QGIS-Plugin-Produce-Training-Samples-For-Deep-Learning
QGIS-Plugin-Produce-Training-Samples-For-Deep-Learning copied to clipboard
PratyushTripathy
About dependencies
How do I install dependencies manually? since it is showing the SSL error despite having openpyxl installed. For installation I had to make a few changes to the original code to make it work on my system. Unfortunately, none of the solutions on github, stackoverflow worked... this is the error WARNING: pip is configured with locations that require TLS/SSL, however the ssl module in Python is not available. WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLError("Can't connect to HTTPS URL because the SSL module is not available.")': /simple/pyrsgis/
Hi @notaopprogrammer, thanks for flagging this. Will try to release an updated version soon. Meanwhile maybe you could consider using the prsgis.ml.raster_to_chips module of the pyrsgis Python package?
Hi @notaopprogrammer, thanks for flagging this. Will try to release an updated version soon. Meanwhile maybe you could consider using the prsgis.ml.raster_to_chips module of the pyrsgis Python package?
Hi thanks for the prompt reply. I tried using raster_to_chips module and I believe my RAM is not enough to process the data. Any solution to this? The tif file is ~4 GB in size
Well, it is difficult to determine how big the data is based only on the TIF file size because the file size of the same TIF depends on other variables too, see here.
Btw, about your original question, I suggest you clip your 4GB raster into smaller image pieces (maybe 4?) and then try generating the chips. The other alternate option is that you upload your data on Google Drive and use Google Colab to run pyrsigs because Colab offers ~12GB RAM for free. But again, when you finally arrive at training the model, you will either need enough ram or use smaller data chunks that can fit in the memory.