GSoC 2023 Project

[Yash Agarwal]

Hey mentors,

My name is Yash Agarwal, and I am a 3rd-year student pursuing Electronics and Communication Engineering at BITS Pilani, India. I am connecting with you regarding the project 'KerasCV: Object Detection API completion or Multi Modal model workflows.' I've read about the project on the project ideas page of GSoC 2023, and it interests me very much.

My current CGPA is 8.98, and I prominently work with Python, Java, C, and C++. I have completed several courses on Coursera on Machine Learning, Deep Learning, and Neural Net. PFA my LinkedIn profile link for the validated certifications.

I have also done a summer internship at the multinational technology company Jio Platforms, the holding company for India's largest mobile network provider, Jio. The project was on 'Adversarial Attack - Implement Face Adversarial attack by adding any attribute to face that causes misclassification of the face.' The project extensively used Keras and TensorFlow framework. PFA the project readme file. I am also currently pursuing a project under a professor at my college, which pertains to 'Comparative Analysis of various Deep Learning Models for Object Detection.' I am comparing state-of-the-art models like Yolov8, SSD, and Faster R-CNN in this project.

I am inclined to build a career in the Machine Learning domain, and working under your esteemed guidance on this project will be a swift learning curve for me. I want to start working with it and I was hoping if you could please give me some pointers on where I can start or read about it to research and learn more about this project.

LinkedIn profile link: https://www.linkedin.com/in/id6109/

Looking forward to a positive reply from you :)

[Sayak Paul]

Kudos to you for your cool projects. Just curious. What is the relevance of CGPA here? 

Sayak Paul | sayak.dev

--
You received this message because you are subscribed to the Google Groups "Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to discuss+u...@tensorflow.org.
To view this discussion on the web visit https://groups.google.com/a/tensorflow.org/d/msgid/discuss/4ba52682-3936-4cd0-9c18-84f9c47a6799n%40tensorflow.org.

[Ayush Agrawal]

Maybe it's better to guide a beginner according to their project interests instead of pointing out nits? There's already too much spam on the group :)

[Sayak Paul]

Yes, Ayush, you're right! I apologize if my tone sounded condescending. 

Yash, glad to know your interests, and it looks like you're quite already on the right track. 

As a 2-times TensorFlow mentor for GSoC, I have some little things to share here. 

First, this guide might be helpful for you to get a general understanding of what I (as a mentor) would look for in an application: https://sayak.dev/gsoc-faqs/. The website is facing some rendering issues (which I hope to fix really soon). 

Second, I would definitely recommend you to check out the existing issues and PRs to deeply understand the developer philosophy and how you might get started to start building for KerasCV. This is a great way for you to start contributing and find a few areas in the library according to your interests. 

Finally, there are some existing materials on KerasCV that you can find below which would help you deepen your understanding of the API usage:

* https://lukewood.xyz/

* https://keras.io/keras_cv/

Hope these help :) 

Sayak Paul | sayak.dev

The information contained in this electronic communication is intended solely for the individual(s) or entity to which it is addressed. It may contain proprietary, confidential and/or legally privileged information. Any review, retransmission, dissemination, printing, copying or other use of, or taking any action in reliance on the contents of this information by person(s) or entities other than the intended recipient is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us by responding to this email or telephone and immediately and permanently delete all copies of this message and any attachments from your system(s). The contents of this message do not necessarily represent the views or policies of BITS Pilani.

[Yash Agarwal]

Thank you.

[Ayush Agrawal]

It's alright, Sayak. Thanks for the pointers. I'm sure your guidance will be really helpful to Yash.

I've been a Tensorflow GSoC Student Dev and Mentor myself, so I know how helpful and enthusiastic the Tensorflow community is. We need to assist each other as much as possible. Just trying to do the same :)

 

- Ayush

[Harshith Mohan Kumar]

Hey,

I've been receiving a large quantity of emails like this where students would briefly introduce themselves and state a list of personal accomplishments and conclude by saying that they are interested in the project listed.

I highly suggest anyone who is truly interested and who wants to maximize their likelihood of being selected to take great initiative in digging all of the publicly available information there is on these project proposals prior to emailing mentors. This act of putting forth the effort first showcases your grit and growth mindset. Infact, this is the first response I provide to any student reaching out to me with a basic introduction in hopes to work on the project I've listed for Red Hen Lab.

Recently I held an online meeting with a few students who were curious to know more about the GSoC proposal building phase. I've recorded this meeting and have uploaded it. Link: https://youtu.be/KMDu0sp2HRE. Maybe this could help you @yash and other first time applicants.

P.S I'm not a mentor for TF, just trying to share some tips :)

Regards,

Harshith MK

To view this discussion on the web visit https://groups.google.com/a/tensorflow.org/d/msgid/discuss/f25d51c5-64aa-4ce6-ba00-ac631a6b3e4cn%40tensorflow.org.

[Yash Agarwal]

Hey Harshith,

I appreciate the suggestion, but I did thoroughly analyze the project before writing this post. To me, discussing the project's specific details would come only after the mentor replies to the preliminary introduction.

Thanks for your insight.