Closed Zhenye-Na closed 3 years ago
Hi @Zhenye-Na, thanks for your feedback. Can you try to restart the runtime and run from the beginning cell?
thans for the reply, I will try to re-run the colab notebook, will attach the error here if it appears again,
still run into the same error:
initialize network with normal
loading renderer from pre-trained checkpoint...
begin drawing...
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-12-8f14c24057d5> in <module>()
1 pt = ProgressivePainter(args=args)
----> 2 final_rendered_image = optimize_x(pt)
<ipython-input-11-25a3da86e7fa> in optimize_x(pt)
51 pt.step_id += 1
52
---> 53 v = pt._normalize_strokes(pt.x)
54 v = pt._shuffle_strokes_and_reshape(v)
55 PARAMS = np.concatenate([PARAMS, v], axis=1)
AttributeError: 'ProgressivePainter' object has no attribute 'x'
maybe it is the size of the image I choose? you can review the colab notebook I used here
It seems to be the problem of values below --max_m_strokes 1000 --max_divide 15 I tried some different values, and I guess the max_m_strokes value has to be divisible by (max_divide * 50)? If you try --max_m_strokes 1500, that might work.
Hello Zhengxia,
Thank you for creating this project, very interesting!
I spent some time playing around with the colab runtime 1, I tried to increase the default hyper-parameters by multiplying
2
, so far so good.However,
I have updated the
args.canvas_size
,args.max_m_strokes
by multiplying2
, whileargs.max_divide
from10
to15
, then in the "drawing" phase:It throws error like this:
Do you know why this error is thrown? Personally, I think maybe this is because that there should be some "math formula" behind these three (or more) parameters ?
What do you think?
Thanks!