amithasen Goto Github PK

feature-engineering

fine-tune-gpt3-model

How you can fine-tune a GPT-3 model with Python with your own data

forecasting-furniture-sales_prophet_with_monthly-sampling

furniture-sales-forecasting

A furniture sales data was provided for each month from 2014 to 2017. Time series forecasting was done using Simple Exponential smoothing, Holt-Winters and exponential smoothing and each model was compared to find out the best acting model. After creating a model the future sales were predicted based on the model selected.

hands-on-feature-engineering-with-python

intermittent-demand-forecasting

Predicting the Sales using Time-series forecasting for month-wise data.

jira-similar-issue-finder-app

A JIRA Bot that can train a machine learning model and comment related JIRA IDs on a list of JIRA issues.

machine-learning-in-project-management

This is a supervised machine learning model to identify and visualize the efficiency of a task aimed at project managers

ml-and-ai

Machine Learning and Artificial Intelligence related projects

mlnd-project-capstone-predicting-device-failures

Udacity machine learning engineer capstone. Using ensmeble models to predict device failures.

model-selection

K Fold Cross Validation and Grid Search for Hyper-parameter fine tuning

mydatascienceportfolio

Applying Data Science and Machine Learning to Solve Real World Business Problems

pneumonia_detection

This is a simplified version of a pneumonia detection using a chest xray dataset with the inceptionv3 image classifier.

pneumoniadetection

RSNA Pneumonia Detection Competition on Kaggle

product-demand-forecasting-using-ml

Using machine learning to solve one of the most common problem of Supply Chain domain, i.e Demand Forecasting.

product-recommendation-system

Product recommendation system on Amazon product dataset using Apache Spark framework

resnets-transfer-learning

Transfer Learning with Resnet trained on ImageNet Dataset

retail-demand-forecasting-model-using-factorization-machines

It is challenging to build useful forecasts for sparse demand products. If the forecast is lower than the actual demand, it can lead to poor assortment and replenishment decisions, and customers will not be able to get the products they want when they need them. If the forecast is higher than the actual demand, the unsold products will occupy inventory shelves, and if the products are perishable, they will have to be liquidated at low costs to prevent spoilage. The overall objective of the model is to use the retail data which provides us with historic sales across various countries and products for a firm. We use this information given, and make use of FM’ s to predict the sparse demand with missing transactions. The above step then enhances the overall demand forecast achieved with LSTM analysis. As part of the this project we answered the following questions: How well does matrix factorization perform at predicting intermittent demand How does matrix factorization approach improve the overall time-series forecasting

retail-giant-sales-forecasting-assignment

Forecast the sales for the next 6 months (test data) for that market segment.

retail_sales_index_forecasting

Time series forecasting applied to the prediction of Singapore's monthly retail sales index

rsna-pneumonia-detection

rsna-pneumonia-detection-challenge

In this competition, you’re challenged to build an algorithm to detect a visual signal for pneumonia in medical images. Specifically, your algorithm needs to automatically locate lung opacities on chest radiographs.

rsna-pneumonia-detection-challenge-1

Kaggle RSNA-Pneumonia-Detection-challenge

rsna-pneumonia-detection-challenge-2

sales-forecasting

Forecast sales data and comparing forecasting models such as moving average, exponential smoothing, and ARIMA.

sales-forecasting-using-time-series-analysis

Business Case of Deere & Co. Deere and copmany forecast higher sales of machinery in the next financial year as the world’s largest tractor manufacturer downplayed the impact of the U.S.-China trade war on soybean prices. Deere also forecast its equipment sales will rise by about 30 percent in the current fiscal year. The company expects farmers’ net returns per acre in 2019 will rise as much as 20 percent to the highest level in about five years, Chief Finance Officer Rajesh Kalathur said on the call. Now with this challenging demand, we need data science team to help them Deere is a tractor and farm equipment manufacturing company, was established in 1838. The company has shown a consistent growth in its revenue from tractor sales since its inception. However, over the years the company has struggled to keep it’s inventory and production cost down because of variability in sales and tractor demand. The management at PowerHorse is under enormous pressure from the shareholders and board to reduce the production cost. Additionally, they are also interested in understanding the impact of their marketing and farmer connect efforts towards overall sales. In the same effort, they have hired you as a data science and predictive analytics consultant. Can you help them in optimizing and solving their business Problem

scraping-jumia-ecommerce

Using the Scrapy framework to scrape data consisting of name, brand, rating, price, product URL and image URLs of laptops on Jumia e-commerce (https://www.jumia.com.ng) into XLSX, SQL and MongoDB.

segmentation_models

Segmentation models with pretrained backbones. Keras and TensorFlow Keras.

shoesstoreproductscraping

sms_spam-detection-using-nlp

Simple Example of NLP(Natural Language Processing