You will get a hyper-parameter tuned AI model
Project details
If you have annotated some images for tasks like image classification, object detection, or segmentation and would like to train a model to test whether it works, this project is for you. I'll program a simple model with a reasonable number of hyperparameter ranges. You'll be able to see how well a model trained on your data performs initially, after which you can decide how you'd like to move forward with the problem you are addressing.
This project is an exploratory step in the journey to a production-ready product. If your project is more complex and/or you'd like additional experimentation, please contact me directly on my profile.
This project is an exploratory step in the journey to a production-ready product. If your project is more complex and/or you'd like additional experimentation, please contact me directly on my profile.
Machine Learning Tools
Azure Machine Learning, MLflow, OpenCV, Python, PyTorch, scikit-learn, TensorFlowWhat's included $1,200
These options are included with the project scope.
$1,200
- Delivery Time 7 days
- Number of Revisions 0
- Number of Model Variations 5
- Number of Graphs/Charts 0
- Model Validation/Testing
- Source Code
Frequently asked questions
23 reviews
(23)
(0)
(0)
(0)
(0)
This project doesn't have any reviews.
PS
Pradeep S.
Jun 2, 2025
Consultation to help fix an issue
I had a great conversation with Rishi to debug the issue with the tensorflow training job I was running into.
CG
Craig G.
May 25, 2024
Extract frames and help with object detection
SK
Selwin K.
Apr 7, 2024
Mask RCNN Expert
MS
Michael S.
Feb 22, 2024
Scripts for handling labels and automate the processes
RS
Rijad S.
Feb 22, 2024
Manual labelling of the Leaf dataset (20K - 22K images)
About Rishi Swethan
Senior AI Engineer specialising in Computer Vision and Multimodal LLMs
100%
Job Success
Barcelona, Spain - 2:23 pm local time
🧰 6+ years of comprehensive experience in AI development for business and research needs.
🚀 Deployed numerous computer vision products used by businesses and governments around the world
✈️ Adaptable work schedule, catering to an international clientele spanning from the US to Australia.
⌛ Strict adherence to time management and budget control, guaranteeing punctual project delivery and efficient resource utilisation.
I've honed my skills by effectively integrating conventional and modern approaches. Image processing and computer vision remain the most challenging and rewarding aspects of my work.
My approach to computer vision-based projects can be summarized in three steps:
* Exploring non-machine-learning approaches where applicable
* Employing hyperparameter tuning to identify the best algorithm and augmentation configurations if Machine Learning is necessary
* Utilizing image processing techniques from step (1) to enhance the results of ML-based algorithms
-> Frequently used image processing techniques:
* Histogram Equalisation
* Canny Edge Detection
* Hough Transform
* Sobel Operator
* Median Filter
* Otsu's Thresholding
* Connected Component Labelling
* Morphological Operations
... and more
-> Deep learning-based approaches:
* Image Segmentation
* Image Classification
* Vision Transformers
* Vision LLMs
* LLM fine-tuning
* GANs
* YOLO
* Pix2Pix
... and more
As a skilled Machine Learning developer with a strong background in R&D, my unique approach to tackling complex computer vision problems ensures optimal results for your projects. Let's connect to discuss how my expertise can help you achieve your goals.
Rishi Swethan Rajkumar
CEO and AI lead at Serna Solutions Limited
Steps for completing your project
After purchasing the project, send requirements so Rishi Swethan can start the project.
Delivery time starts when Rishi Swethan receives requirements from you.
Rishi Swethan works on your project following the steps below.
Revisions may occur after the delivery date.
Programming and training
I will write the program needed to train the model and train it locally