TypeError: super(type, obj): obj must be an instance or subtype of type

[kumarshrestha009]

TypeError
in ()
1 import UnariesNet
----> 2 uNet = UnariesNet.unariesNet()

/home/kumar/DeepOcclusion/UnariesNet.py in init(self, load_pretrained)
80 c53_r = mNet.c53_r
81
---> 82 op = ROIPoolingOp(pooled_h=p_h, pooled_w=p_w, spatial_scale=1.0)
83
84

/home/kumar/DeepOcclusion/roi_pooling.py in init(self, pooled_h, pooled_w, spatial_scale)
18
19 def init(self, pooled_h, pooled_w, spatial_scale):
---> 20 super(ROIPoolingOp, self).init(self.func_file,
21 self.func_name)
22

TypeError: super(type, obj): obj must be an instance or subtype of type

How can i fix this ? Help will be highly appreciated ?

[DhanasekarChellamuthu]

Hi @kumarshrestha009, before that in same roi_pooling.py 7 th line from theano.sandbox.cuda import as_cuda_ndarray_variable, GpuOp i am getting error like cannot import name 'as_cuda_ndarray_variable' #38. can you help me to resolve this issue
I am having doubt at Dependencies 2 nd point:

You also need to install the theano ROI-Pooling Layer from https://github.com/ddtm/theano-roi-pooling to get the unary potentials to work.

can you comment system dependencies

[kumarshrestha009]

You can type
import theano.sandbox.cuda as cuda_ndarray_variable

Theano is the main dependency. I had used theano 0.9.

[DhanasekarChellamuthu]

Thanks for answering @kumarshrestha009 but you said like calling whole package in the name as_cuda_ndarray_variable but in that code it is written as calling one function from that package that both make different meaning right.

[Kirmani66]

@kumarshrestha009 Did you solved the super(type, obj) issue. If yes please share the solution

[leconghieu99]

@DhanasekarChellamuthu can you please send me the .zip files. Looks like the link is broken. Email: [email protected]