hi! Does anybody batch the graphs in the batch processing?
I defined a graph, which can help me do calibration, deburst and filter, and I want to process this graph for several images, so I think it should be processed in the batch.
but after the batch process, there is only one band in the result, so what wrong with it?
Thank you in advance
Each run of the batch processing executes the graph once, but with a different input product.
So for each input product there should be one output product in the target folder. It depends on the defined graph how many bands they have.
Can you please share a screenshot of your batch processing window and upload the graph xml?
so, you mean the output path should be defined in the graph before bath process?
but I want to apply one graph to several images, so can I output them to same fold with different name?
Yes, the target file usually keeps the name of the input file, but you should define a new folder for the outputs of the batch script to avoid that it overwrites the input data.
also, it is good to remove the target directoy in the “Write” tab of the graph. Sometimes the one entered here is taken instead of the Target Folder of the batch window.
I try to remove the ‘write’ tab in the graph, and defined a new folder in the target folder of batch process,but it is strange, there is nothing in the folder. can you recommend me some video information about batch processing in SNAP?
do not remove the Write tab, but remove the target directory and name which is entered in there (it is there because it was automatically generated when you made the graph).
but if the name and directory is empty, the graph can not be saved.
I try it again and it worked. Thanks.
I just delete the target product name when building the graph, the first time I save graph, it says you don’t have a name, can’t save it, but click OK and you can save it again.
load it to the batch process,no change were made in the batch process interface. and then the custom graph batch was successfully implemented.
just for your reference, and thanks @ABraun again.