Spatial_Markov gives its own set of 'k' classes even though explicitly mentioned as paramter value about giddy HOT 4 CLOSED

Hi @romass12! Thank you for reporting this.

Could you please provide the output (error message) in a screenshot and possibly how your data (rpci) look like so that we can help you debug?

from giddy.

Hi @weikang9009
Yes, the rpci shape is (23,203) with 4 classes.
The distribution of 3 classes where the event occurs and is divided into classes are,
1 -> 1300
2 -> 587
3 -> 379
The rest is '0' class, where the event did not occur.

array([[1, 1, 0, ..., 0, 0, 0], [0, 0, 2, ..., 1, 2, 0], [2, 0, 0, ..., 2, 1, 1], ..., [0, 0, 0, ..., 0, 0, 0], [0, 0, 0, ..., 0, 0, 0], [0, 0, 0, ..., 0, 0, 0]])

Spatial_markov is quantizing the input even though I have already binned the data

The output says:
/usr/local/lib/python3.7/dist-packages/mapclassify/classifiers.py:239: UserWarning: Warning: Not enough unique values in array to form k classes "Warning: Not enough unique values in array to form k classes", UserWarning /usr/local/lib/python3.7/dist-packages/mapclassify/classifiers.py:241: UserWarning: Warning: setting k to 2 Warn("Warning: setting k to %d" % k_q, UserWarning)

from giddy.

Hi @romass12, thank you for the description of your data.

Since your data has been discretized/classified, you should set the parameter discrete=True when initializing the Spatial_Markov class. The spatial lag will be calculated as the most common category among neighboring observations. Please refer to the Spatial_Markov API documentation and pay special attention to the parameter discrete and look at the example (4) where the discrete case is handled.

from giddy.

Thank you so much @weikang9009

from giddy.