{
  "id": 198230,
  "title": "Image Statistics DATASET Compiled - For EDA",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/198230",
  "author_name": "Aditya Baurai",
  "post_date": "2020-11-20T10:33:44.732000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hello all!!</p>\n<p>I created this metadata .csv file (<strong>train_image_statistics.csv</strong>) encapsulating the compiled statistics of all our training images, provided by Kaggle for this competition. </p>\n<p>Attributes such as <em>shape, skewness, mean of each color channel, path, etc are all tabulated for each image</em>, stored, and exported as a .csv file. </p>\n<hr>\n<p><strong>Here is the notebook which I used for creating it. You can add or remove features accordingly</strong> : <br>\n<strong><a href=\"https://www.kaggle.com/fireheart7/cassava-metadata-preparation?scriptVersionId=47305070\" target=\"_blank\">https://www.kaggle.com/fireheart7/cassava-metadata-preparation?scriptVersionId=47305070</a></strong></p>\n<p>The exported dataset (<strong>compiled image statistics</strong>) is available at : <br>\n<strong><a href=\"https://www.kaggle.com/fireheart7/cassavaimagestatistics\" target=\"_blank\">https://www.kaggle.com/fireheart7/cassavaimagestatistics</a></strong></p>\n<hr>\n<p>Hope this boosts up your EDA in the competition. Soon, I'll append the custom TFrecords too. </p>\n<p>All the very best<br>\n~<em>Aditya Baurai</em></p>",
  "messages": [
    {
      "id": 1084728,
      "postDate": "2020-11-20T10:33:44.733Z",
      "content": "<p>Hello all!!</p>\n<p>I created this metadata .csv file (<strong>train_image_statistics.csv</strong>) encapsulating the compiled statistics of all our training images, provided by Kaggle for this competition. </p>\n<p>Attributes such as <em>shape, skewness, mean of each color channel, path, etc are all tabulated for each image</em>, stored, and exported as a .csv file. </p>\n<hr>\n<p><strong>Here is the notebook which I used for creating it. You can add or remove features accordingly</strong> : <br>\n<strong><a href=\"https://www.kaggle.com/fireheart7/cassava-metadata-preparation?scriptVersionId=47305070\" target=\"_blank\">https://www.kaggle.com/fireheart7/cassava-metadata-preparation?scriptVersionId=47305070</a></strong></p>\n<p>The exported dataset (<strong>compiled image statistics</strong>) is available at : <br>\n<strong><a href=\"https://www.kaggle.com/fireheart7/cassavaimagestatistics\" target=\"_blank\">https://www.kaggle.com/fireheart7/cassavaimagestatistics</a></strong></p>\n<hr>\n<p>Hope this boosts up your EDA in the competition. Soon, I'll append the custom TFrecords too. </p>\n<p>All the very best<br>\n~<em>Aditya Baurai</em></p>",
      "rawMarkdown": "Hello all!!\n\nI created this metadata .csv file (**train_image_statistics.csv**) encapsulating the compiled statistics of all our training images, provided by Kaggle for this competition. \n\nAttributes such as *shape, skewness, mean of each color channel, path, etc are all tabulated for each image*, stored, and exported as a .csv file. \n****\n**Here is the notebook which I used for creating it. You can add or remove features accordingly** : \n**https://www.kaggle.com/fireheart7/cassava-metadata-preparation?scriptVersionId=47305070**\n\nThe exported dataset (**compiled image statistics**) is available at : \n**https://www.kaggle.com/fireheart7/cassavaimagestatistics**\n****\nHope this boosts up your EDA in the competition. Soon, I'll append the custom TFrecords too. \n\nAll the very best\n~*Aditya Baurai*",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1084728": "Hello all!!\n\nI created this metadata .csv file (**train_image_statistics.csv**) encapsulating the compiled statistics of all our training images, provided by Kaggle for this competition. \n\nAttributes such as *shape, skewness, mean of each color channel, path, etc are all tabulated for each image*, stored, and exported as a .csv file. \n****\n**Here is the notebook which I used for creating it. You can add or remove features accordingly** : \n**https://www.kaggle.com/fireheart7/cassava-metadata-preparation?scriptVersionId=47305070**\n\nThe exported dataset (**compiled image statistics**) is available at : \n**https://www.kaggle.com/fireheart7/cassavaimagestatistics**\n****\nHope this boosts up your EDA in the competition. Soon, I'll append the custom TFrecords too. \n\nAll the very best\n~*Aditya Baurai*"
  }
}