torch_ecg.utils#

This module contains a collection of utility functions and classes that are used throughout the package.

Neural network auxiliary functions and classes #

`extend_predictions`(preds, classes, ...)	Extend the prediction arrays to prediction arrays in larger range of classes
`compute_output_shape`(layer_type, input_shape)	Compute the output shape of a (transpose) convolution/maxpool/avgpool layer.
`compute_conv_output_shape`(input_shape[, ...])	Compute the output shape of a convolution layer.
`compute_deconv_output_shape`(input_shape[, ...])	Compute the output shape of a transpose convolution layer
`compute_maxpool_output_shape`(input_shape[, ...])	Compute the output shape of a maxpool layer.
`compute_avgpool_output_shape`(input_shape[, ...])	Compute the output shape of a avgpool layer.
`compute_sequential_output_shape`(model[, ...])	Compute the output shape of a sequential model.
`compute_module_size`(module[, requires_grad, ...])	compute the size (number of parameters) of a `Module`.
`default_collate_fn`(batch)	Default collate functions for model training.
`compute_receptive_field`([kernel_sizes, ...])	Compute the receptive field of several types of `Module`.
`adjust_cnn_filter_lengths`(config, fs[, ...])	Adjust the filter lengths in the config for convolutional neural networks, according to the new sampling frequency.
`SizeMixin`()	Mixin class for size related methods
`CkptMixin`()	Mixin class for loading from checkpoint class methods

`smooth`(x[, window_len, window, mode, keep_dtype])	Smooth the 1d data using a window with requested size.
`resample_irregular_timeseries`(sig[, ...])	Resample the 2d irregular timeseries sig into a 1d or 2d regular time series with frequency output_fs, elements of sig are in the form `[time, value]`, where the unit of time is ms.
`detect_peaks`(x[, mph, mpd, threshold, ...])	Detect peaks in data based on their amplitude and other features.
`remove_spikes_naive`(sig[, threshold, inplace])	Remove signal spikes using a naive method.
`butter_bandpass_filter`(data, lowcut, ...[, ...])	Butterworth bandpass filtering the signals.
`get_ampl`(sig, fs[, fmt, window, critical_points])	Get amplitude of a signal (near critical points if given).
`normalize`(sig, method[, mean, std, sig_fmt, ...])	Normalize a signal.
`normalize_t`(sig[, method, mean, std, ...])	Perform z-score normalization on `sig`, to make it has fixed mean and standard deviation, or perform min-max normalization on `sig`, or normalize `sig` using `mean` and `std` via \((sig - mean) / std\).
`resample_t`(sig[, fs, dst_fs, siglen, inplace])	Resample signal tensors to a new sampling frequency or a new signal length.

`get_mask`(shape, critical_points, left_bias, ...)	Get the mask around the given critical points.
`class_weight_to_sample_weight`(y[, class_weight])	Transform class weight to sample weight.
`ensure_lead_fmt`(values[, n_leads, fmt])	Ensure the multi-lead (ECG) signal to be of specified format.
`ensure_siglen`(values, siglen[, fmt, tolerance])	Ensure the (ECG) signal to be of specified length.
`masks_to_waveforms`(masks, class_map, fs[, ...])	Convert masks into lists of `ECGWaveForm` for each lead.
`mask_to_intervals`(mask[, vals, right_inclusive])	Convert a mask into a list of intervals, or a dict of lists of intervals.
`uniform`(low, high, num)	Generate a list of numbers uniformly distributed.
`stratified_train_test_split`(df, stratified_cols)	Perform stratified train-test split on the dataframe.
`one_hot_encode`(cls_array[, num_classes, dtype])	Convert a categorical array to a one-hot array.
`generate_weight_mask`(target_mask, fg_weight, ...)	Generate weight mask for a binary target mask, accounting the foreground weight and boundary weight.

`overlaps`(interval, another)	Find the overlap between two intervals.
`validate_interval`(interval[, join_book_endeds])	Check whether interval is an Interval or a GeneralizedInterval.
`in_interval`(val, interval[, left_closed, ...])	Check whether val is inside interval or not.
`in_generalized_interval`(val, ...[, ...])	Check whether val is inside generalized_interval or not.
`intervals_union`(interval_list[, ...])	Find the union of intervals.
`generalized_intervals_union`(interval_list[, ...])	Calculate the union of a list (or tuple) of `GeneralizedInterval`.
`intervals_intersection`(interval_list[, ...])	Calculate the intersection of all intervals in interval_list.
`generalized_intervals_intersection`(...[, ...])	calculate the intersection of intervals.
`generalized_interval_complement`(...)	Calculate the complement of an interval in another interval.
`get_optimal_covering`(total_interval, ...[, ...])	Compute an optimal covering of to_cover by intervals.
`interval_len`(interval)	Compute the length of an interval.
`generalized_interval_len`(generalized_interval)	Compute the length of an interval.
`find_extrema`(signal[, mode])	Locate local extrema points in a 1D signal.
`is_intersect`(interval, another_interval)	Determines if two (generalized) intervals intersect or not.
`max_disjoint_covering`(intervals[, ...])	Find the largest (the largest interval length) covering of a sequence of intervals.

`top_n_accuracy`(labels, outputs[, n])	Compute top n accuracy.
`confusion_matrix`(labels, outputs[, num_classes])	Compute a binary confusion matrix
`ovr_confusion_matrix`(labels, outputs[, ...])	Compute binary one-vs-rest confusion matrices.
`metrics_from_confusion_matrix`(labels, outputs)	Compute macro metrics, and metrics for each class.
`compute_wave_delineation_metrics`(...[, ...])
`QRS_score`(rpeaks_truths, rpeaks_preds, fs[, thr])	QRS accuracy score, proposed in CPSC2019.

`add_docstring`(doc[, mode])	Decorator to add docstring to a function or a class.
`remove_parameters_returns_from_docstring`(doc)	Remove parameters and/or returns from docstring, which is of the format of numpydoc.
`default_class_repr`(c[, align, depth])	Default class representation.
`ReprMixin`()	Mixin class for enhanced `__repr__()` and `__str__()` methods.
`CitationMixin`()	Mixin class for getting citations from DOIs.
`get_kwargs`(func_or_cls[, kwonly])	Get the kwargs of a function or class.
`get_required_args`(func_or_cls)	Get the required positional arguments of a function or class.
`add_kwargs`(func, **kwargs)	Add keyword arguments to a function.

Get the list of records in a recursive manner.

`dict_to_str`(d[, current_depth, indent_spaces])	Convert a (possibly) nested dict into a str of json-like formatted form.
`str2bool`(v)	Converts a "boolean" value possibly in the format of `str` to `bool`.
`nildent`(text)	Kill all leading white spaces in each line of text, while keeping all lines (including empty)
`get_date_str`([fmt])	Get the current time in the `str` format.

ecg_plot(ecg, sample_rate, columns, ...[, ...])

Function to plot raw ECG signal.

`init_logger`([log_dir, log_file, log_name, ...])	Initialize a logger.
`list_sum`(lst)	Sum a sequence of lists.
`dicts_equal`(d1, d2[, allow_array_diff_types])	Determine if two dicts are equal.
`MovingAverage`([data])	Class for computing moving average.
`Timer`([name, verbose])	Context manager to time the execution of a block of code.
`timeout`(duration)	A context manager that raises a `TimeoutError` after a specified time (in seconds).
`np_topk`(arr, k[, dim, largest, sorted])	Find the k largest elements of an array along a specified axis.
`select_k`(arr, k[, dim, largest, sorted])	Select elements from an array along a specified axis of specific rankings.