qiskit-hackathon-korea-22
qiskit-hackathon-korea-22 copied to clipboard
qiskit-community
Hangul 한글 characters classification by Quantum Machine Learning
Abstract
Construct Hangul MNIST classifier using Qiskit Quantum Machine Learning.
Description
We have MNIST classifier made by 1) Quantum Support Vector Machine - Qiskit Global Summer School 2021, 2)Torch Connector and Hybrid QNNs. Let’s try to make Hangul characters classifier using Qiskit and adjust the method to improve the accuracy.
We can use Hangul Datasets which Sophy found for us or Handwritten Hangul Characters.
Members
- @FmnFlower
- @kjchoiquantum
- @Jason-L1
- @jin-castle
- @jhlee29
- @githubhandle
- @githubhandle - Slack:
@slackhandleemail:[email protected] - Qiskit Coach: @kifumi @attp @String137 @tula3and
Deliverable
GitHub repo
https://github.com/jhlee29/quantum-meets-hangul
I'm also interested in mentoring this project! Can I join as a mentor?
@kifumi Hojun is one of our mentor, this would be a good chance for Hojun learn more about QML from awesome senior mentors
Hi. This is Dongyoung, who I told you earlier on gather that I will be contacting you. I am reading (https://qiskit.org/textbook/ch-machine-learning/machine-learning-qiskit-pytorch.html) now. However, it seems that it will take some time to understand the code of pytorch because I have only used tensorflow before. Since you said you use SVM, we are going to use CNN. But honestly, I don't know if that's possible. Because like I said, I've only used TensorFlow.
As I said our team is already 4 people, so it would be difficult for all of our teams to participate in your team. So, as you said, I propose to split the team. I thought about dividing the team and trying to solve the same topic in different ways.
But in order to do that, our team first needs to understand the open code shown in the link. So, let's see if our team can understand that code, and if we can process the code at our level, we'll let you know. I'll tell you by tomorrow afternoon at the latest. thank you. If I'm wrong, or if you don't want to collaborate with us, don't hesitate to tell me.
Hello! I would like to join this project. Is there still a room in this group?
@0sophy1 How about dividing the team into two if the members exceed the limit?
If you are OK with that I'm definitely follow your opinion :) Thank you so much!
@tula3and Maybe this team will split to two teams, so the mentors should also split to the two teams. So, please join us!
As I said, our team is likely to proceed with the project using the Hangul dataset and QCNN. However, the dataset may change.
@ktasha45 Yes, I think the dataset is key for this project.. Could you please open new issue? And, please tell your members to leave the comment on the new issue.
@ktasha45 Yes, I think the dataset is key for this project.. Could you please open new issue? And, please tell your members to leave the comment on the new issue.
I'll do that. as soon as possible.
@FmnFlower @kjchoiquantum @Jason-L1 @jin-castle Hi everyone, could you please tell me your slack name? Or, please join the Room on veertly! Veertly에서 왼쪽 메뉴에 Rooms를 클릭하면 이슈에 맞는 방이 만들어져 있어요! 찾기 힘드시면 알려주세요~!