render() return TensorXf values outside the range [0.0,1.0]

[nebnes]

After rendering, I get a TensorXf with values outside the range [0.0,1.0]
I get several values above 1.0 around 94.00
It seems out of range values match overly bright pixels. Is it correct to clip the tensor 's high values back to 1.0 ?
np.clip(img,0.0,1.0)

Is it correct ?

Thanks in advance

[bathal1]

The output of a rendering in Mitsuba is in a linear color space, so there is no reason for the values to be in the [0, 1] range.

What you do to get it to this range entirely depends on what you intend to do with the image after rendering. For example, in order to write it in a common format like JPEG or PNG, one would apply tonemapping to convert it to the correct color space.

Why do you need values to be in that range ?

[nebnes]

@bathal1
I didn't particularly expect to be in this range. It's just that I had some big gaps in a rendering. I understand now, thank you.

[bathal1]

If you want to know more about how images are handled in Mitsuba, I recommend you take a look at the image manipulation tutorial. The "Bitmap conversion" section might be of particular interest to you.

[nebnes]

I'll take a look
thanks @bathal1