cuda

Perceptrons¶ Implementation of the classic Perceptron by Frank Rosenblatt for binary classification (here: 0/1 class labels) in PyTorch Imports¶ In [1]: import numpy as np import matplotlib.pyplot as plt import torch %matplotlib inline Preparing a toy dataset¶ In [2]: ########################## ### DATASET ########################## data = np.genfromtxt(‘perceptron_toydata.txt’, delimiter=’\t’) X, y = data[:, :2], data[:, 2] y = y.astype(np.int) […]

Solving the XOR Problem¶ Toy Dataset¶ In [16]: import torch import torch.nn.functional as F import pandas as pd import time import matplotlib.pyplot as plt %matplotlib inline RANDOM_SEED = 123 DEVICE = (‘cuda:0’ if torch.cuda.is_available() else ‘cpu’) In [17]: df = pd.read_csv(‘xor.csv’) X = df[[‘x1’, ‘x2’]].values y = df[‘class label’].values In [18]: plt.scatter(X[y==0, 0], X[y==0, 1], marker=’o’) plt.scatter(X[y==1, 0],

Logistic Regression with L2 Regularization¶ Implementation of classic logistic regression for binary class labels. Imports¶ In [ ]: %matplotlib inline import matplotlib.pyplot as plt import numpy as np import torch import torch.nn.functional as F device = torch.device(‘cuda:0’ if torch.cuda.is_available() else ‘cpu’) LAMBDA = 2 Preparing a toy dataset¶ In [ ]: ########################## ### DATASET ########################## data = np.genfromtxt(‘data/toydata.txt’, delimiter=’\t’)

Matrix Multiplication using SUMMA and Canon CMPSC 450 Matrix Multiplication Exploded (CUDA?) *=*=*= *=*=*= *=*=*= CMPSC 450 SUMMA Step 1 *=*=*= *=*=*= *=*=*= Locally Owned A Locally Owned B Mult Input A (temp) Mult Input B (temp) Partial Locally Owned Answer (1/3) CMPSC 450 SUMMA Step 2 *=*=*= *=*=*= *=*=*= Locally Owned A Locally Owned