01_functional_properties_with_python_measurements_with_stability.pkl

The .pkl file integrates all the processed data from the .nwb files into a single DataFrame for easier access, without the need to load large files every time, where each row represents one single neuron's data.
Data structure (for example, read the file using pandas.read_pickle() as 'df' in Python):
Each row represents one single neuron's data, and columns represent specific features. For example:
df['spike_times'] - This is a vector of time points where spike occurrence is relative to the start of recording.
df['unit_name'] - Unit ID from SpikeInterface.
df['quality'] - Based on df['l_ratio'], df['isi_violations_ratio'], and df['amplitude_median'], this filters well-isolated single units as "good" and others as "noise."
df['buzaki_py_cell_type'] - This classifies the units into "pyramidal" or "interneurons" based on df['peak_to_valley'], df['tau_rise_python'], and df['mean_firing_rate'].