allensdk.brain_observatory.dff module¶
-
allensdk.brain_observatory.dff.compute_dff(traces, save_plot_dir=None, mode_kernelsize=5400, mean_kernelsize=3000)[source]¶ Compute dF/F of a set of traces using a low-pass windowed-mode operator. The operation is basically:
T_mm = windowed_mean(windowed_mode(T))
T_dff = (T - T_mm) / T_mm
Parameters: - traces: np.ndarray
2D array of traces to be analyzed
Returns: - np.ndarray with the same shape as the input array.
-
allensdk.brain_observatory.dff.movingaverage(x, kernelsize, y)[source]¶ Compute the windowed average of an array.
Parameters: - x: np.ndarray
Array to be analyzed
- kernelsize: int
Size of the moving window
- y: np.ndarray
Output array to store the results
-
allensdk.brain_observatory.dff.movingmode_fast(x, kernelsize, y)[source]¶ Compute the windowed mode of an array. A running mode is initialized with a histogram of values over the initial kernelsize/2 values. The mode is then updated as the kernel moves by adding and subtracting values from the histogram.
Parameters: - x: np.ndarray
Array to be analyzed
- kernelsize: int
Size of the moving window
- y: np.ndarray
Output array to store the results