Pinned Repositories
Computer_vision_techniques
This repository consist of Image processing techniques done using OpenCV with different set of examples which is useful for all the real time applications
Customer-Churn-Prediction
This case requires trainees to develop a model for predicting customer churn at a fictitious wireless telecom company and use insights from the model to develop an incentive plan for enticing would-be churners to remain with company. Data for the case are available in csv format. The data are a scaled down version of the full database generously donated by an anonymous wireless telephone company.
Data-structures-
Practicing Algorithms and Data structures in python that can organize and stores data efficiently
Decision-Tree-
Decision Tree is the model that contains Tree like structure (if else ...) which has Nodes in it. The splitting of dataset into true Node and False Node is done by checking Impurity in the data and maximum Information gain's attribute is split as next internal node(classification) , orelse by Standard deviation (Regression).
Diabetics_or_not
Prima Indians Dataset contains several people's body condition to check whether that person will be affected due to diabetes or not.
Face-Mask-Detection
This repository consist of an approach to reduce risk of Coronavirus spread. Wearing a mask is among the non-pharmaceutical intervention measures that can be used to cut the primary source of SARS-CoV2 droplets expelled by an infected individual. The outstanding performance of the proposed model is highly suitable for video surveillance devices.. Though should improve more on accuracy of the model in the future.
Handwritten-digits-classifier
Classifying Handwritten digits,by picking datasets from MNIST .This is done by Supervised Algorithms (classification),by giving almost 98% accuracy.
Pneumothorax-Segmentation-from-Chest-X-ray-
Pneumothorax is a relatively common respiratory disease that can occur when air leaks into the space between the lungs and chest wall. A blunt or penetrating chest injury, certain medical procedures or lung disease can cause a pneumothorax. Doctors will use chest X-rays to identify Pneumothorax. Based on that, doctor can suggest suitable treatment. Using a dataset with x-ray images and their diagnosis with the exact place of the air in the pleura, a model could be trained to recognize Pneumothorax.
Real-time-Fruit-Recognition-
This repository consist of a real time fruit recognition system in grocery stores where the model classifies the fruits from the tray which is starting point of cashier-less stores with customers able to purchase fruits without being checked out by a cashier or using a self-checkout station. The model provides 96% accuracy.
Time-Series-Modeling
The dataset contains each household's electric power consumption in the entire area which consist of 2075259 records (houses) located in Sceaux (7km of Paris, France) between December 2006 and November 2010 (47 months). The task is to predict the electricity power consumption for unknown houses in that area with respect to the measurement we had. Though ,this is Supervised Learning problem as the dataset consist of target value also and this is the regression problem as the prediction is of power consumption ie. continuous value.
Harisuryatm's Repositories
Harisuryatm/Real-time-Fruit-Recognition-
This repository consist of a real time fruit recognition system in grocery stores where the model classifies the fruits from the tray which is starting point of cashier-less stores with customers able to purchase fruits without being checked out by a cashier or using a self-checkout station. The model provides 96% accuracy.
Harisuryatm/Time-Series-Modeling
The dataset contains each household's electric power consumption in the entire area which consist of 2075259 records (houses) located in Sceaux (7km of Paris, France) between December 2006 and November 2010 (47 months). The task is to predict the electricity power consumption for unknown houses in that area with respect to the measurement we had. Though ,this is Supervised Learning problem as the dataset consist of target value also and this is the regression problem as the prediction is of power consumption ie. continuous value.
Harisuryatm/Computer_vision_techniques
This repository consist of Image processing techniques done using OpenCV with different set of examples which is useful for all the real time applications
Harisuryatm/Customer-Churn-Prediction
This case requires trainees to develop a model for predicting customer churn at a fictitious wireless telecom company and use insights from the model to develop an incentive plan for enticing would-be churners to remain with company. Data for the case are available in csv format. The data are a scaled down version of the full database generously donated by an anonymous wireless telephone company.
Harisuryatm/Data-structures-
Practicing Algorithms and Data structures in python that can organize and stores data efficiently
Harisuryatm/Decision-Tree-
Decision Tree is the model that contains Tree like structure (if else ...) which has Nodes in it. The splitting of dataset into true Node and False Node is done by checking Impurity in the data and maximum Information gain's attribute is split as next internal node(classification) , orelse by Standard deviation (Regression).
Harisuryatm/Diabetics_or_not
Prima Indians Dataset contains several people's body condition to check whether that person will be affected due to diabetes or not.
Harisuryatm/Face-Mask-Detection
This repository consist of an approach to reduce risk of Coronavirus spread. Wearing a mask is among the non-pharmaceutical intervention measures that can be used to cut the primary source of SARS-CoV2 droplets expelled by an infected individual. The outstanding performance of the proposed model is highly suitable for video surveillance devices.. Though should improve more on accuracy of the model in the future.
Harisuryatm/Handwritten-digits-classifier
Classifying Handwritten digits,by picking datasets from MNIST .This is done by Supervised Algorithms (classification),by giving almost 98% accuracy.
Harisuryatm/Pneumothorax-Segmentation-from-Chest-X-ray-
Pneumothorax is a relatively common respiratory disease that can occur when air leaks into the space between the lungs and chest wall. A blunt or penetrating chest injury, certain medical procedures or lung disease can cause a pneumothorax. Doctors will use chest X-rays to identify Pneumothorax. Based on that, doctor can suggest suitable treatment. Using a dataset with x-ray images and their diagnosis with the exact place of the air in the pleura, a model could be trained to recognize Pneumothorax.
Harisuryatm/K---Nearest-Neighbors
K - Nearest Neighbors is used in both Regression and Classification problems , mostly this algorithm is good for classifying objects. This Algorithm is done by taking minimum distance of k-nearest points while predicting the new data and selects outcome as maximum of the class within that k-range.Gives better accuracy and Learns complex models easily.
Harisuryatm/Linear-Regression-
In simple terms, Linear Regression is the model to train and find the best fit line equation for the given dataset. This is done by minimizing the cost(Error) function between actual datapoints with respect to predicted line equation. Our goal is to update the weights and biases using gradient descent technique.
Harisuryatm/Loan_Dataset_Cleansing
Dataset contains consists of some customer information and their associated loan information. The loan was taken to buy a product and hence the product ID and their details are also provided by the company that are unstructured . By analyzing this data or exploratory data analysis on it and infer from the data about the trends in loans taken and products bought.
Harisuryatm/Multi-Class-Aerial-Segmentation
Semantic segmentation on aerial imagery using Torch implementations of UNet and UNet++.
Harisuryatm/Naive-Bayes-
Naive-Bayes Algorithm is a classification problem,by using Bayes Theorem (calculates posterior probability) it works on large dataset . The features are assumed to be independent and made as equal importance .
Harisuryatm/Neural_Networks_insights
contains basic ANN structure with implementing the gates (basics of NN) to get the insights of Forward and Backward propogation
Harisuryatm/Random-Forest-Algorithm
Random Forest Algorithm is an ensembling method for classification and regression that is operated by construction of multiple trees that makes model robust. that works on large dataset with high precision and accuracy.
Harisuryatm/Support-Vector-Machine
Support Vector Machine model is used inboth classification and regression.Mostly used in Classification task by finding hyperplane that differentiate two classes very well.Same as Linear Regression here we find both cost function and gradient descent (by updating weights and biases). Reguralization parameter (C) - to balance the hyperplane margin