Open qianmeil opened 10 months ago
Hello,
It looks like you have a memory error in the computation of the spatial weights matrix (https://stackoverflow.com/questions/8378797/stdbad-alloc-am-i-using-too-much-memory). The Plugin is tested for some thousand points only... and 130k double-precision point coordinates may be too much for the plugin.
For such datasets, I suggest moving to Python directly.
Daniele
Hello,
Thank you for your quick reply! I have no idea how to run such a plugin on python directly,because there are many scripts and the plugin is related to QGIS,and I can not find out any useful information. I wonder if you can tell me detailed procedures or if you know any similar examples about running QGIS plugins on python directly.
Looking forward to your reply!
I suggest using directly PySAL (on which the Plugin is based). A relevant reference to start can be: https://geographicdata.science/book/notebooks/07_local_autocorrelation.html
Additionally, you may try other free software like GeoDa for computing LISA on your dataset. Have a look!
Best,
Daniele
When I run the plugin using a few point for test,it works,but when I run this plugin with around 130000 latitude and longitude points,it report an error: Traceback (most recent call last): File "C:\Users/10784/AppData/Roaming/QGIS/QGIS3\profiles\default/python/plugins\HotspotAnalysis\hotspot_analysis.py", line 610, in run w = DistanceBand(t, threshold1, p=2, binary=False) File "C:\Users\10784\AppData\Roaming\Python\Python39\site-packages\pysal\lib\weights\distance.py", line 781, in init self._band() File "C:\Users\10784\AppData\Roaming\Python\Python39\site-packages\pysal\lib\weights\distance.py", line 848, in _band self.dmat = self.kdtree.sparse_distance_matrix( File "D:\PROGRA~1\apps\Python39\lib\site-packages\scipy\spatial_kdtree.py", line 865, in sparse_distance_matrix return super().sparse_distance_matrix( File "_ckdtree.pyx", line 1528, in scipy.spatial._ckdtree.cKDTree.sparse_distance_matrix MemoryError: std::bad_alloc Windows 11 QGIS:3.34.2 python:3.9.5 Looking forward to your reply!