/tvm-benchmark

A set of benchmarks for TVM

Primary LanguagePython

TVM Benchmark

A benchmark set for TVM. We mainly target the applications that are NOT used for deep learning.

Case List

Clustering

K-Means

A method of vector quantization which is popular for cluster analysis in data mining.

Classification

Logistic Regression

The implementation of multinomial logistic regression using logistic loss gradient decent.

Linear Support Vector Machine

The implementation of binary classification using hinge loss gradient decent.

(TODO: Add regularizer. Now we simply use zero regularizer.)

Regression

Linear Least Square

The implementation of linear regression using squared loss gradient decent.

Image Processing

Image Blurring

Simple image blurring process that generates a 224x244 gray image and blurs it using a 3x3 window.

(TODO: Implement reading images from files.)