shivakrishna2497
Strong engineering professional with a Master of Technology (M.Tech.) in Software Engineering from VIT University, Vellore, focused in Machine learning
Finnovation Tech Pvt LtdBangalore
Pinned Repositories
Convolutional-Neural-Network-to-classify-Handwritten-Digits
Convolutional Neural Network to classify hand written digits with Tensorflow’s Keras API and achieved an accuracy over 98%
DataCamp
Fraud-Reports
Fuzzy-String-Matching
HeartDiseasePrediction
The project involved analysis of the heart disease patient dataset with proper data processing. Then, 6 models were trained and tested with maximum scores as follows: 1)K Neighbors Classifier: 87% 2)Support Vector Classifier: 83% 3)Decision Tree Classifier: 79% 4)Random Forest Classifier: 84% 5)Logistic Regression:84% 6)Naive Bayes:80% In the end, K Neighbors Classifier achieved the highest score of 87% with 8 nearest neighbors.
Intelligent-document-processing-system-
Multi Class Document Classification
Predicting-Hospital-Readmissions-From-Diabetes-Dataset
In this project I will demonstrate how to build a model predicting readmission in Python using the following steps data exploration feature engineering building training/validation/test samples model selection model evaluation
Predicting-the-on-site-of-Diabetes-using-Keras-using-
I collected the Pima Indians onset of diabetes dataset from UCI Machine Learning repository,It describes patient medical record data for Pima Indians and whether they had an onset of diabetes within five years, As such, it is a binary classification problem (onset of diabetes as 1 or not as 0). I built my first neural network using keras which takes numerical input and numerical output Number of Instances: 768 Number of Attributes: 8 plus class For Each Attribute: (all numeric-valued) 1. Number of times pregnant 2. Plasma glucose concentration a 2 hours in an oral glucose tolerance test 3. Diastolic blood pressure (mm Hg) 4. Triceps skin fold thickness (mm) 5. 2-Hour serum insulin (mu U/ml) 6. Body mass index (weight in kg/(height in m)^2) 7. Diabetes pedigree function 8. Age (years) 9. Class variable (0 or 1) Steps I followed in building a neural network using Keras: 1)Load Data I have loaded the file directly using the NumPy function loadtxt(). There are eight input variables and one output variable ,Once loaded I split the dataset into input variables (X) and the output class variable (Y) 2)Define Network I created a Sequential() model and added layers one at a time , first layer has 12 neurons and expects 8 input variables, second hidden layer has 8 neurons and finally, the output layer has 1 neuron to predict the class (onset of diabetes or not) 3) Compile Network I Compiled the model using tensorflow as back-end,I used “binary_crossentropy“(loss function) and default gradient descent algorithm “adam” 4)Fit Network I fit the network with my training set(80%) by calling the fit() function on the model,For this problem, I gave a small number of epochs=150 and a small batch size of 10 5)Evaluate Network I evaluated the performance of the network on the same training dataset and got a training accuracy of 76.55% and after making predictions in next step I evaluated the performance of the network using testdataset where I got a testing accuracy of 76.62% 6)Make Predictions predictions will be in the range between 0 and 1 as there's a sigmoid activation function on the output layer and I converted them into a binary prediction for this classification task by rounding them
Prediction-of-House-Price-using-Decision-Tree
n this project, I developed and evaluated the performance and the predictive power of a model trained and tested on data collected from houses in Melbourne. Once I got a good fit, I used this model to predict the monetary value of a house located at the Melbourne’s area. A model like this would be very valuable for a real state agent who could make use of the information provided in a daily basis.
Sentiment-Analysis-of-IMDB-Movie-Reviews
The IMDB movie reviews dataset is a set of 50,000 reviews, half of which are positive and the other half negative. The dataset is available online and can be either directly downloaded from Stanford’s website or obtained by running in a terminal (Linux): wget http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz Then, we need to extract the dowloaded files. You can once again either do it manually or by running: tar -zxvf aclImdb_v1.tar.gz
shivakrishna2497's Repositories
shivakrishna2497/Sentiment-Analysis-of-IMDB-Movie-Reviews
The IMDB movie reviews dataset is a set of 50,000 reviews, half of which are positive and the other half negative. The dataset is available online and can be either directly downloaded from Stanford’s website or obtained by running in a terminal (Linux): wget http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz Then, we need to extract the dowloaded files. You can once again either do it manually or by running: tar -zxvf aclImdb_v1.tar.gz
shivakrishna2497/Predicting-the-on-site-of-Diabetes-using-Keras-using-
I collected the Pima Indians onset of diabetes dataset from UCI Machine Learning repository,It describes patient medical record data for Pima Indians and whether they had an onset of diabetes within five years, As such, it is a binary classification problem (onset of diabetes as 1 or not as 0). I built my first neural network using keras which takes numerical input and numerical output Number of Instances: 768 Number of Attributes: 8 plus class For Each Attribute: (all numeric-valued) 1. Number of times pregnant 2. Plasma glucose concentration a 2 hours in an oral glucose tolerance test 3. Diastolic blood pressure (mm Hg) 4. Triceps skin fold thickness (mm) 5. 2-Hour serum insulin (mu U/ml) 6. Body mass index (weight in kg/(height in m)^2) 7. Diabetes pedigree function 8. Age (years) 9. Class variable (0 or 1) Steps I followed in building a neural network using Keras: 1)Load Data I have loaded the file directly using the NumPy function loadtxt(). There are eight input variables and one output variable ,Once loaded I split the dataset into input variables (X) and the output class variable (Y) 2)Define Network I created a Sequential() model and added layers one at a time , first layer has 12 neurons and expects 8 input variables, second hidden layer has 8 neurons and finally, the output layer has 1 neuron to predict the class (onset of diabetes or not) 3) Compile Network I Compiled the model using tensorflow as back-end,I used “binary_crossentropy“(loss function) and default gradient descent algorithm “adam” 4)Fit Network I fit the network with my training set(80%) by calling the fit() function on the model,For this problem, I gave a small number of epochs=150 and a small batch size of 10 5)Evaluate Network I evaluated the performance of the network on the same training dataset and got a training accuracy of 76.55% and after making predictions in next step I evaluated the performance of the network using testdataset where I got a testing accuracy of 76.62% 6)Make Predictions predictions will be in the range between 0 and 1 as there's a sigmoid activation function on the output layer and I converted them into a binary prediction for this classification task by rounding them
shivakrishna2497/Convolutional-Neural-Network-to-classify-Handwritten-Digits
Convolutional Neural Network to classify hand written digits with Tensorflow’s Keras API and achieved an accuracy over 98%
shivakrishna2497/DataCamp
shivakrishna2497/Fraud-Reports
shivakrishna2497/Fuzzy-String-Matching
shivakrishna2497/HeartDiseasePrediction
The project involved analysis of the heart disease patient dataset with proper data processing. Then, 6 models were trained and tested with maximum scores as follows: 1)K Neighbors Classifier: 87% 2)Support Vector Classifier: 83% 3)Decision Tree Classifier: 79% 4)Random Forest Classifier: 84% 5)Logistic Regression:84% 6)Naive Bayes:80% In the end, K Neighbors Classifier achieved the highest score of 87% with 8 nearest neighbors.
shivakrishna2497/Intelligent-document-processing-system-
Multi Class Document Classification
shivakrishna2497/Predicting-Hospital-Readmissions-From-Diabetes-Dataset
In this project I will demonstrate how to build a model predicting readmission in Python using the following steps data exploration feature engineering building training/validation/test samples model selection model evaluation
shivakrishna2497/Prediction-of-House-Price-using-Decision-Tree
n this project, I developed and evaluated the performance and the predictive power of a model trained and tested on data collected from houses in Melbourne. Once I got a good fit, I used this model to predict the monetary value of a house located at the Melbourne’s area. A model like this would be very valuable for a real state agent who could make use of the information provided in a daily basis.
shivakrishna2497/Sequence-Models
Week 1
shivakrishna2497/WebScrapping-of-Techolution-Careers-website
Problem statement: I need to scrap the data from techolution website and store the data in a dataframe according to the date posted Step 1: Created virtualenvironment for the task needed to be done and installed required dependencies 1.beautifulsoup4==4.7.1 2.lxml==4.3.3 3.numpy==1.16.3 4.pandas==0.24.2 5.urllib3==1.24.2 Step 2 : Pulled the data from https://techolution.app.param.ai/jobs/ using BeautifulSoup Step 3 : I tried to pull the all the data regarding Job Titles and Date posted at once but I was getting empty list , So I pulled out each and every header tag which contained job title and stored it in “jobtitle.txt” file , Similarly stored all the data regarding date posted in “date.txt” Step 4 : Opened two text files and wrote down the entire details in third file “techolution.txt” Step 5 : Then I loaded the “techolution.txt” file into a dataframe according to the date posted with “Job Title” and “Date Posted” as Column names and stored the dataframe to a csv “Techolution.csv”