graph LR
mne_io["mne.io"]
mne_io_base_BaseRaw["mne.io.base.BaseRaw"]
mne__fiff_meas_info_Info["mne._fiff.meas_info.Info"]
mne_annotations_Annotations["mne.annotations.Annotations"]
mne_epochs_Epochs["mne.epochs.Epochs"]
mne_evoked_Evoked["mne.evoked.Evoked"]
mne_cov_Covariance["mne.cov.Covariance"]
mne_datasets["mne.datasets"]
mne_datasets -- "provides data for" --> mne_io
mne_io -- "produces" --> mne_io_base_BaseRaw
mne_io_base_BaseRaw -- "contains" --> mne__fiff_meas_info_Info
mne_io_base_BaseRaw -- "contains" --> mne_annotations_Annotations
mne_epochs_Epochs -- "consumes" --> mne_io_base_BaseRaw
mne_epochs_Epochs -- "consumes" --> mne_annotations_Annotations
mne_epochs_Epochs -- "contains" --> mne__fiff_meas_info_Info
mne_evoked_Evoked -- "consumes" --> mne_epochs_Epochs
mne_evoked_Evoked -- "contains" --> mne__fiff_meas_info_Info
mne_cov_Covariance -- "consumes" --> mne_io_base_BaseRaw
mne_cov_Covariance -- "consumes" --> mne_epochs_Epochs
mne_cov_Covariance -- "contains" --> mne__fiff_meas_info_Info
The Data Management & Core Models component in MNE-Python is fundamental to the library's operation, serving as the backbone for all neuroimaging data handling. It adheres to the "Data-Centric Architecture" pattern, where core data objects are central, and other components interact with these objects. The "Modular Design" pattern is also evident, as different data formats are handled by specific sub-modules within mne.io, and each core data structure is encapsulated in its own module.
This component is responsible for reading and writing various neuroimaging data formats (e.g., FIF, EDF, BrainVision, KIT). It acts as a facade for different raw data readers, providing a unified interface for loading diverse datasets.
Related Classes/Methods:
mne.io(1:1)mne.io.fiff.raw(1:1)mne.io.edf.edf(1:1)
This is the foundational abstract class for representing continuous raw neurophysiological data. It provides common methods and attributes for all raw data types, ensuring consistency across different acquisition systems.
Related Classes/Methods:
This class stores comprehensive metadata about a neuroimaging recording, including channel names, types, sampling frequency, sensor locations, and acquisition parameters. It's crucial for correctly interpreting and processing the data.
Related Classes/Methods:
This component manages annotations and events within the data, such as bad segments, experimental events, or physiological markers. These annotations are critical for data cleaning, epoching, and event-related analysis.
Related Classes/Methods:
This class represents data segmented into trials or epochs, typically extracted around specific events. It's a key data structure for event-related potential (ERP) and event-related field (ERF) analysis.
Related Classes/Methods:
This class represents averaged data across multiple epochs, typically used to visualize and analyze event-related potentials/fields. Averaging improves the signal-to-noise ratio.
Related Classes/Methods:
This component handles the noise covariance matrix, which is essential for source localization and inverse modeling. It characterizes the noise present in the data.
Related Classes/Methods:
This component provides access to various publicly available neuroimaging datasets. These datasets are invaluable for examples, tutorials, testing, and demonstrating MNE-Python's capabilities.
Related Classes/Methods:
mne.datasets(1:1)