【筆記】動手學深度學習-預備知識

預備知識
2.1 資料操作

import torch

x = torch.arange(12)

print(x.shape)

print(torch.Size(x))

print(x.numel())

X = x.reshape(3, 4)

print(X)

print(torch.ones((2, 3, 4)))

print(torch.randn(3, 4))

print(torch.tensor([[2, 1, 3, 4], [1, 2, 3, 4], [3, 4, 5, 6]]))


x = torch.tensor([1.0, 2, 4, 8])
y = torch.tensor([2, 2, 2, 2])
print(x + y, x - y, x * y, x / y, x ** y)  # **運算子是求冪運算

print(torch.exp(x))


X = torch.arange(12, dtype=torch.float32).reshape((3,4))
Y = torch.tensor([[2.0, 1, 4, 3], [1, 2, 3, 4], [4, 3, 2, 1]])
print(torch.cat((X, Y), dim=0))
print(torch.cat((X, Y), dim=1))

print(X == Y)

print(X.sum())

a = torch.arange(3).reshape((3, 1))
b = torch.arange(2).reshape((1, 2))
print(a)
print(b)
print(a+b)

print(X[-1])
print(X[1:3])

X[1, 2] = 9
print(X)

X[0:2, :] = 12
print(X)

before = id(Y)
Y = Y + X
print(id(Y) == before)

Z = torch.zeros_like(Y)
print('id(Z):', id(Z))
Z[:] = X + Y
print('id(Z):', id(Z))

A = X.numpy()
B = torch.tensor(A)
print(type(A))
print(type(B))

a = torch.tensor([3.5])
print(a)
print(a.item())
print(float(a))
print(int(a))

2.2 資料預處理

import pandas as pd

data = pd.read_csv(data_file)
inputs, outputs = data.iloc[:, 0:2], data.iloc[:, 2]
inputs = inputs.fillna(inputs.mean())
print(inputs)

import torch

X = torch.tensor(inputs.to_numpy(dtype=float))
y = torch.tensor(outputs.to_numpy(dtype=float))
print(x)
print(y)

2.3 線性代數

import torch

x = torch.tensor(3.0)
y = torch.tensor(2.0)

print(x + y)
print(x * y)
print(x / y)
print(x**y)

x = torch.arange(4)
print(x)

print(x[3])

print(len(x))

print(x.shape)

A = torch.arange(20).reshape(5, 4)
print(A)

print(A.T)

B = torch.tensor([[1, 2, 3], [2, 0, 4], [3, 4, 5]])
print(B)

print(B == B.T)

print(X = torch.arange(24).reshape(2, 3, 4))
print(X)


A = torch.arange(20, dtype=torch.float32).reshape(5, 4)
B = A.clone()  # 透過分配新記憶體，將A的一個副本分配給B
print(A)
print(A + B)

print(A * B)

a = 2
X = torch.arange(24).reshape(2, 3, 4)
print(a + X)
print((a * X).shape)

x = torch.arange(4, dtype=torch.float32)
print(x)
print(x.sum())

A_sum_axis0 = A.sum(axis=0)
print(A_sum_axis0)
print(A_sum_axis0.shape)

A_sum_axis1 = A.sum(axis=1)
print(A_sum_axis1)
print(A_sum_axis1.shape)

print(A.sum(axis=[0, 1]))

print(A.mean())
print(A.sum())
print(A.numel())

print(A.mean(axis=0), A.sum(axis=0) / A.shape[0])

sum_A = A.sum(axis=1, keepdims=True)
print(sum_A)

print(A / sum_A)

print(A.cumsum(axis=0))

y = torch.ones(4, dtype = torch.float32)
print(x)
print(y)
print(torch.dot(x, y))
print(torch.sum(x * y))

print(A.shape)
print(x.shape)
print(torch.mv(A, x))

B = torch.ones(4, 3)
print(torch.mm(A, B))

u = torch.tensor([3.0, -4.0])
print(torch.norm(u))

print(torch.abs(u).sum())

print(torch.norm(torch.ones((4, 9))))