{
  "id": 157902,
  "title": "Late Submission - CNN Data Pipeline + Problem Solving Tutorial",
  "url": "/competitions/bengaliai-cv19/discussion/157902",
  "author_name": "James McGuigan",
  "post_date": "2020-06-12T13:12:26.922000",
  "votes": 3,
  "comment_count": 0,
  "views": 0,
  "content": "<p>As part of my learning journey, this was my first serious attempt to build a CNN neural network for a real-world dataset after completing the DeepLearning.ai and Tensorflow in Practice Specializations. I was previously a professional full stack data engineer.</p>\n\n<p>I didn't make the deadline (technical submission errors) and my top score was 0.9381 on public. However, my focus was on good engineering practices for a production-grade data pipeline. The result of this was a 2000 line GitHub codebase and a collection of Notebooks.</p>\n\n<p>I have done a problem solving tutorial/writeup explaining the numerous technical issues I encountered along the way and how I resolved them from a Data Engineering perspective.</p>\n\n<ul>\n<li><a href=\"https://www.kaggle.com/jamesmcguigan/bengali-ai-cnn-data-pipeline-problem-solving\">https://www.kaggle.com/jamesmcguigan/bengali-ai-cnn-data-pipeline-problem-solving</a></li>\n</ul>\n\n<p>Enjoy</p>",
  "messages": [
    {
      "id": 883224,
      "postDate": "2020-06-12T13:12:26.923Z",
      "content": "<p>As part of my learning journey, this was my first serious attempt to build a CNN neural network for a real-world dataset after completing the DeepLearning.ai and Tensorflow in Practice Specializations. I was previously a professional full stack data engineer.</p>\n\n<p>I didn't make the deadline (technical submission errors) and my top score was 0.9381 on public. However, my focus was on good engineering practices for a production-grade data pipeline. The result of this was a 2000 line GitHub codebase and a collection of Notebooks.</p>\n\n<p>I have done a problem solving tutorial/writeup explaining the numerous technical issues I encountered along the way and how I resolved them from a Data Engineering perspective.</p>\n\n<ul>\n<li><a href=\"https://www.kaggle.com/jamesmcguigan/bengali-ai-cnn-data-pipeline-problem-solving\">https://www.kaggle.com/jamesmcguigan/bengali-ai-cnn-data-pipeline-problem-solving</a></li>\n</ul>\n\n<p>Enjoy</p>",
      "rawMarkdown": "As part of my learning journey, this was my first serious attempt to build a CNN neural network for a real-world dataset after completing the DeepLearning.ai and Tensorflow in Practice Specializations. I was previously a professional full stack data engineer.\n\nI didn't make the deadline (technical submission errors) and my top score was 0.9381 on public. However, my focus was on good engineering practices for a production-grade data pipeline. The result of this was a 2000 line GitHub codebase and a collection of Notebooks.\n\nI have done a problem solving tutorial/writeup explaining the numerous technical issues I encountered along the way and how I resolved them from a Data Engineering perspective.\n\n- https://www.kaggle.com/jamesmcguigan/bengali-ai-cnn-data-pipeline-problem-solving\n\nEnjoy",
      "votes": 3
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "883224": "As part of my learning journey, this was my first serious attempt to build a CNN neural network for a real-world dataset after completing the DeepLearning.ai and Tensorflow in Practice Specializations. I was previously a professional full stack data engineer.\n\nI didn't make the deadline (technical submission errors) and my top score was 0.9381 on public. However, my focus was on good engineering practices for a production-grade data pipeline. The result of this was a 2000 line GitHub codebase and a collection of Notebooks.\n\nI have done a problem solving tutorial/writeup explaining the numerous technical issues I encountered along the way and how I resolved them from a Data Engineering perspective.\n\n- https://www.kaggle.com/jamesmcguigan/bengali-ai-cnn-data-pipeline-problem-solving\n\nEnjoy"
  }
}