René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"control.timeresp.TimeResponseData.to_pandas() failing","articleBody":"Hi, today I was using some step responses and noticed that the `.to_pandas()` is not actually working.\r\n\r\nI managed to workaround it by creating my own function to translate the response into a dataframe.\r\n\r\nExample of code not working:\r\n\r\n```Python\r\nimport control as ct\r\nimport numpy as np\r\n\r\nmodel = ct.rss(states=['x0', 'x1'], outputs=['y0', 'y1'], inputs=['u0', 'u1'], name='My Model')\r\n\r\nT = np.linspace(0, 10, 100, endpoint=False)\r\nX0 = np.zeros(model.nstates)\r\n\r\nres = ct.step_response(model, T=T, X0=X0, input=0)\r\n\r\ndf = res.to_pandas()\r\n```\r\n\r\nError:\r\n\r\n```python\r\n---------------------------------------------------------------------------\r\nValueError                                Traceback (most recent call last)\r\nCell In[140], line 6\r\n      3 T = np.linspace(0, 10, 100, endpoint=False)\r\n      4 X0 = np.zeros(model.nstates)\r\n----\u003e 6 res = ct.step_response(model, T=T, X0=X0, input=0).to_pandas()\r\n\r\nFile ~.env/lib/python3.10/site-packages/control/timeresp.py:723, in TimeResponseData.to_pandas(self)\r\n    719 if self.nstates \u003e 0:\r\n    720     data.update(\r\n    721         {name: self.x[i] for i, name in enumerate(self.state_labels)})\r\n--\u003e 723 return pandas.DataFrame(data)\r\n\r\nFile ~.env/lib/python3.10/site-packages/pandas/core/frame.py:778, in DataFrame.__init__(self, data, index, columns, dtype, copy)\r\n    772     mgr = self._init_mgr(\r\n    773         data, axes={\"index\": index, \"columns\": columns}, dtype=dtype, copy=copy\r\n    774     )\r\n    776 elif isinstance(data, dict):\r\n    777     # GH#38939 de facto copy defaults to False only in non-dict cases\r\n--\u003e 778     mgr = dict_to_mgr(data, index, columns, dtype=dtype, copy=copy, typ=manager)\r\n    779 elif isinstance(data, ma.MaskedArray):\r\n    780     from numpy.ma import mrecords\r\n\r\nFile ~.env/lib/python3.10/site-packages/pandas/core/internals/construction.py:503, in dict_to_mgr(data, index, columns, dtype, typ, copy)\r\n    499     else:\r\n    500         # dtype check to exclude e.g. range objects, scalars\r\n    501         arrays = [x.copy() if hasattr(x, \"dtype\") else x for x in arrays]\r\n--\u003e 503 return arrays_to_mgr(arrays, columns, index, dtype=dtype, typ=typ, consolidate=copy)\r\n\r\nFile ~.env/lib/python3.10/site-packages/pandas/core/internals/construction.py:114, in arrays_to_mgr(arrays, columns, index, dtype, verify_integrity, typ, consolidate)\r\n    111 if verify_integrity:\r\n    112     # figure out the index, if necessary\r\n    113     if index is None:\r\n--\u003e 114         index = _extract_index(arrays)\r\n    115     else:\r\n    116         index = ensure_index(index)\r\n\r\nFile ~.env/lib/python3.10/site-packages/pandas/core/internals/construction.py:664, in _extract_index(data)\r\n    662         raw_lengths.append(len(val))\r\n    663     elif isinstance(val, np.ndarray) and val.ndim \u003e 1:\r\n--\u003e 664         raise ValueError(\"Per-column arrays must each be 1-dimensional\")\r\n    666 if not indexes and not raw_lengths:\r\n    667     raise ValueError(\"If using all scalar values, you must pass an index\")\r\n\r\nValueError: Per-column arrays must each be 1-dimensional\r\n```\r\n\r\n---\r\n\r\nThe code I'm using to workaround it is the following:\r\n\r\n```Python\r\n\r\nimport matplotlib.pyplot as plt\r\nimport control as ct\r\nimport numpy as np\r\n\r\ndef step_response_to_pandas(step_response):\r\n    return pd.DataFrame(\r\n        {'trace_label': np.array([[label] * (len(res.time)) for label in res.trace_labels]).ravel()} |\r\n        {'time': res.time.repeat(len(res.trace_labels))} |\r\n        {label: res.inputs[i].ravel() for i,label in enumerate(res.input_labels)} |\r\n        {label: res.outputs[i].ravel() for i,label in enumerate(res.output_labels)} |\r\n        {label: res.states[i].ravel() for i,label in enumerate(res.state_labels)}\r\n    )\r\n\r\ndef plot_step_response_dataframe(df):\r\n    grouped = df.groupby(level='trace_label')\r\n    row_size = 1\r\n\r\n    for trace_label, group in grouped:\r\n        fig, axes = plt.subplots(len(group.columns), 1, figsize=(6.4, len(group.columns) * row_size), sharex=True)\r\n        fig.suptitle(f'Trace: {trace_label}', fontsize=16)\r\n        \r\n        if len(group.columns) == 1:\r\n            axes = [axes]\r\n        \r\n        for ax, (signal_name, signal_data) in zip(axes, group.items()):\r\n            ax.plot(group.index.get_level_values('time'), signal_data, label=signal_name)\r\n            ax.grid(True)\r\n            ax.set_ylabel(signal_name)\r\n        \r\n        axes[-1].set_xlabel('Time')\r\n        \r\n        plt.tight_layout()\r\n        plt.show()\r\n\r\n\r\nmodel = ct.rss(states=['x0', 'x1'], outputs=['y0', 'y1'], inputs=['u0', 'u1'], name='My Model')\r\n\r\nT = np.linspace(0, 10, 100, endpoint=False)\r\nX0 = np.zeros(model.nstates)\r\n\r\nres = ct.step_response(model, T=T, X0=X0)\r\n\r\ndf = step_response_to_pandas(res)\r\ndf = df.set_index(['trace_label', 'time'])\r\n\r\nplot_step_response_dataframe(df)\r\n\r\ndisplay(df)\r\n\r\n```\r\n\r\nExample of output:\r\n\r\n![image](https://github.com/user-attachments/assets/caf0cbf0-29af-4478-adc1-14169a36013c)\r\n![image](https://github.com/user-attachments/assets/5028e6e8-4a29-46cb-aeef-a4233cfea04c)\r\n![image](https://github.com/user-attachments/assets/29ea4ddf-325c-45d8-9470-1f993ec20283)\r\n\r\n\r\n---\r\n\r\nThanks!\r\n","author":{"url":"https://github.com/joaoantoniocardoso","@type":"Person","name":"joaoantoniocardoso"},"datePublished":"2024-12-30T19:41:30.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/1087/python-control/issues/1087"}