Scripts to get data from the OCP server. Feel free to submit bug reports or feature requests here.
These scripts are written to work with 2.7 and (ideally, but untestedly) Python 3. An example (how to download the CELL, Kasthuri (July 30 2015) data) is included inside /examples.
Use pip to install the latest stable ocpy package from the Python Package Index:
pip install ocpy
You may need to satisfy requirements as listed in the requirements file: You can use pip for this as well, and run pip install -r requirements from inside the package directory, or you can manually install each of the libraries listed in the requirements file.
You can use the function ocpy.access.get_data() to retrieve data from the OCP servers.
| Argument | Required | Description |
|---|---|---|
token |
Yes | The project token to download. |
x_start |
Yes | The low bound in dimension X |
x_stop |
Yes | The high bound in dimension X |
y_start |
Yes | The low bound in dimension Y |
y_stop |
Yes | The high bound in dimension Y |
z_start |
Yes | The low bound in dimension Z |
z_stop |
Yes | The high bound in dimension Z |
fmt |
No (hdf5) |
The format in which to download code. (Currently only hdf5 is legal.) |
resolution |
Yes | The resolution level at which to download data |
server |
No (http://openconnecto.me) |
The server at which to request data |
location |
No (./) |
The location on-disk (locally) where you'd like to save the data. Two subdirectories will be created: /hdf5 and /tiff. |
See the guide for more examples and use cases.
import ocpy.access
ocpy.access.get_data(
token = "kasthuri11",
x_start = 5000, x_stop = 5500,
y_start = 5000, y_stop = 5500,
z_start = 1, z_stop = 3,
location = "data"
)
The above script downloads 500x500x2 voxels of data from the kasthuri11 dataset and saves them on your hard-drive inside a subdirectory called data.
Note: There is a known bug in the scipy/PIL TIFF converter that prevents accurate hdf5-tiff conversion of non-uint8 data.
This package was generated using this tutorial.