search

BrandonSmithJ / MDN

Mixture Density Network for water constituent estimation
GNU General Public License v3.0
40 stars 34 forks source link

I have some problems with the running of the program #11

Closed worldso closed 2 years ago

worldso commented 2 years ago

Recently downloaded and debugged your project, I found that I have some problems in running, what is the input and output of the program and how to make the program work normally，If possible, can you leave your email address to facilitate our communication?

worldso commented 2 years ago

AssertionError: No datasets found for ['chl'] within D:\Data\Insitu/*/OLI

worldso commented 2 years ago

what meaning this is assert(len(data_folder)), f'No datasets found for {products} within {data_path}//{sensor}' assert(len(data_keys)), f'No variables found for {products} within {data_path}//{sensor}'

BrandonSmithJ commented 2 years ago

There is an example in the readme on how to use the program. Inputs are meant to be Rrs, outputs are water quality products (chlorophyll, TSS, etc.).

The assertion error means you are trying to train a new model using the products interface, which won't work unless you have Rrs/chl data. If you just want a general MDN, look at just using the MDN model and how it is fit within get_estimates.py.

worldso commented 2 years ago

There is an example in the readme on how to use the program. Inputs are meant to be Rrs, outputs are water quality products (chlorophyll, TSS, etc.).

The assertion error means you are trying to train a new model using the products interface, which won't work unless you have Rrs/chl data. If you just want a general MDN, look at just using the MDN model and how it is fit within get_estimates.py.

I can't understand the meaning of "//" in the path "AssertionError: No datasets found for ['chl'] within D:\Data\Insitu//OLI", I put the data in "D:\Data\Insitu\ OLI" program still getting the same assert error

worldso commented 2 years ago

I am a Chinese university student, and our team is also doing research similar to your team, using deep learning to analyze remote sensing data and then make a judgment on water quality, but I am not very proficient in python, and there are still many questions to be asked. seniors to answer. Thank you very much.

BrandonSmithJ commented 2 years ago

There should not be // in your path. I believe you're missing the 'location' folder in your data path, i.e. D:/Data/Insitu/location/OLI - where location is meant to signify the name of the dataset. You can have multiple datasets within the Insitu folder, and all of the ones which contain the parameters of interest you are trying to train on (e.g. chl) will be used.