Python

Search

pickle.load from gpu device to cpu

#if you are using pickle load on a cpu device, from a gpu device
#you will need to override the pickle load functionality like this:
#note: if loading directly in torch, just add map_location='cpu' in load()

class CPU_Unpickler(pickle.Unpickler):
    def find_class(self, module, name):
        if module == 'torch.storage' and name == '_load_from_bytes':
            return lambda b: torch.load(io.BytesIO(b), map_location='cpu')
        else: return super().find_class(module, name)

#contents = pickle.load(f) becomes...
contents = CPU_Unpickler(f).load()

Comment

PREVIOUS	NEXT

Code Example
Python :: how to draw threshold line in bar graph python
Python :: pd df replace
Python :: python loop nest shorthand
Python :: sqlalchemy create engine Oracle
Python :: how to use ActionChains selenium python with WebDriverWait
Python :: how to customize simplejwt error response message in django restframework
Python :: how to get maximum value of number in python
Python :: how to slice a set in python
Python :: query set
Python :: plotly ylog
Python :: #Check if list1 contains all elements of list2 using all()
Python :: pil saves blue images
Python :: pymongo dynamic structure
Python :: flask tutorial
Python :: pandas array of dataframes
Python :: pandas sequential numbering within group
Python :: get the first element that is larger than
Python :: python boucle for
Python :: how to encrypt and decrypt strings python
Python :: pyqt popup yes no
Python :: printing coloured and bold text in python
Python :: round to the nearest 0.5
Python :: stripping whitespace in python
Python :: print(f ) python
Python :: noob python
Python :: python remove table widget numbers
Python :: remove items from list while iterating python
Python :: django table view sort filter
Python :: from future import division
Python :: 2 functions at a time py

Search

PYTHON

pickle.load from gpu device to cpu

ADD CONTENT