Peri Akiva's repositories
RoboBull
A HackRU winner project. a user-friendly web-interface robo-adviser for novice traders that want to focus on day and short term trading. Project is using candlestick and moving average algorithms to predict suggestions (buy, sell, hold) for users. Product is also notifying users via SMS when the suggestion is changed; users are able to subscribe to multiple stocks in order to get real time suggestions on those chosen stocks. The algorithm is using real time data and provides real time visualization flowing chart for chosen stocks. Used Python, Flask, MongoDB, Node.js.
finding_berries
PyTorch implementation of the paper "Finding Berries: Segmentation and Counting of Cranberries using Point Supervision and Shape Priors". Peri Akiva, Kristin Dana, Peter Oudemans, Michael Mars. CVPRW2020.
semanticAnalysis
Analysis of sentiment resulting from public figure's statements on social media against his/her public approval ratings. Sentiment was extracted and classified using scarpped data, Naive Bayes classifier and linear SVM. Both classifiers were used and compared for benchmark purposes. Used Pandas, sklearn, and Python.
Bag-of-Words
Bag of Words implementation using SVM and SIFT
Parallel-Quicksort
Parallel implementation for quicksort using pthreads
DigitalEqualizer
Matlab implementation for a digital equalizer.
generate_dots
generating random dots on an image and the option to save it
Google-Interview-Dev-Guide-Sols
Python solutions to google dev guide solutions
Image_Cropping_and_Stitching
cropping and stitching images
locating-objects-without-bboxes
PyTorch code for "Locating objects without bounding boxes" - Loss function and trained models
mmpose
OpenMMLab Pose Estimation Toolbox and Benchmark.
Parallel-Cube
Cuda implemendation to cubing every variable in an integer array
PyMesh
Geometry Processing Library for Python
Sparse-To-CSR
Taking sprae matrix and convert it to compressed row storage
Take-A-Break
A program that works in the background that reminds the user to take a break every 2 hours with a lifting video
TCN_HPL
PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡
topic-modeling-presidents
The Most Important Topics in United States History According to Past Presidents using K-Means and LDA