Anshit V. Profile

5.0

(1 review)

You're now following

Error following user.

This user does not allow users to follow them.

You are already following this user.

Your membership plan only allows 0 follows. Upgrade here.

Successfully unfollowed

Error unfollowing user.

You have successfully recommended

Error recommending user.

Something went wrong. Please refresh the page and try again.

Email successfully verified.

$10 USD / hour

katni,

india

$10 USD / hour

It's currently 8:12 PM here

Joined April 3, 2022

0 Recommendations

Anshit V.

@anshitfreelancin

5.0

(1 review)

2.0

100%

$10 USD / hour

katni,

india

$10 USD / hour

100%

Jobs Completed

100%

On Budget

100%

On Time

N/A

Repeat Hire Rate

Data Science Consultant

I am a self-taught data scientist who has practiced in this field for more than four years. I completed my undergraduate from IIT Roorkee in 2021, and I worked with LTI Infotech as Senior Software Engineer. Throughout the years, I have familiarized myself with the ins and outs of the world of Data Science, starting from handling missing values in the data, performing necessary data preprocessing steps, applying the best possible machine learning algorithm, to giving the detailed data visualizations to the end-user. I am fluent in Python and C++ programming languages and capable of writing any Machine Learning and Deep Learning algorithm using Python. Check out my work in the Portfolio items:

See all Python Developers in India

Contact Anshit V. about your job

Portfolio

9954871

-Worked on real-time detection and classification of vehicles on a live feed video captured at a toll plaza
-Implemented YOLOv4 on the video frames using Darknet, an open-source neural network framework
-Also worked on extracting text from number plates of vehicles using Python-tesseract, an optical character recognition tool of python

Number plate text detection using YOLO and python tesseract

9936941

A retail firm names 'Poseidon' has employed a chatbot on its progressive web app to interact with its customers. My job was to determine the core intent of the customer's interaction (query) and classify it into one of the following ten different tags :
0 - order.cancellation
1 - order.concern.delay
2 - order.modification
3 - order.status
4 - product.browse
5 - product.reviews
6 - shipping.address.modification
7 - shipping.plans.browse
8 - store.browse
9 - store.timing
Queries are customer interactions with a chatbot; the tags classify them broadly. To train a learning algorithm on this text data, I used Long short-term memory, an artificial recurrent neural network architecture used in deep learning. Unlike standard feedforward neural networks, LSTM has feedback connections. It can process single data points (such as images) and entire sequences of data (such as speech or video). My model was able to achieve a pretty decent accuracy of 83.98% accuracy on the test set.

Sentimental Analysis

9936938

Introduction: This documentation gives the precise method and algorithm applied in Melodious Medley. We have to suggest a new song to a person based on his/her preferences and music taste by making a predictive model.
Feature extraction from Datasets: The Dataset consists of 15 features in which I have used 14 features(excluding ID as it does not have a major impact on prediction). The Data does not have any missing data, making the work a little easier. Missing data in the pandas dataframe can be checked using the dataset.IsNull().sum().sum(), which gives you the total number of missing data in your raw data. The most difficult and important feature to deal with is ts_listen. The Time-Series Data can be used in the pandas dataframe using dataset['ts_listen']=pd.to_datetime(dataset.ts_listen)
This line converts the raw data into a systematic date-time format that can be further used to extract different values. E.g., Date, Hour, seconds, milliseconds, day of the week, Year, etc.

Suggest a new song to a person based on his/her preferences

9936853

-A Lock that unlocks itself only if the user’s face is
recognized orthe userinputs the PIN correctly
-Executed FaceNet Algorithm, which uses
Convolutional Neural Networks to recognize faces
from a real-time video
-Used a 12V Push-Pull Solenoid for making the
locking system, which was connected to the
Raspberry-Pi through a MOSFET transistor so that
12V power can be supplied externally
-Technologies used: Python, Tensorflow, Deep Learning,
Raspberry-Pi, OpenCV

Bio-metric Face Recognition Lock

9936860

ChatBot

9936856

-Deployed an end-to-end model that can predict
whether a bank should grant the loan to a particular
person based on credit rating defining features like
Real estate, Income, age, previous transactions, etc.
-Tested algorithms like XGBoost, LightGBM, Logistics
Regression, Random Forest and SVM
-Technologies used: Python, Pandas, Sckit-Learn, Machine
Learning, MS-Excel

Prediction of Loan approval using Machine Learning

9954871

Number plate text detection using YOLO and python tesseract

9936941

Sentimental Analysis

9936938

Suggest a new song to a person based on his/her preferences

9936853

Bio-metric Face Recognition Lock

9936860

ChatBot

9936856

Prediction of Loan approval using Machine Learning

Reviews

Changes saved

Showing 1 - 1 out of 1 reviews

Filter reviews by:

5.0

$100.00 USD

Project 33599339 has been deleted

He is good. I would like to work woth him again

Neha G.

@gtmneha

•

2 years ago

Experience

LTI Infotech

Jul 2021 - Dec 2021 (5 months, 1 day)

• Using RestAPI calls to retrieve data from the cloud server • Worked on technologies like Threading, Multiprocessing and Asyncio to speed up the whole process by almost four times

Enercomp Solutions Pvt. Ltd.

Jul 2020 - Sep 2020 (2 months, 1 day)

• Worked on real-time detection and classification of vehicles on a live feed video captured at a toll plaza • Implemented YOLOv4 on the video frames using Darknet, an open-source neural network framework • Also worked on extracting text from number plates of vehicles using Python-tesseract, an optical character recognition tool of python.

SensoVision Systems Pvt. Ltd.

Dec 2019 - Jan 2020 (1 month, 1 day)

• Worked on developing advanced image processing and computer vision C++ libraries to be deployed in automated industrial inspection systems • Worked extensively on real-time steel surface inspection systems, which apart from anomaly detection also involves measuring the parameters in real-time and filtering out the ones outside the bounds of user-specified tolerance