Freelancer logo How It Works Browse Jobs Log In Sign Up Post a Project Profile cover photoundefined
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 user.
Error unfollowing user.
You have successfully recommended
Error recommending user.
Email successfully verified.
User Avatar
$30 USD / hour
Flag of INDIA
mumbai, india
$30 USD / hour
It's currently 3:26 PM here
Joined August 13, 2017
4 Recommendations

Ujjwal K.

@ujjwal1996

5.0 (52 reviews)
5.5
5.5
$30 USD / hour
Flag of INDIA
mumbai, india
$30 USD / hour
97%
Jobs Completed
96%
On Budget
97%
On Time
26%
Repeat Hire Rate

Machine Learning | Python | Data Science

Thank you for visiting my profile. ★★★★ My service providing skill sets include but are not limited to ★★★★ ✅ ENGINEERING AND TECHNOLOGY ✔️ Python, MATLAB, MS Excel ✔️ AI, ML, and Algorithms (Data Structures, Neural Networks, Data Science, NLP) ✔️ AI and ML Framework (TensorFlow, Keras, RNN, CNN, GAN, DNN, ...) ✔️ Probability and Statistics (Hypothesis testing, Forecasting, T-test, ANOVA, ...) ✔️ Data Analysis (Python Pandas, Jupyter Notebook, Anaconda) ✔️ Technical Writing ✔️ Web Scraping (BeautifulSoup, Selenium, requests, urllib, ...) ✔️ OpenCV (Object Segmentation, Object Detection, Image Processing, ...) ✔️ Optimizations (Genetic Algorithms, Swarm Optimization, Ant Colony Optimization, ...) ✔️ Mechanical and Electrical Engineering (Signal Processing, Control Systems, Robotics, CFD, Mechanics of Solids, ...) ------------------------------------------------------------------------------------------------------------------------------ ➤➤ I can spend full time on Freelancer and serve 24x7. ➤➤ Rest assured with 100% accuracy and efficiency on my part. ➤➤ I will provide you the project before the deadline and I am comfortable with negotiation in pricing. Let us build a great relationship for the successful implementation of your project. Best Regards
Freelancer Python Developers India

Contact Ujjwal K. about your job

Log in to discuss any details over chat.

Portfolio Items

To forecast the travel times of the assigned route arc on every working
day for the whole current 2020 year (excluding January 1st, December
25th and every Sunday) based on the previous data.

Methodology:
The entire project is divided into 3 steps:
1. Pre-processing the data
2. Performing statistical analysis on the data
3. Forecasting the travel times using LSTM neural net architecture
Time Series Forecasting
To forecast the travel times of the assigned route arc on every working
day for the whole current 2020 year (excluding January 1st, December
25th and every Sunday) based on the previous data.

Methodology:
The entire project is divided into 3 steps:
1. Pre-processing the data
2. Performing statistical analysis on the data
3. Forecasting the travel times using LSTM neural net architecture
Time Series Forecasting
To forecast the travel times of the assigned route arc on every working
day for the whole current 2020 year (excluding January 1st, December
25th and every Sunday) based on the previous data.

Methodology:
The entire project is divided into 3 steps:
1. Pre-processing the data
2. Performing statistical analysis on the data
3. Forecasting the travel times using LSTM neural net architecture
Time Series Forecasting
To forecast the travel times of the assigned route arc on every working
day for the whole current 2020 year (excluding January 1st, December
25th and every Sunday) based on the previous data.

Methodology:
The entire project is divided into 3 steps:
1. Pre-processing the data
2. Performing statistical analysis on the data
3. Forecasting the travel times using LSTM neural net architecture
Time Series Forecasting
To forecast the travel times of the assigned route arc on every working
day for the whole current 2020 year (excluding January 1st, December
25th and every Sunday) based on the previous data.

Methodology:
The entire project is divided into 3 steps:
1. Pre-processing the data
2. Performing statistical analysis on the data
3. Forecasting the travel times using LSTM neural net architecture
Time Series Forecasting
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
In order to be used for GMM, images were converted to a 5–dimensional hypercube, where each pixel was converted to a feature
vector of size 5

GaussianMixture() function from scipy library was used to implement GMM. The hypercube data was t to a GMM model and with the
objective function of BIC and the best number of clusters or classes were then chosen based on lowest value of BIC
GMM Clustering
The paper demonstrates one of the major application of CV which is Facial
expression classification through facial emotions using Convolution Neural Network
(CNN). 

We have also utilized Graphics Processing Unit (GPU) computation (available
by notebook in the Google colab) so that we can accelerate the training process. 

Now we have built this CNN model with the help of Python using a well known Tensorflow module
called Keras. 

Different types of expressions that can be classified by our model is based on
fer2013 dataset.
Convolution Neural Network: Facial Expression classification
The paper demonstrates one of the major application of CV which is Facial
expression classification through facial emotions using Convolution Neural Network
(CNN). 

We have also utilized Graphics Processing Unit (GPU) computation (available
by notebook in the Google colab) so that we can accelerate the training process. 

Now we have built this CNN model with the help of Python using a well known Tensorflow module
called Keras. 

Different types of expressions that can be classified by our model is based on
fer2013 dataset.
Convolution Neural Network: Facial Expression classification
The paper demonstrates one of the major application of CV which is Facial
expression classification through facial emotions using Convolution Neural Network
(CNN). 

We have also utilized Graphics Processing Unit (GPU) computation (available
by notebook in the Google colab) so that we can accelerate the training process. 

Now we have built this CNN model with the help of Python using a well known Tensorflow module
called Keras. 

Different types of expressions that can be classified by our model is based on
fer2013 dataset.
Convolution Neural Network: Facial Expression classification
The paper demonstrates one of the major application of CV which is Facial
expression classification through facial emotions using Convolution Neural Network
(CNN). 

We have also utilized Graphics Processing Unit (GPU) computation (available
by notebook in the Google colab) so that we can accelerate the training process. 

Now we have built this CNN model with the help of Python using a well known Tensorflow module
called Keras. 

Different types of expressions that can be classified by our model is based on
fer2013 dataset.
Convolution Neural Network: Facial Expression classification
The paper demonstrates one of the major application of CV which is Facial
expression classification through facial emotions using Convolution Neural Network
(CNN). 

We have also utilized Graphics Processing Unit (GPU) computation (available
by notebook in the Google colab) so that we can accelerate the training process. 

Now we have built this CNN model with the help of Python using a well known Tensorflow module
called Keras. 

Different types of expressions that can be classified by our model is based on
fer2013 dataset.
Convolution Neural Network: Facial Expression classification
The paper demonstrates one of the major application of CV which is Facial
expression classification through facial emotions using Convolution Neural Network
(CNN). 

We have also utilized Graphics Processing Unit (GPU) computation (available
by notebook in the Google colab) so that we can accelerate the training process. 

Now we have built this CNN model with the help of Python using a well known Tensorflow module
called Keras. 

Different types of expressions that can be classified by our model is based on
fer2013 dataset.
Convolution Neural Network: Facial Expression classification
• To derive the stiffness matrix for a bar element.
• To illustrate how to solve a bar assemblage by the direct
stiffness method.
• To introduce guidelines for selecting displacement
functions.
• To describe the concept of transformation of vectors in
two different coordinate systems in the plane.
• To derive the stiffness matrix for a bar arbitrarily oriented
in the plane.
• To demonstrate how to compute stress for a bar in the
plane.
• To show how to solve a plane truss problem.
• To develop the transformation matrix in threedimensional space and show how to use it to derive the
stiffness matrix for a bar arbitrarily oriented in space.
• To demonstrate the solution of space trusses.
FEM Analysis using Python
• To derive the stiffness matrix for a bar element.
• To illustrate how to solve a bar assemblage by the direct
stiffness method.
• To introduce guidelines for selecting displacement
functions.
• To describe the concept of transformation of vectors in
two different coordinate systems in the plane.
• To derive the stiffness matrix for a bar arbitrarily oriented
in the plane.
• To demonstrate how to compute stress for a bar in the
plane.
• To show how to solve a plane truss problem.
• To develop the transformation matrix in threedimensional space and show how to use it to derive the
stiffness matrix for a bar arbitrarily oriented in space.
• To demonstrate the solution of space trusses.
FEM Analysis using Python
• To derive the stiffness matrix for a bar element.
• To illustrate how to solve a bar assemblage by the direct
stiffness method.
• To introduce guidelines for selecting displacement
functions.
• To describe the concept of transformation of vectors in
two different coordinate systems in the plane.
• To derive the stiffness matrix for a bar arbitrarily oriented
in the plane.
• To demonstrate how to compute stress for a bar in the
plane.
• To show how to solve a plane truss problem.
• To develop the transformation matrix in threedimensional space and show how to use it to derive the
stiffness matrix for a bar arbitrarily oriented in space.
• To demonstrate the solution of space trusses.
FEM Analysis using Python
• To derive the stiffness matrix for a bar element.
• To illustrate how to solve a bar assemblage by the direct
stiffness method.
• To introduce guidelines for selecting displacement
functions.
• To describe the concept of transformation of vectors in
two different coordinate systems in the plane.
• To derive the stiffness matrix for a bar arbitrarily oriented
in the plane.
• To demonstrate how to compute stress for a bar in the
plane.
• To show how to solve a plane truss problem.
• To develop the transformation matrix in threedimensional space and show how to use it to derive the
stiffness matrix for a bar arbitrarily oriented in space.
• To demonstrate the solution of space trusses.
FEM Analysis using Python
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Use historical data to predict:

▪ future sales (Time series analysis)
▪ The probability of adding a new fund in the future
▪ Use different ML models to create lift and decile chart
Assist sales and marketing by improving their targeting
Found the underlying buying patterns of the customers of an automobile part manufacturer based on the past 3 years of the Company's transaction data

Pattern finding was based upon RFM analysis which helped to create customer segments

Provided identified segments to the company so that they can devise a strategy to deal with such segments
RFM Analysis
Found the underlying buying patterns of the customers of an automobile part manufacturer based on the past 3 years of the Company's transaction data

Pattern finding was based upon RFM analysis which helped to create customer segments

Provided identified segments to the company so that they can devise a strategy to deal with such segments
RFM Analysis
Found the underlying buying patterns of the customers of an automobile part manufacturer based on the past 3 years of the Company's transaction data

Pattern finding was based upon RFM analysis which helped to create customer segments

Provided identified segments to the company so that they can devise a strategy to deal with such segments
RFM Analysis
Found the underlying buying patterns of the customers of an automobile part manufacturer based on the past 3 years of the Company's transaction data

Pattern finding was based upon RFM analysis which helped to create customer segments

Provided identified segments to the company so that they can devise a strategy to deal with such segments
RFM Analysis

Reviews

Changes saved
Showing 1 - 5 out of 50+ reviews
Filter reviews by: 5.0
$100.00 USD
Good, will corporate more.
Python Machine Learning (ML) Mathematics Deep Learning
+1 more
K
Flag of Yuehong L. @katherine176
2 months ago
0.0
$30.00 AUD
The project is not completed according to the dispute
Python Data Processing Machine Learning (ML) Data Analysis Deep Learning
G
Flag of Ali S. @gis2020research
5 months ago
5.0
$200.00 USD
Nice working with Ujjwal again. Thank you!
Engineering Matlab and Mathematica Algorithm Electrical Engineering Machine Learning (ML)
User Avatar
Flag of Sinisa S. @sinisaskoric1969
7 months ago
5.0
$70.00 USD
Good, will corporate more.
Python Machine Learning (ML) Mathematics Deep Learning
+1 more
K
Flag of Yuehong L. @katherine176
7 months ago
5.0
£120.00 GBP
He was quick and his budget was at a good price.
Python Machine Learning (ML) Mathematics Deep Learning
+1 more
J
Flag of John M. @johnmoday99
7 months ago

Experience

Research associate

National University of Singapore
Jun 2019 - Aug 2019 (2 months, 2 days)
Quantified peristaltic motion of soft robot using object segmentation and multi task learning

Machine Learning intern

ISMRITI
Jun 2018 - Aug 2018 (2 months, 2 days)
Developed scalable Natural language semantic text search engine and image hashing search engine. Also mentored a class of 400 students in an AI workshop conducted by ISMRITI.

Education

Dual degree

Indian Institute of Technology, Kanpur, India 2019 - 2020
(1 year)

Bachelors of technology

Indian Institute of Technology, Kanpur, India 2015 - 2019
(4 years)

Qualifications

Best project award

Science and Technology Club, IIT Kanpur
2018
Awarded the best project award for student search app using face recognition by IIT Kanpur

Contact Ujjwal K. about your job

Log in to discuss any details over chat.

Verifications

Preferred Freelancer
Identity Verified
Payment Verified
Phone Verified
Email Verified
Facebook Connected

Certifications

preferredfreelancer-1.png Preferred Freelancer Program SLA 1 97%

Top Skills

Python 45 Machine Learning (ML) 36 Mathematics 34 Deep Learning 31 MATLAB 31

Browse Similar Freelancers

Python Developers in India
Python Developers
Machine Learning Experts
Mathematics Teachers

Browse Similar Showcases

Python
Machine Learning (ML)
Mathematics
Deep Learning
Previous User
Next User
Invite sent successfully!
Registered Users Total Jobs Posted
Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)
Copyright © 2022 Freelancer Technology Pty Limited (ACN 142 189 759)
There is no internet connection
Loading preview