Please enable JavaScript.

Coggle requires JavaScript to display documents.

PyTorch - Coggle Diagram

- - - - getitem
        
        Implement logic to retrieve a sample from the dataset by index provided to "getitem" function, and return.
      - len
        
        Return the length of the dataset
      - init
        
        Import the dataset here using various methods like readcsv(), etc.
    - - train_iter = iter(trainloader)
      - train_iter.next()
  - - - Constructor
      - Forward
- - - - CPU
        
        Floating Point
        
        16-bit floating Point
        
        torch.float16 / torch.half
        
        32-bit floating Point
        
        torch.float32 / torch.float
        
        64-bit floating Point
        
        torch.float64 / torch.double
        
        Integer
        
        Unsigned
        
        8-bit Unsigned
        
        torch.uint8
        
        Signed
        
        8-bit
        
        torch.int8
        
        16-bit
        
        torch.int16
        
        32-bit
        
        torch.int32
        
        64-bit
        
        torch.int64
        
        Boolean
        
        torch.bool
      - GPU
        
        Same as CPU Tensors
        
        Floating Point
        
        16-bit floating Point
        
        32-bit floating Point
        
        64-bit floating Point
        
        Integer
        
        Unsigned
        
        8-bit Unsigned
        
        Signed
        
        8-bit
        
        16-bit
        
        32-bit
        
        64-bit
        
        Boolean