Learning Pytorch for Mathematical computation

Commute the n-dimentional matrix (tensors)

Pytorch: An tool with easy to use API for higher-dimention math computaion.

torch.clamp()
torch.sigmoid()
torch.std()
torch.std_mean()
torch.unique()

# Import torch and other required modules
#!conda install -y pytorch -c pytorch 
import torch

tourch.clamp(input_tensor,min_value,max_value,output_tensor)

This function helps in clamping the matrix's values in the given range(min,max).

### tensor representing the data(sweets): sweet seasonality(True/False), sweet current demand (in tons), sweet price (INR/kg)
t = torch.Tensor([[True, 20, 600],
                  [False, 15, 360],
                  [True, 50, 960],
                  [False, 55, 1660],
                  [True, 35, 600]])

## validate the tensor
t

tensor([[1.0000e+00, 2.0000e+01, 6.0000e+02],
        [0.0000e+00, 1.5000e+01, 3.6000e+02],
        [1.0000e+00, 5.0000e+01, 9.6000e+02],
        [0.0000e+00, 5.5000e+01, 1.6600e+03],
        [1.0000e+00, 3.5000e+01, 6.0000e+02]])