cyberdoggos / generator Goto Github PK

Hello how can i solve this warn?
Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.

I changed the resolution through config-norm yml, and added the new trait named as same as old names, this only two changes that I made, it gives me only 1/1000 img then this error appears, can you please help me, thanks.

[Running] python -u "c:\Users*\Documents*\generator-main\generate.py"
Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.
Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.
Traceback (most recent call last):
File "c:\Users***\Documents\ *\generator-main\generate.py", line 192, in
main()
File "c:\Users*\Documents\ ****\generator-main\generate.py", line 181, in main
gen.generate_normal(i, mistakes)
File "c:\Users\ ****\Documents\ ****\generator-main\generate.py", line 30, in generate_normal
image, ids[idx], name, value, trait = self.add_layer(image, layer_id, layer, traits)
File "c:\Users\ ****\Documents\ ****\generator-main\generate.py", line 102, in add_layer
image = alpha * layer + (1 - alpha) * image
ValueError: operands could not be broadcast together with shapes (100,100,0) (100,100,3)

hey,
thanks dude for sharing the code.
after I added some traits and changed most of the names, also I am sure about prop is equal 1
error.txt

Hi,

I have been trying to use your code, but I am having troubles with this error.

Could you help me?

Hi my file size is already 3000 / 3000 pixels but when I run script, it generates the images twice, one normal while the other, it appends _large.

How can I stop it from creating the large one? Thanks