You can download a file from OSF to your Colab notebook using !wget, which will store it as a local file. Depending on what is in the file it can then be loaded using the usual methods. Here is an example for creating a numpy.npz file with multiple named matrices:
#@title Data retrieval
#@markdown This cell downloads the example dataset that we will use in this tutorial.
import io
import requests
r = requests.get('https://osf.io/sy5xt/download')
if r.status_code != 200:
print('Failed to download data')
else:
spike_times = np.load(io.BytesIO(r.content), allow_pickle=True)['spike_times']
And this allows treating the downloaded data set as a file, although it is less secure:
#np.savez('W1D2_data.npz', judgments=judgments, opticflow=opticflow, vestibular=vestibular)
!wget https://osf.io/c5xyf/download -O W1D2_data.npz
filez = np.load(file='W1D2_data.npz', allow_pickle=True)
judgments = filez['judgments']
opticflow = filez['opticflow']
vestibular = filez['vestibular']
Important is to get the URL correct: https://osf.io/c5xyf/download which can be found by clicking on the file in OSF after upload, then right-click on 'Download' in the upper right and copying the link / URL / address.
