Greetings Kagglers! 👋 Recently, I delved into the Digit Recognizer competition and experimented with various Deep Learning Architectures.
\nBelow are the notebooks detailing my journey:
\nMultilayer Perceptron - Introduction to a basic MLP architecture with a commendable accuracy of 97.832%.
CNN Architecture - A deep dive into Convolutional Neural Networks, achieving an impressive score of 98.810%.
Vanilla ResNet18 Architecture - Implementation and insights into ResNet18, securing an accuracy of 98.885%.
ResNet18 with Augmentation - Leveraging data augmentation with ResNet18 for a remarkable accuracy of 99.260%.
🔍 Key Highlights:
\n🙏 Acknowledgments:
\nSpecial thanks to @rasbtn for creating the Deep Learning Fundamentals course. This work would not have been possible without learning about these architectures from this course.
\n💬 Your Feedback Matters:
\nFeel free to explore the notebooks and share your insights. Have questions or suggestions? Let's discuss this in the comments!
\nHappy coding, and may your models be ever accurate! 🚀✨
", "messages": [ { "id": 2607315, "postDate": "2024-01-18T07:00:27.770Z", "content": "Greetings Kagglers! 👋 Recently, I delved into the Digit Recognizer competition and experimented with various Deep Learning Architectures.
\nBelow are the notebooks detailing my journey:
\nMultilayer Perceptron - Introduction to a basic MLP architecture with a commendable accuracy of 97.832%.
CNN Architecture - A deep dive into Convolutional Neural Networks, achieving an impressive score of 98.810%.
Vanilla ResNet18 Architecture - Implementation and insights into ResNet18, securing an accuracy of 98.885%.
ResNet18 with Augmentation - Leveraging data augmentation with ResNet18 for a remarkable accuracy of 99.260%.
🔍 Key Highlights:
\n🙏 Acknowledgments:
\nSpecial thanks to @rasbtn for creating the Deep Learning Fundamentals course. This work would not have been possible without learning about these architectures from this course.
\n💬 Your Feedback Matters:
\nFeel free to explore the notebooks and share your insights. Have questions or suggestions? Let's discuss this in the comments!
\nHappy coding, and may your models be ever accurate! 🚀✨
", "rawMarkdown": "Greetings Kagglers! 👋 Recently, I delved into the Digit Recognizer competition and experimented with various Deep Learning Architectures. \n\n**Below are the notebooks detailing my journey:**\n\n- [**Multilayer Perceptron**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p1-mlp-arch-0-97832) - Introduction to a basic MLP architecture with a commendable **accuracy of 97.832%**.\n\n- [**CNN Architecture**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p2-cnn-arch-0-98810) - A deep dive into Convolutional Neural Networks, achieving an **impressive score of 98.810%**.\n\n- [**Vanilla ResNet18 Architecture**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p3-resnet18-arch-0-98885) - Implementation and insights into ResNet18, **securing an accuracy of 98.885%**.\n\n- [**ResNet18 with Augmentation**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p4-resnet18-with-aug-0-9926) - Leveraging data augmentation with ResNet18 for a **remarkable accuracy of 99.260%**.\n\n**🔍 Key Highlights:**\n\n- Scores are based on the public leaderboard.\n- Explore the methodology and nuances of each architecture.\n\n**🙏 Acknowledgments:**\n\nSpecial thanks to @rasbtn for creating the [Deep Learning Fundamentals](https://lightning.ai/courses/deep-learning-fundamentals/) course. This work would not have been possible without learning about these architectures from this course.\n\n**💬 Your Feedback Matters:**\n\nFeel free to explore the notebooks and share your insights. Have questions or suggestions? Let's discuss this in the comments!\n\nHappy coding, and may your models be ever accurate! 🚀✨", "votes": 2 } ], "comments": [], "raw_markdown_by_id": { "2607315": "Greetings Kagglers! 👋 Recently, I delved into the Digit Recognizer competition and experimented with various Deep Learning Architectures. \n\n**Below are the notebooks detailing my journey:**\n\n- [**Multilayer Perceptron**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p1-mlp-arch-0-97832) - Introduction to a basic MLP architecture with a commendable **accuracy of 97.832%**.\n\n- [**CNN Architecture**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p2-cnn-arch-0-98810) - A deep dive into Convolutional Neural Networks, achieving an **impressive score of 98.810%**.\n\n- [**Vanilla ResNet18 Architecture**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p3-resnet18-arch-0-98885) - Implementation and insights into ResNet18, **securing an accuracy of 98.885%**.\n\n- [**ResNet18 with Augmentation**](https://www.kaggle.com/code/pardeep19singh/pytorch-lightning-p4-resnet18-with-aug-0-9926) - Leveraging data augmentation with ResNet18 for a **remarkable accuracy of 99.260%**.\n\n**🔍 Key Highlights:**\n\n- Scores are based on the public leaderboard.\n- Explore the methodology and nuances of each architecture.\n\n**🙏 Acknowledgments:**\n\nSpecial thanks to @rasbtn for creating the [Deep Learning Fundamentals](https://lightning.ai/courses/deep-learning-fundamentals/) course. This work would not have been possible without learning about these architectures from this course.\n\n**💬 Your Feedback Matters:**\n\nFeel free to explore the notebooks and share your insights. Have questions or suggestions? Let's discuss this in the comments!\n\nHappy coding, and may your models be ever accurate! 🚀✨" } }