About the input of ccg calculation

[PathwayinGithub]

Hi！
I read your neuron paper and have an issue about the input of jittered ccg calculation in ccg.py .
I don‘t understand the meaning of the four dimensions of the input variable（spikes: neuronoritrial*time）and can't find the explanation.
We used neuropixels3B1 and spikeglx software to record the neurons. And I have some data now：a list of spike times（spike_times.npy)，a list of the cluster id of each spike time(spike_cluster.npy)…… How can I transform these data into the structure your code used？
Thanks！