os
+import json
+import csv
+import re
+
+_ITERATION_RE = re.compile(r'^iteration(\d+)$')
+
+
+def find_output_dir(run_path, model_basename):
+    # config.json's Problem.Output is a path relative to config.json's own directory
+    config_path = os.path.join(run_path, 'config.json')
+    output_rel = None
+    if os.path.isfile(config_path):
+        try:
+            with open(config_path, 'r', encoding='utf-8') as f:
+                config = json.load(f)
+            output_rel = config.get('Problem', {}).get('Output')
+        except (OSError, json.JSONDecodeError):
+            output_rel = None
+    if not output_rel:
+        output_rel = 'output/' + model_basename
+    return os.path.normpath(os.path.join(run_path, output_rel))
+
+
+def _read_palace_json(dir_path):
+    path = os.path.join(dir_path, 'palace.json')
+    if not os.path.isfile(path):
+        return None
+    try:
+        with open(path, 'r', encoding='utf-8') as f:
+            data = json.load(f)
+    except (OSError, json.JSONDecodeError):
+        return None
+    return {
+        'duration_s': data.get('ElapsedTime', {}).get('Durations', {}).get('Total'),
+        'peak_ram_mb': data.get('PeakMemoryMegabytes', {}).get('Total'),
+        'dof': data.get('Problem', {}).get('DegreesOfFreedom'),
+        'mesh_elements': data.get('Problem', {}).get('MeshElements'),
+    }
+
+
+def _read_error_indicators(dir_path):
+    path = os.path.join(dir_path, 'error-indicators.csv')
+    if not os.path.isfile(path):
+        return None
+    try:
+        with open(path, 'r', encoding='utf-8', newline='') as f:
+            rows = [row for row in csv.reader(f, skipinitialspace=True) if row]
+    except OSError:
+        return None
+    if len(rows) < 2:
+        return None
+    headers = [h.strip() for h in rows[0]]
+    values = [v.strip() for v in rows[1]]
+    fields = dict(zip(headers, values))
+    try:
+        return {
+            'norm': float(fields['Norm']),
+            'minimum': float(fields['Minimum']),
+            'maximum': float(fields['Maximum']),
+            'mean': float(fields['Mean']),
+        }
+    except (KeyError, ValueError):
+        return None
+
+
+def _list_iteration_dirs(output_dir):
+    if not os.path.isdir(output_dir):
+        return []
+    found = []
+    for name in os.listdir(output_dir):
+        match = _ITERATION_RE.match(name)
+        full_path = os.path.join(output_dir, name)
+        if match and os.path.isdir(full_path):
+            found.append((int(match.group(1)), full_path))
+    found.sort(key=lambda item: item[0])
+    return [full_path for _, full_path in found]
+
+
+def _format_duration(seconds):
+    if seconds is None:
+        return 'n/a'
+    if seconds < 60:
+        return f"{seconds:.1f} s"
+    minutes, secs = divmod(int(round(seconds)), 60)
+    hours, minutes = divmod(minutes, 60)
+    if hours:
+        return f"{hours}h {minutes}m {secs}s"
+    return f"{minutes}m {secs}s"
+
+
+def _format_ram(mb):
+    if mb is None:
+        return 'n/a'
+    if mb >= 1024:
+        return f"{mb / 1024:.2f} GB"
+    return f"{mb:.1f} MB"
+
+
+def _format_int(n):
+    return 'n/a' if n is None else f"{n:,}"
+
+
+def _format_sci(x):
+    return 'n/a' if x is None else f"{x:.3e}"
+
+
+def build_results_summary(run_path, model_basename):
+    """Return a formatted multi-line summary of Palace results found under run_path,
+    or an explanatory message if nothing is there yet.
+    """
+    output_dir = find_output_dir(run_path, model_basename)
+    if not os.path.isdir(output_dir):
+        return (f"No Palace results found yet (expected output directory: {output_dir}) "
+                 "-- has the simulation finished?")
+
+    iteration_dirs = _list_iteration_dirs(output_dir)
+
+    if not iteration_dirs:
+        summary = _read_palace_json(output_dir)
+        errors = _read_error_indicators(output_dir)
+        if summary is None:
+            return (f"No palace.json found yet in {output_dir} "
+                     "-- has the simulation finished?")
+        lines = [
+            f"=== Simulation results: {model_basename} ===",
+            f"Degrees of freedom : {_format_int(summary['dof'])}",
+            f"Mesh elements      : {_format_int(summary['mesh_elements'])}",
+            f"Simulation time    : {_format_duration(summary['duration_s'])}",
+            f"Peak RAM           : {_format_ram(summary['peak_ram_mb'])}",
+        ]
+        if errors:
+            lines.append(
+                f"Error indicator    : Norm={_format_sci(errors['norm'])}  "
+                f"Max={_format_sci(errors['maximum'])}  Mean={_format_sci(errors['mean'])}"
+            )
+        lines.append("=" * 40)
+        return "\n".join(lines)
+
+    # Adaptive mesh refinement: one row per iteration subfolder, plus the root as "Final"
+    rows = [(os.path.basename(it_dir), _read_palace_json(it_dir), _read_error_indicators(it_dir))
+            for it_dir in iteration_dirs]
+    rows.append(("Final", _read_palace_json(output_dir), _read_error_indicators(output_dir)))
+
+    headers = ["Iteration", "DOF", "Mesh elems", "Error Norm", "Error Max", "Error Mean", "Time", "Peak RAM"]
+    table_rows = []
+    for label, summary, errors in rows:
+        summary = summary or {}
+        errors = errors or {}
+        table_rows.append([
+            label,
+            _format_int(summary.get('dof')),
+            _format_int(summary.get('mesh_elements')),
+            _format_sci(errors.get('norm')),
+            _format_sci(errors.get('maximum')),
+            _format_sci(errors.get('mean')),
+            _format_duration(summary.get('duration_s')),
+            _format_ram(summary.get('peak_ram_mb')),
+        ])
+
+    widths = [max(len(headers[i]), *(len(r[i]) for r in table_rows)) for i in range(len(headers))]
+
+    def fmt_row(cells):
+        return " | ".join(
+            cell.ljust(widths[i]) if i == 0 else cell.rjust(widths[i])
+            for i, cell in enumerate(cells)
+        )
+
+    lines = [f"=== Simulation results: {model_basename} (adaptive mesh refinement) ==="]
+    header_line = fmt_row(headers)
+    lines.append(header_line)
+    lines.append("-+-".join("-" * w for w in widths))
+    for row in table_rows:
+        lines.append(fmt_row(row))
+    lines.append("=" * len(header_line))
+    return "\n".join(lines)

{setupem-0.4.0 → setupem-0.5.0}/src/setupEM/setupEM.py

@@ -17,9 +17,10 @@
 ########################################################################
 
 
-import sys, json, os, pathlib, ast, webbrowser, argparse
+import sys, json, os, pathlib, ast, webbrowser, argparse, shutil, re
 import numpy as np
 import importlib.metadata
+import importlib.util
 import requests
 import gdspy
 from scipy.interpolate import interp1d
@@ -54,7 +55,7 @@ if __package__ in (None, ""):
         VectorWidget, PopUpWindow, CreateModelTabBase, MainWindowBase,
         epsilon_to_color, default_stackup_dielectric_label, default_stackup_metal_label,
     )
-    from palace_results import build_results_summary
+    from palace_results import build_results_summary, find_output_dir
 else:
     from .setup_common import (
         EDIT_STYLE_OPTIONAL, EDIT_STYLE_REQUIRED, COMBO_STYLE_REQUIRED, COMBO_STYLE_OPTIONAL,
@@ -62,7 +63,7 @@ else:
         VectorWidget, PopUpWindow, CreateModelTabBase, MainWindowBase,
         epsilon_to_color, default_stackup_dielectric_label, default_stackup_metal_label,
     )
-    from .palace_results import build_results_summary
+    from .palace_results import build_results_summary, find_output_dir
 
 
 '''
@@ -1126,11 +1127,85 @@ class CreateModelTab(CreateModelTabBase):
         # simulation run apart from a mesh-creation run (self.process is reused for both).
         self._process_purpose = None
 
-        # S-parameter result viewer: appended here (not in the shared CreateModelTabBase)
-        # since setupThermal has no S-parameters and must not show this button.
+        # S-parameter result viewer + model fit: appended here (not in the shared
+        # CreateModelTabBase) since setupThermal has no S-parameters and must not
+        # show these buttons. Added as a row of the base class's buttons_grid
+        # (not a separate layout) so this row's two-thirds/one-third split lines
+        # up exactly with Preview/Create Mesh/Start Simulation above, in the same
+        # Actions group.
+        row = self.buttons_grid.rowCount()
         self.view_results_btn = QPushButton("📈 View Results...")
         self.view_results_btn.clicked.connect(self.MainWindow.open_result_viewer)
-        self.file_layout.addWidget(self.view_results_btn)
+        self.buttons_grid.addWidget(self.view_results_btn, row, 0)
+        self.model_fit_btn = QPushButton("🧩 Model Fit...")
+        self.model_fit_btn.clicked.connect(self.open_model_fit)
+        self.buttons_grid.addWidget(self.model_fit_btn, row, 1)
+
+    def open_model_fit(self):
+        SNP2LE_URL = "https://github.com/iic-jku/snp2le"
+
+        # invalidate_caches(): after a just-completed "pip install snp2le" run (see
+        # on_finished() below), the import system's directory-listing cache for
+        # site-packages can still be stale in this same process, so a find_spec()
+        # right after install could still report "not found" without this.
+        importlib.invalidate_caches()
+        if importlib.util.find_spec("snp2le") is None:
+            box = QMessageBox(self)
+            box.setIcon(QMessageBox.Question)
+            box.setWindowTitle("Model Fit")
+            box.setText("snp2le is not installed. Install it now with pip?")
+            install_btn = box.addButton("Install", QMessageBox.AcceptRole)
+            box.addButton("Cancel", QMessageBox.RejectRole)
+            box.setDefaultButton(install_btn)
+            box.exec()
+
+            if box.clickedButton() is install_btn:
+                self.log_area.appendPlainText("Installing snp2le with pip ...")
+                self._process_purpose = "install_snp2le"
+                self.process.start(sys.executable, ["-m", "pip", "install", "snp2le"])
+            else:
+                self.log_area.appendPlainText("⚠️ snp2le is not installed. Install it with: pip install snp2le")
+                self.log_area.appendPlainText(SNP2LE_URL)
+            return
+
+        # local import: only needed once snp2le is actually launched, same lazy-import
+        # style used by open_result_viewer() for its own heavy imports.
+        if __package__ in (None, ""):
+            from result_viewer import find_touchstone_files
+        else:
+            from .result_viewer import find_touchstone_files
+
+        if self.MainWindow.PalaceMode:
+            run_path = saved_values['sim_path'] + "/palace_model/" + saved_values['model_basename'] + "_data"
+        else:
+            run_path = saved_values['sim_path'] + "/elmer_model/" + saved_values['model_basename'] + "_data"
+
+        raw_files = [
+            path for path in find_touchstone_files(run_path)
+            if '_dc' not in os.path.basename(path) and '_deembedded' not in os.path.basename(path)
+        ]
+
+        if not raw_files:
+            QMessageBox.warning(self, "Model Fit",
+                                 f"No raw S-parameter result file found in {run_path}.\nRun a simulation first.")
+            return
+        if len(raw_files) > 1:
+            self.log_area.appendPlainText("⚠️ Multiple raw S-parameter result files found, using the first one:")
+            for path in raw_files:
+                self.log_area.appendPlainText("  " + path)
+        raw_file = raw_files[0]
+
+        # snp2le's GUI mode takes no filename argument (only its "-b" batch mode does) -
+        # the file has to be loaded by hand via the GUI's own file picker, so point the
+        # user at it explicitly and start snp2le in that file's directory, so its file
+        # picker opens there instead of wherever setupEM happened to be launched from.
+        raw_dir = os.path.dirname(raw_file)
+        self.log_area.appendPlainText("Starting snp2le ...")
+        self.log_area.appendPlainText(SNP2LE_URL)
+        self.log_area.appendPlainText(f"In the snp2le window, load: {raw_file}")
+        self._process_purpose = "model_fit"
+        self.process.setWorkingDirectory(raw_dir)
+        self.process.start(sys.executable, ["-m", "snp2le"])
 
     def _append_results_summary(self):
         # Palace-only: parse palace.json / error-indicators.csv and append a results summary
@@ -1144,6 +1219,18 @@ class CreateModelTab(CreateModelTabBase):
         # Auto-append the results summary after a real simulation run (not after mesh creation)
         if self._process_purpose == "run_simulation" and self.MainWindow.PalaceMode:
             self._append_results_summary()
+        elif self._process_purpose == "install_snp2le":
+            importlib.invalidate_caches()
+            if exit_code == 0 and importlib.util.find_spec("snp2le") is not None:
+                self.log_area.appendPlainText("snp2le installed successfully.\n")
+                # re-enter open_model_fit(): this time find_spec() succeeds, so it
+                # goes straight to finding the raw result file and launching snp2le.
+                self.open_model_fit()
+            else:
+                self.log_area.appendPlainText(
+                    f"⚠️ snp2le installation failed (exit code {exit_code}). "
+                    f"Try manually: pip install snp2le\n"
+                )
 
     def create_model(self):
         # Request all tabs to save values again,
@@ -1180,9 +1267,64 @@ class CreateModelTab(CreateModelTabBase):
 
 
 
+    def _confirm_clear_previous_results(self, run_path):
+        """If run_path already holds solver output from a previous run, ask the user
+        (default: delete) whether to remove it before starting a new simulation. Only
+        ever touches solver-OUTPUT data - never the mesh/config/run script that "Create
+        Mesh" just wrote, since those are still needed to start this run.
+
+        Palace writes all of its results into one dedicated subdirectory (named in
+        config.json's Problem.Output, resolved via find_output_dir()), so that whole
+        subtree is the deletion target. Elmer has no such subdirectory - its solver
+        writes scalar_results.*/fields*.vtu*/*.sNp files directly into run_path
+        alongside the mesh/config - so those are matched and removed individually.
+        """
+        if self.MainWindow.PalaceMode:
+            output_dir = find_output_dir(run_path, saved_values['model_basename'])
+            if not os.path.isdir(output_dir) or not os.listdir(output_dir):
+                return
+            targets = [output_dir]
+        else:
+            targets = []
+            if os.path.isdir(run_path):
+                for fn in os.listdir(run_path):
+                    full_path = os.path.join(run_path, fn)
+                    if not os.path.isfile(full_path):
+                        continue
+                    if fn in ("scalar_results.dat", "scalar_results.names") or \
+                       fn.startswith("fields") or re.search(r'\.s\d+p$', fn, re.IGNORECASE):
+                        targets.append(full_path)
+            if not targets:
+                return
+
+        box = QMessageBox(self)
+        box.setIcon(QMessageBox.Question)
+        box.setWindowTitle("Previous simulation results found")
+        box.setText(
+            f"This run's output directory already contains results from a previous "
+            f"simulation:\n\n{run_path}\n\nDelete the existing results before starting?"
+        )
+        delete_btn = box.addButton("Delete", QMessageBox.AcceptRole)
+        box.addButton("Keep", QMessageBox.RejectRole)
+        box.setDefaultButton(delete_btn)
+        box.exec()
+
+        if box.clickedButton() is delete_btn:
+            for target in targets:
+                if os.path.isdir(target):
+                    shutil.rmtree(target)
+                else:
+                    os.remove(target)
+
     def run_model(self):
         # Run model that we created before
 
+        if self.MainWindow.PalaceMode:
+            run_path = saved_values['sim_path'] + "/palace_model/" + saved_values['model_basename'] + "_data"
+        else:
+            run_path = saved_values['sim_path'] + "/elmer_model/" + saved_values['model_basename'] + "_data"
+        self._confirm_clear_previous_results(run_path)
+
         # clear log
         self.log_area.clear()
         self._process_purpose = "run_simulation"
@@ -1629,14 +1771,25 @@ class MainWindow(MainWindowBase):
     def show_version(self):
         setupEM_version = importlib.metadata.version("setupEM")
         gds2palace_version = importlib.metadata.version("gds2palace")
-        version_info = f"Installed:\nsetupEM {setupEM_version}\ngds2palace {gds2palace_version}"
+        # snp2le (used by Model Fit) is an optional dependency, not installed by default
+        try:
+            snp2le_version = importlib.metadata.version("snp2le")
+        except importlib.metadata.PackageNotFoundError:
+            snp2le_version = "not installed"
+        version_info = f"Installed:\nsetupEM {setupEM_version}\ngds2palace {gds2palace_version}\nsnp2le {snp2le_version}"
 
         # get latest available version information
         latest_setupEM = self.get_latest_version("setupEM")
         latest_gds2palace = self.get_latest_version("gds2palace")
         latest_info = f"Latest version:\nsetupEM {latest_setupEM}\ngds2palace : {latest_gds2palace}"
+        upgrade_info = "\n\nYou can update using\n  pip install gds2palace --upgrade\n  pip install setupEM --upgrade"
+        if snp2le_version != "not installed":
+            # only worth checking/offering an upgrade for a package the user actually has
+            latest_snp2le = self.get_latest_version("snp2le")
+            latest_info += f"\nsnp2le {latest_snp2le}"
+            upgrade_info += "\n  pip install snp2le --upgrade"
+        upgrade_info += '\nafter exiting this program\n'    
         version_info = version_info + '\n\n' + latest_info
-        upgrade_info = "\n\nYou can update using\n  pip install gds2palace --upgrade\n  pip install setupEM --upgrade\nafter exiting this program"
 
         QMessageBox.information(self,"Version information",version_info + upgrade_info)
 

{setupem-0.4.0 → setupem-0.5.0}/src/setupEM/setup_common.py

@@ -39,7 +39,7 @@ import requests
 import gdspy
 from scipy.interpolate import interp1d
 from PySide6.QtWidgets import (
-    QMainWindow, QWidget, QHBoxLayout, QVBoxLayout,
+    QMainWindow, QWidget, QHBoxLayout, QVBoxLayout, QGridLayout,
     QLabel, QLineEdit, QComboBox,
     QPushButton, QFileDialog, QMessageBox, QGroupBox,
     QCheckBox, QPlainTextEdit, QDialog, QSizePolicy, QFrame,
@@ -1208,42 +1208,49 @@ class CreateModelTabBase(QWidget):
         self.label1.setFixedWidth(120)
         self.modelname_layout.addWidget(self.label1)
         self.modelname_edit = QLineEdit("")
-        self.modelname_edit.setFixedWidth(250)
         self.modelname_edit.setStyleSheet(EDIT_STYLE_OPTIONAL)
         # install event filter, so we capture when edit looses focus
         self.modelname_edit.editingFinished.connect(self.on_modelname_edit_done)
-
         self.modelname_layout.addWidget(self.modelname_edit)
-        self.modelname_layout.addStretch()
         self.file_layout.addLayout(self.modelname_layout)
 
+        self.file_group.setLayout(self.file_layout)
+
+        # Actions group - kept visually separate (its own framed group) from the
+        # input fields above. Preview/Create Mesh/Start Simulation/Terminate (plus,
+        # in setupEM's subclass, View Results/Model Fit) all share this grid so
+        # their right edges line up at exactly the same two-thirds/one-third split
+        # - a QGridLayout keeps columns aligned across rows; independent
+        # QHBoxLayouts can't guarantee that once some rows have two widgets (e.g.
+        # Start Simulation/Terminate) and others have one (Preview/Create Mesh).
+        self.actions_group = QGroupBox("Actions")
+        self.actions_layout = QVBoxLayout()
+        self.buttons_grid = QGridLayout()
+        self.buttons_grid.setColumnStretch(0, 2)  # primary column: two thirds
+        self.buttons_grid.setColumnStretch(1, 1)  # secondary column: one third
+
         self.preview_model_btn = QPushButton("⚙️ Preview model geometry in gmsh")
         self.preview_model_btn.clicked.connect(self.preview_model)
-        self.file_layout.addWidget(self.preview_model_btn)
+        self.buttons_grid.addWidget(self.preview_model_btn, 0, 0)
 
         self.create_model_btn = QPushButton("⚙️ Create mesh and simulation settings file")
         self.create_model_btn.clicked.connect(self.create_mesh)
-        self.file_layout.addWidget(self.create_model_btn)
+        self.buttons_grid.addWidget(self.create_model_btn, 1, 0)
 
-        self.run_layout = QHBoxLayout()
         self.create_run_btn = QPushButton("▶️ Start Simulation")
         self.create_run_btn.clicked.connect(self.run_model)
-        self.run_layout.addWidget(self.create_run_btn)
+        self.buttons_grid.addWidget(self.create_run_btn, 2, 0)
         self.kill_btn = QPushButton("🛑 Terminate ")
         self.kill_btn.clicked.connect(self.terminate_run)
-        self.run_layout.addWidget(self.kill_btn)
-        self.run_layout.setStretch(0, 5)  # index 0 = Run
-        self.run_layout.setStretch(1, 1)  # index 1 = Terminate
+        self.buttons_grid.addWidget(self.kill_btn, 2, 1)
 
-        self.file_layout.addLayout(self.run_layout)
+        self.actions_layout.addLayout(self.buttons_grid)
 
-        self.file_group.setLayout(self.file_layout)
-
-        # Log group
-
-        self.log_group = QGroupBox("Log file")
-        self.log_layout = QVBoxLayout()
-        self.log_group.setLayout(self.log_layout)
+        # Log file: kept inside the Actions frame (not its own group box) since it is
+        # the direct output of the actions above (Preview/Create Mesh/Start Simulation),
+        # not an independent input section.
+        self.actions_layout.addSpacing(10)
+        self.actions_layout.addWidget(QLabel("Log file:"))
         self.log_area = QPlainTextEdit()
         self.log_area.setReadOnly(True)
         log_font = QFont()
@@ -1256,11 +1263,13 @@ class CreateModelTabBase(QWidget):
         log_font.setFixedPitch(True)
         log_font.setPointSize(9)
         self.log_area.setFont(log_font)
-        self.log_layout.addWidget(self.log_area)
+        self.actions_layout.addWidget(self.log_area)
+
+        self.actions_group.setLayout(self.actions_layout)
 
         self.main_layout.addWidget(self.file_group)
         self.main_layout.addSpacing(20)
-        self.main_layout.addWidget(self.log_group)
+        self.main_layout.addWidget(self.actions_group)
         self.setLayout(self.main_layout)
 
         # --- QProcess setup ---

{setupem-0.4.0 → setupem-0.5.0/src/setupEM.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: setupEM
-Version: 0.4.0
+Version: 0.5.0
 Summary: Python tool for configuration of gds2palace workflow with GUI.
 Author-email: Volker Muehlhaus <volker@muehlhaus.com>
 License-Expression: GPL-3.0-or-later
@@ -21,10 +21,14 @@ Dynamic: license-file
 
 ## What's New
 
-- A built-in **Result Viewer** (Create Model tab > View Results...) to browse and plot Touchstone S-parameter results without leaving setupEM
-- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
+setupEM includes two built-in tools for working with simulation results directly, without external scripts:
+
+- **Result Viewer** (Create Model tab > View Results...) plots Touchstone S-parameter results - dB/phase, Smith chart, zoomed Smith chart - without leaving setupEM. See section "[Result Viewer](#result-viewer)".
+- **Model Fit** (Create Model tab > Model Fit...) launches [snp2le](https://github.com/iic-jku/snp2le), an external open-source tool that extracts a lumped-element SPICE/Spectre netlist from S-parameter results - offering to install it via pip if it isn't already present. See section "[Model Fit](#model-fit)".
+
 - A graphical **Stackup XML Editor** (Tools > Edit Stackup XML...), including Variables, Reference-relative positioning, Derived Layers, and Thermal Tables
 - Input Files tab can now **override stackup Variables** (e.g. `total_thickness`) directly, without hand-editing the XML or the generated model
+- **setupThermal**, a companion app for Elmer thermal simulation, alongside setupEM
 
 See [CHANGES.md](https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/CHANGES.md) for details.
 
@@ -195,6 +199,8 @@ The buttons are used top down: You can first preview the resulting model geometr
 
 To start simulation, use the "Run palace" button. If you are on Linux, this will start Palace using script "run_sim". If you are on Windows, this will start the Linux Subsystem for Windows (WSL) and open a command prompt in the simulation directory.
 
+If the output directory already has results from a previous run, you are asked whether to delete or keep them before starting - default is to delete, so a rerun with different settings doesn't leave stale results mixed in with the new ones.
+
 <img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/createmodel3.png" alt="create" width="700">
 
 To start simulation on Linux, it is required that you have configured a script "run_sim" as described in the [gds2palace documentation](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/blob/main/doc/gds2palace_workflow_userguide.pdf). You can find a template [here](https://github.com/VolkerMuehlhaus/gds2palace_ihp_sg13g2/tree/main/scripts) in the gds2palace repository.
@@ -230,6 +236,24 @@ The matplotlib toolbar above the plot (pan/zoom/save as PNG) works as usual. A f
 
 Result Viewer can also be run standalone, without the full setupEM GUI, either directly (`python result_viewer.py [target_dir]`) or via the `resultViewer` console script installed with the package.
 
+## Model Fit
+
+Click **Model Fit...** on the Create Model tab (next to **View Results...**) to extract a lumped-element netlist from the current run's S-parameter result, using [snp2le](https://github.com/iic-jku/snp2le) - an external, open-source tool, not part of setupEM.
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/resultviewer_button.png" alt="model fit button" width="700">
+
+If snp2le isn't installed, setupEM offers to install it for you via pip:
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/modelfit1.png" alt="snp2le not installed" width="350">
+
+Choosing **Install** runs `pip install snp2le` in the Log panel and, once it succeeds, continues automatically - no need to click Model Fit a second time. Choosing **Cancel** logs the manual install command and the project link instead.
+
+Once snp2le is available, Model Fit locates the raw (not `_dc`, not `_deembedded`) Touchstone result file for the current run and launches the snp2le GUI in that file's directory. snp2le's GUI has no command-line option to preload a file, so the exact path is printed to the Log panel - load it via snp2le's own file picker:
+
+<img src="https://raw.githubusercontent.com/VolkerMuehlhaus/setupEM/main/doc/png/modelfit2.png" alt="snp2le starting" width="700">
+
+If no raw result file exists yet (no simulation has been run), Model Fit shows a warning instead of starting snp2le - run a simulation first.
+
 ## Code 
 Behind the scenes, the setupEM user interface created Python model code for gds2palace, and you can check the resulting code on the "Code" tab.