PyTorch functions

[nishnik]

• torch.empty(5, 3)
• torch.rand(5, 3)
• torch.zeros(5, 3, dtype=torch.long)
• torch.tensor([5.5, 3])
• x = x.newones(5, 3, dtype=torch.double) # new* methods take in sizes x = torch.randn_like(x, dtype=torch.float) # override dtype!
• torch.add(x, y) y.add_(x)
• x = torch.randn(4, 4)
• y = x.view(16)
• z = x.view(-1, 8) # the size -1 is inferred from other dimensions
• x = torch.randn(1) print(x.item())
• b = a.numpy() # The Torch Tensor and NumPy array will share their underlying memory locations (if the Torch Tensor is on CPU), and changing one will change the other.
• a = np.ones(5) b = torch.from_numpy(a) np.add(a, 1, out=a) # changing the np array changed the Torch Tensor automatically
• device = torch.device("cuda") # a CUDA device object y = torch.ones_like(x, device=device) # directly create a tensor on GPU x = x.to(device) # or just use strings .to("cuda")
• torch.cuda.is_available()

Autograd

• x = torch.ones(2, 2, requires_grad=True)
• y = x + 2 # y was created as a result of an operation, so it has a grad_fn.

[nishnik]

• Scikit Cheat Sheet
• Scikit Example : nishnik/GreatStep-Data-Analytics
• PyTorch Cheat Sheet
• PyTorch mnist example
• PyTorch nishnik/LSTMs-with-Attention-For-Aggression-Detection