Pinned Repositories
autotrader
Classify-bio-images-For-Protein-Localization-using-AL
Most proteins localize to specific regions where they perform their biological function. Fluorescent microscopy can reveal the subcellular localization patterns of tagged proteins. The goal of this project is to use active learning to build a classifier that capable of classifying bioimages (encoded as feature vectors) according to subcellular localization patterns. There are three data pools: Easy: A low-noise data pool Moderate: This pool has some noise (labels and features) Difficult: The points in this pool have a larger number of features than those in the easy and moderate pools. Some of these features are irrelevant. Your algorithm will need to perform active learning and feature selection. Each data pool consists of 4120 training images and 1000 test images. Each image is represented as a feature vector (you do not need to do feature extraction yourself). There are 8 subcellular localization patterns: (i) Endosomes; (ii) Lysosomes; (iii) Mitochondria; (iv) Peroxisomes; (v) Actin; (vi) Plasma Membrane; (vii) Microtubules; and (viii) Endoplasmic Reticulum. The data are based on those released by Dr. Nicholas Hamilton for his paper Statistical and visual differentiation of high throughput subcellular imaging, N. Hamilton, J. Wang, M.C. Kerr and R.D. Teasdale, BMC Bioinformatics 2009, 10:94. Select and implement a suitable active learning algorithm and apply it to the training data. Additionally, implement a random learner that selects random images in the training data. Using a budget of 2,500 calls to the oracle, compute and plot the test errors for each algorithm as a function of the number of calls to the oracle. Use the test data to compute the test errors. Repeat this for the easy and moderate data pools. If you are working on a team, or want extra credit, apply your algorithm to the difficult pool as well.
forensic-audio-analysis
MLSP - 2017
Frame-Level-Speech-Recognition-with-Neural-Networks
Kaggle_HomeDepot
Turing Test's Solution for Home Depot Product Search Relevance Competition on Kaggle (https://www.kaggle.com/c/home-depot-product-search-relevance)
LSTM_TextGeneration
Personal-Website
Personal Website being hosted at www.andrew.cmu.edu/user/ramesho/
USNewsScraping
Scraping through Health US News Search Results
RameshOswal's Repositories
RameshOswal/USNewsScraping
Scraping through Health US News Search Results
RameshOswal/Personal-Website
Personal Website being hosted at www.andrew.cmu.edu/user/ramesho/
RameshOswal/alpa
Training and serving large-scale neural networks
RameshOswal/awesome-open-mlops
The Fuzzy Labs guide to the universe of open source MLOps
RameshOswal/binderhub
Run your code in the cloud, with technology so advanced, it feels like magic!
RameshOswal/cml
♾️ CML - Continuous Machine Learning | CI/CD for ML
RameshOswal/cookiecutter-pypackage
Cookiecutter template for a Python package.
RameshOswal/DeepSpeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
RameshOswal/demo-self-driving
Streamlit app demonstrating an image browser for the Udacity self-driving-car dataset with realtime object detection using YOLO.
RameshOswal/easy-application
Over 400 software engineering companies that are easy to apply to
RameshOswal/FAQChatbot
Set of scripts to build a chatbot which will answer based on the FAQs supplied.
RameshOswal/industry-machine-learning
A curated list of applied machine learning and data science notebooks and libraries across different industries.
RameshOswal/kafka-python
Python client for Apache Kafka
RameshOswal/langchain
⚡ Building applications with LLMs through composability ⚡
RameshOswal/machine-learning-engineering-for-production-public
Public repo for DeepLearning.AI MLEP Specialization
RameshOswal/machine-learning-systems-design
A booklet on machine learning systems design with exercises
RameshOswal/matminer
data mining for Materials Science
RameshOswal/mltrace
Coarse-grained lineage and tracing for machine learning pipelines.
RameshOswal/monk_v1
Monk is a low code Deep Learning tool and a unified wrapper for Computer Vision.
RameshOswal/nsetools
Realtime Data From National Stock Exchange (India)
RameshOswal/ocp
https://opencatalystproject.org/
RameshOswal/PyPDF2
A utility to read and write PDFs with Python
RameshOswal/python-sample-package
Python seed for writing a simple script of python package
RameshOswal/python_cmd_pkg_template
Template Repo to make Python Cmd line pkgs.
RameshOswal/ramesh_helpers
RameshOswal/rameshoswal.github.io
RameshOswal/redner
A differentiable Monte Carlo path tracer
RameshOswal/setu-python-sdk
SDK to help Pythonistas integrate with Setu's APIs
RameshOswal/tapas
End-to-end neural table-text understanding models.
RameshOswal/toy-ml-pipeline
Toy example of an applied ML pipeline for me to experiment with MLOps tools.