open-solution-mapping-challenge
open-solution-mapping-challenge copied to clipboard
neptune-ai
UnicodeDecodeError: 'gbk' codec can't decode byte 0xa5 in position 404: illegal multibyte sequence
Hi, I'm sorry to bother you. There are some problems in the replication experiment, and the description of the problem is
reader.py，line 178, in update_raw data = self.stream.read(size) UnicodeDecodeError: 'gbk' codec can't decode byte 0xa5 in position 404: illegal multibyte sequence
i don't know how to deal with this problem,would you mind giving me some advices?
Hi there @pnzsl and sorry for late reply. It looks like something is wrong with the files that are stored on your storage.
Hmm, could you check that you can read the images with no problems in a notebook or something?
Hi there @pnzsl and sorry for late reply. It looks like something is wrong with the files that are stored on your storage.
Hmm, could you check that you can read the images with no problems in a notebook or something?
I have solved this problem.thanks!
Hi，sorry to occupy your rest time again. hmm，I have completed the steps of mask and metadata. I have encountered some troubles when training the train data to obtain the model.
well， I can’t adjust the parameter of a. According to my understanding, a is a parameter in the setup function. How can I adjust it？please look at the following question.
looking forward to your reply！
@pnzsl where do you adjust/change the a?
I cannot find it in the neptune.yaml config file.
As shown in the figure, the a is in the generic. py file. It's just a formal parameter in a function. The computer is not with me for the time being. I will provide you with specific information tomorrow. thank you sir.
It may be that your training/validation data is of 0 size.
Hmm, I suggest that you go to the https://github.com/neptune-ml/open-solution-mapping-challenge/blob/master/src/pipeline_manager.py#L98 and print the shapes of train/valid just to make sure that those make sense.
I think we need to make sure that the data which goes into the:
pipeline.fit_transform(data)
actually makes sense first.
I suspect that there is something fishy going on here.