Skip to content
Branch: master
Find file History
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
dbscan Use enclave_info.toml instead of txt (#155) Dec 14, 2019
gaussian_mixture_model Use enclave_info.toml instead of txt (#155) Dec 14, 2019
gaussian_processes Use enclave_info.toml instead of txt (#155) Dec 14, 2019
gbdt Use enclave_info.toml instead of txt (#155) Dec 14, 2019
gen_linear_model Use enclave_info.toml instead of txt (#155) Dec 14, 2019
image_resizing Use enclave_info.toml instead of txt (#155) Dec 14, 2019
kmeans Use enclave_info.toml instead of txt (#155) Dec 14, 2019
lin_reg Use enclave_info.toml instead of txt (#155) Dec 14, 2019
logistic_reg Use enclave_info.toml instead of txt (#155) Dec 14, 2019
naive_bayes Use enclave_info.toml instead of txt (#155) Dec 14, 2019
neural_net Use enclave_info.toml instead of txt (#155) Dec 14, 2019
online_decrypt Use enclave_info.toml instead of txt (#155) Dec 14, 2019
private_join_and_compute Use enclave_info.toml instead of txt (#155) Dec 14, 2019
py_file Use enclave_info.toml instead of txt (#155) Dec 14, 2019
py_logistic_reg Use enclave_info.toml instead of txt (#155) Dec 14, 2019
py_matrix_multiply Use enclave_info.toml instead of txt (#155) Dec 14, 2019
quickstart Rewrite quickstart example command line interface (#186) Dec 20, 2019
quickstart_c Use enclave_info.toml instead of txt (#155) Dec 14, 2019
rsa_sign Use enclave_info.toml instead of txt (#155) Dec 14, 2019
svm Use enclave_info.toml instead of txt (#155) Dec 14, 2019
README.md Implement an example about Logistic Regression in MesaPy (#58) Nov 18, 2019

README.md

Examples

This directory contains several examples to illustrate the application scenarios. The case studies document contains more descriptive information.

  • quickstart
    • Quickstart demo of how to invoke MesaTEE services. This can be a great starting point for new users.
  • image_resizing
    • One can invoke MesaTEE services similar to AWS Lambda. On data uploading or new events coming, MesaTEE function services are immediately triggered. For example, you can use MesaTEE to thumbnail images, transcode videos, index files, process logs, validate content, and aggregate and filter data in real-time. In this specific example, we demonstrate image resizing.
  • gbdt
    • MesaTEE also supports a variety of big data analyses and machine learning algorithms, such as GBDT, Linear Regression, as well as neural networks. In this specific example, we demonstrate how to utilize a GBDT model to perform data prediction -- in the trusted secure fashion, without concerning privacy leakage.
  • rsa_sign and online_decrypt
    • Another killing feature of MesaTEE is to serve as a key vault or an HSM. MesaTEE can conveniently provide secret management (securely store and control accesses to tokens, passwords, certificates, API keys, and other secrets), key management (create and control encryption keys), and certificate management (provision and manage certificates).
  • private_join_and_compute
    • When cross-department or cross-company data collaboration happens, privacy concerns arise. Thus secure multi-party computation (SMC) has become more and more important nowadays to enable joint big data analyses. However, traditional crypto-based SMC has quite a few limitations, and MesaTEE can solve them effectively, with way better performance/flexibility improvements. Details are discussed here.
  • py_matrix_multiply
    • In the era of FaaS and AI, Python rules them all. So we have another dedicated project called MesaPy. In this specific example, we demonstrate how to invoke the MesaPy engine integrated into MesaTEE.
  • py_logistic_reg
    • MesaTEE supports secondary AI development all in Python with the help of MesaPy Engine. Here is an example about logistic regression model including training and prediction.
  • DBSCAN
    • Provides an implementaton of DBSCAN clustering.
  • Generalized Linear Model
    • Contains implemention of generalized linear models using iteratively reweighted least squares.The model will automatically add the intercept term to the input data.
  • Gaussian Mixture Models
    • Provides implementation of GMMs using the EM algorithm.
  • Gaussian Processes
    • Provides implementation of gaussian process regression.
  • Linear Regression
    • Contains implemention of linear regression using OLS and gradient descent optimization.
  • Logistic Regression
    • Contains implemention of logistic regression using gradient descent optimization.
  • Naive Bayes Classifiers
    • The classifier supports Gaussian, Bernoulli and Multinomial distributions.A naive Bayes classifier works by treating the features of each input as independent observations. Under this assumption we utilize Bayes' rule to compute the probability that each input belongs to a given class.
  • Neural Network
    • Contains implementation of simple feed forward neural network.
  • Support Vector Machine
    • Contains implementation of Support Vector Machine using the Pegasos training algorithm.The SVM models currently only support binary classification. The model inputs should be a matrix and the training targets are in the form of a vector of -1s and 1s..
You can’t perform that action at this time.