manishkumarkeshri Goto Github PK
Name: Manish Kumar Keshri
Type: User
Company: Dell EMC
Bio: writing code.
Twitter: manish_k_keshri
Location: United States
Blog: https://www.linkedin.com/in/manish-kumar-keshri-b094158b/
Name: Manish Kumar Keshri
Type: User
Company: Dell EMC
Bio: writing code.
Twitter: manish_k_keshri
Location: United States
Blog: https://www.linkedin.com/in/manish-kumar-keshri-b094158b/
There exists a puzzle state which comprises of 8 numbers and an empty slot E randomly arranged. The goal state is to reach the puzzle state that will have the numbers arranged in order with the 0 coming after all other numbers have been arranged. For example, consider the initial puzzle state as (E 1 3 4 2 5 7 8 6), then the final or the goal state will be (1 2 3 4 5 6 7 8 E). We should be using A* algorithm to solve the 8-puzzle problem with the heuristic being the total number of misplaced tiles in the given puzzle state. The empty slot can move only once to generate the new puzzle state.
Built an LSTM model to predict apple share price. Trained it past 6 months daily price data.
Implementation of Haar-Cascades algorithm for frontal face and eyes detection and Fisherfaces algorithm for face recognition using OpenCV C++ Library
Iteratively Re-Weighted Least Squares method based Logistic method
Achieved speed-up of 402x in similaritymatrix calculation on NVIDIA GTX 1080; used coalescing, thread coarsening and shared memory tiling.
• Implemented FP Tree algorithm to find frequently purchased itemsets and corresponding association rules in more than 63000 transactions using C++ • Compared results for different values of minimum support and confidence.
Built a deep neural network with Batch Normalization and tuned Hyperparameters that would facilitate communications from a speech-impaired person to someone who doesn’t understand sign language (Used: Python: scikit-learn, TensorFlow) - Used dataset of images provided by Coursera to train to test the algorithm (Accuracy on -Train: 0.91, -Test: 0.84)
Notes and experiments to understand deep learning concepts
Built a deep network, and applied it to classify images of cat vs non-cat.
Simulated a geo-distributed recommendation system using a decentralized SVD algorithm
CNN based image recognition model using tflearn
• Explored dataset with 146 data points (i.e. ”employees of Enron”), each of which has 21 features( i.e. email info, salary, compensation etc.) to find the fraud. • Implemented features selection, outliers removal, classification and validation. • Compared confusion matrices and F1 scores of Decision tree, SVM and AdaBoost classifiers (Used: Python).
Implemented feed forward neural network with backpropagation to recognize the digit in each image
Built a model uses Keras, a deep learning library, to generate jazz music. Specifically, it builds an LSTM network, learning from the given MIDI file. It uses deep learning to make music -- something that's considered as deeply human.
There are 15 Missionary, 15 Cannibal on left side of the river. There is a boat which can carry 6 people at most. The goal is to safely transfer all the missionary and cannibals to the right side of the river. By safely we mean that at any instant like the left side, right side or even on the boat the no. of cannibal should not be greater than the no. of cannibals
Resources for "Natural Language Processing" Coursera course.
Built an NLP system using RNN with LSTM units to assign an emoji to a sentence. Used 50-dimensional GloVe Word Embedding as the features for sentence words and emoji's were assigned using 5 directional SoftMax output.
Pravega - Streaming as a new software defined storage primitive
Pravega Kubernetes Operator
Sample Applications for Pravega.
Tooling for Pravega
Computation using data flow graphs for scalable machine learning
The Incredible PyTorch: a curated list of tutorials, papers, projects, communities and more relating to PyTorch.
Constructed a speech dataset and implemented an algorithm for trigger word detection (sometimes also called keyword detection, or wakeword detection). Trigger word detection is the technology that allows devices like Amazon Alexa, Google Home, Apple Siri, and Baidu DuerOS to wake up upon hearing a certain word.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.