I/O operations

Functions used to read data from various sources and store them in a numpy array.

Read files

Read image, video and numerical data as a numpy array:

• bigfish.stack.read_image()

• bigfish.stack.read_dv()

• bigfish.stack.read_array()

Read a zipped archive of files as a dictionary-like object:

• bigfish.stack.read_uncompressed()

• bigfish.stack.read_cell_extracted()

Read CSV file:

• bigfish.stack.read_array_from_csv()

• bigfish.stack.read_dataframe_from_csv()

bigfish.stack.read_image(path, sanity_check=False)

Read an image with png, jpg, jpeg, tif or tiff extension.

Parameters:

pathstr

Path of the image to read.

sanity_checkbool

Check if the array returned fits with bigfish pipeline.

Returns:

imagendarray, np.uint or np.int

Image read.

bigfish.stack.read_dv(path, sanity_check=False)

Read a video file with dv extension.

Parameters:

pathstr

Path of the file to read.

sanity_checkbool

Check if the array returned fits with bigfish pipeline.

Returns:

videondarray

Video read.

bigfish.stack.read_array(path)

Read a numpy array with npy extension.

Parameters:

pathstr

Path of the array to read.

Returns:

arrayndarray

Array read.

bigfish.stack.read_uncompressed(path, verbose=False)

Read a NpzFile object with npz extension.

Parameters:

pathstr

Path of the file to read.

verbosebool

Return names of the different objects.

Returns:

dataNpzFile object

NpzFile read.

bigfish.stack.read_cell_extracted(path, verbose=False)

Read a NpzFile object with npz extension, previously written with bigfish.stack.save_cell_extracted().

Parameters:

pathstr

Path of the file to read.

verbosebool

Return names of the different objects.

Returns:

cell_resultsDict

Dictionary including information about the cell (image, masks, coordinates arrays). Minimal information are:

• cell_id: Unique id of the cell.

• bbox: bounding box coordinates with the order (min_y, min_x, max_y, max_x).

• cell_coord: boundary coordinates of the cell.

• cell_mask: mask of the cell.

bigfish.stack.read_array_from_csv(path, dtype=None, delimiter=';', encoding='utf-8', skiprows=0)

Read a numpy array saved in a csv file.

Parameters:

pathstr

Path of the csv file to read.

dtypetype, optional

Expected dtype to cast the array.

delimiterstr, default=”;”

Delimiter used to separate columns.

encodingstr, default=”utf-8”

Encoding to use.

skiprowsint, default=0

Skip the first skiprows lines of the file. Useful to skip the first rows of a csv with header.

Returns:

arrayndarray

Array read.

bigfish.stack.read_dataframe_from_csv(path, delimiter=';', encoding='utf-8')

Read a numpy array or a pandas object saved in a csv file.

Parameters:

pathstr

Path of the csv file to read.

delimiterstr

Delimiter used to separate columns.

encodingstr

Encoding to use.

Returns:

dfpd.DataFrame

Pandas object read.

---

Write files

Save numpy array:

• bigfish.stack.save_image()

• bigfish.stack.save_array()

Save cell-level results in a zipped archive of files:

• bigfish.stack.save_cell_extracted()

Save tabular data in a CSV file:

• bigfish.stack.save_data_to_csv()

bigfish.stack.save_image(image, path, extension='tif')

Save an image.

The input image should have between 2 and 5 dimensions, with boolean, (unsigned) integer, or float.

The dimensions should be in the following order: (round, channel, z, y, x).

Parameters:

imagenp.ndarray

Image to save.

pathstr

Path of the saved image.

extensionstr

Default extension to save the image (among png, jpg, jpeg, tif or tiff).

Notes

• If the image has more than 2 dimensions, tif and tiff extensions are required (png extension does not handle 3-d images other than (M, N, 3) or (M, N, 4) shapes).

• A 2-d boolean image can be saved in png, jpg or jpeg (cast in np.uint8).

• A multidimensional boolean image should be saved with bigfish.stack.save_array() or as a boolean images with tif/ tiff extension.

bigfish.stack.save_array(array, path)

Save an array in a npy extension file.

The input array should have between 2 and 5 dimensions, with boolean, (unsigned) integer, or float.

Parameters:

arraynp.ndarray

Array to save.

pathstr

Path of the saved array.

bigfish.stack.save_cell_extracted(cell_results, path)

Save cell-level results from bigfish.stack.extract_cell() in a NpzFile object with npz extension.

Parameters:

cell_resultsDict

Dictionary including information about the cell (image, masks, coordinates arrays). Minimal information are:

• cell_id: Unique id of the cell.

• bbox: bounding box coordinates with the order (min_y, min_x, max_y, max_x).

• cell_coord: boundary coordinates of the cell.

• cell_mask: mask of the cell.

pathstr

Path of the saved array.

bigfish.stack.save_data_to_csv(data, path, delimiter=';')

Save a numpy array or a pandas object into a csv file.

The input should be a pandas object (Series or DataFrame) or a numpy array with 2 dimensions and (unsigned) integer or float.

Parameters:

datanp.ndarray, pd.Series or pd.DataFrame

Data to save.

pathstr

Path of the saved csv file.

delimiterstr

Delimiter used to separate columns.