chrislemke

import functools
import os
import shutil
import subprocess
from contextlib import contextmanager
from typing import Any, Callable, Dict, Iterator, Tuple

from dotenv import load_dotenv

chrislemke

from typing import Tuple

import matplotlib.pyplot as plt
import numpy as np
from numpy.typing import NDArray
import pandas as pd
from sklearn.metrics import (
    ConfusionMatrixDisplay,
    accuracy_score,
    auc,

chrislemke

import torch

# Batch matrix multiplication
X = torch.arange(24).reshape(2, 3, 4)
Y = torch.arange(40).reshape(2, 4, 5)

A = torch.einsum('ijk, ikl->ijl', X, Y)
torch.bmm(X, Y)

print(A)

chrislemke

import torch

# Matrix-Matrix multiplication
X = torch.arange(6).reshape(2, 3)
Y = torch.arange(12).reshape(3, 4)

A = torch.einsum('ij, jk->ik', X, Y)
torch.mm(X, Y)

print(A)

chrislemke

import torch

# Matrix-Vector multiplication
X = torch.rand((3, 3))
y = torch.rand((1, 3))

A = torch.einsum('ij, kj->ik', X, y)
torch.mm(X, torch.transpose(y, 0, 1))  # or torch.mm(X, y.T)

print(A)

chrislemke

import torch

# Outer product
v = torch.rand((3))
t = torch.rand((3))

A = torch.einsum('i, j->ij', v, t)
torch.outer(v, t)

print(A)

chrislemke

import torch

# Dot product
v = torch.rand((3))
c = torch.rand((3))

a = torch.einsum('i, i->', v, c)
torch.dot(v, c)

print(a)  # tensor(0.5750)

chrislemke

import torch

# Element wise multiplication
X = torch.rand((3, 2))
Y = torch.rand((3, 2))

A = torch.einsum('ij, ij->ij', X, Y)
torch.mul(X, Y)  # or X * Y

print(A)

chrislemke

import torch

X = torch.rand((2, 3))

# Row summation
a = torch.einsum('ij->i', X)
torch.sum(X, axis=1)

print(a)  # tensor([1.4088, 1.7803])

chrislemke

import torch

# Summation
X = torch.rand((2, 3))

a = torch.einsum('ij->', X)
torch.sum(X)

print(a) # tensor(3.5585)