{
  "id": 198731,
  "title": "Looking what CNN learns",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/198731",
  "author_name": "",
  "post_date": "2020-11-22T18:09:09.355182Z",
  "votes": 7,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I was just curious as I wanted to know what my cnn model was learning, anyone figuring it out?<br>\nIf you know any notebooks, resources  which make it easy to find out what the cnn's learning please give a link in the comments.<br>\nThank you.</p>",
  "messages": [
    {
      "id": "1087458",
      "postDate": "11/22/2020 18:09:09",
      "content": "<p>I was just curious as I wanted to know what my cnn model was learning, anyone figuring it out?<br>\nIf you know any notebooks, resources  which make it easy to find out what the cnn's learning please give a link in the comments.<br>\nThank you.</p>",
      "rawMarkdown": "I was just curious as I wanted to know what my cnn model was learning, anyone figuring it out?\nIf you know any notebooks, resources  which make it easy to find out what the cnn's learning please give a link in the comments.\nThank you.",
      "votes": null
    },
    {
      "id": "1087476",
      "postDate": "11/22/2020 18:33:53",
      "content": "<p>You can use Grad-Cam <a href=\"url\" target=\"_blank\">https://arxiv.org/pdf/1610.02391.pdf</a><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2383059%2F090f47a02ec5ce1235d2f2afb8645a99%2F100.png?generation=1606070105173155&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "You can use Grad-Cam [https://arxiv.org/pdf/1610.02391.pdf](url)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2383059%2F090f47a02ec5ce1235d2f2afb8645a99%2F100.png?generation=1606070105173155&alt=media)",
      "votes": null
    },
    {
      "id": "1087547",
      "postDate": "11/22/2020 20:38:38",
      "content": "<p>Topic <a href=\"https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/160147\" target=\"_blank\">CNN Input Size Explained\n</a> by <a href=\"https://www.kaggle.com/cdeotte\" target=\"_blank\">@cdeotte</a> </p>\n<p>Notebook <a href=\"https://www.kaggle.com/andradaolteanu/convolutional-neural-nets-cnns-explained\" target=\"_blank\">CNNs Explained</a> by <a href=\"https://www.kaggle.com/andradaolteanu\" target=\"_blank\">@andradaolteanu</a> </p>\n<p>I think you'll quickly find all you need to know in these two resources </p>\n<p>Good luck ! </p>",
      "rawMarkdown": "Topic [CNN Input Size Explained\n](https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/160147) by @cdeotte \n\nNotebook [CNNs Explained](https://www.kaggle.com/andradaolteanu/convolutional-neural-nets-cnns-explained) by @andradaolteanu \n\nI think you'll quickly find all you need to know in these two resources \n\nGood luck !",
      "votes": null
    },
    {
      "id": "1087627",
      "postDate": "11/22/2020 23:14:46",
      "content": "<p>I have done this a couple of time using SHAP, here for <a href=\"https://www.kaggle.com/dimitreoliveira/diabetic-retinopathy-shap-model-explainability\" target=\"_blank\">Diabetic retinopathy</a> and here for <a href=\"https://www.kaggle.com/dimitreoliveira/melanoma-classification-shap-model-explained\" target=\"_blank\">Melanoma Classification</a>.</p>",
      "rawMarkdown": "I have done this a couple of time using SHAP, here for [Diabetic retinopathy](https://www.kaggle.com/dimitreoliveira/diabetic-retinopathy-shap-model-explainability) and here for [Melanoma Classification](https://www.kaggle.com/dimitreoliveira/melanoma-classification-shap-model-explained).",
      "votes": null
    },
    {
      "id": "1087741",
      "postDate": "11/23/2020 03:48:15",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/mrinath\" target=\"_blank\">@mrinath</a>  this will help you<br>\n<a href=\"https://www.kaggle.com/vanvalkenberg/what-a-cnn-sees-plotting-feature-maps-of-cnn\" target=\"_blank\">https://www.kaggle.com/vanvalkenberg/what-a-cnn-sees-plotting-feature-maps-of-cnn</a></p>\n<p>it's a script for plotting Cnn's feature maps.<br>\nThanks.</p>",
      "rawMarkdown": "Hi @mrinath  this will help you\nhttps://www.kaggle.com/vanvalkenberg/what-a-cnn-sees-plotting-feature-maps-of-cnn\n\nit's a script for plotting Cnn's feature maps.\nThanks.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1087476,
      "author_name": "socratis",
      "author_url": "",
      "post_date": "11/22/2020 18:33:53",
      "content": "<p>You can use Grad-Cam <a href=\"url\" target=\"_blank\">https://arxiv.org/pdf/1610.02391.pdf</a><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2383059%2F090f47a02ec5ce1235d2f2afb8645a99%2F100.png?generation=1606070105173155&amp;alt=media\" alt=\"\"></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1087547,
      "author_name": "serigne",
      "author_url": "",
      "post_date": "11/22/2020 20:38:38",
      "content": "<p>Topic <a href=\"https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/160147\" target=\"_blank\">CNN Input Size Explained\n</a> by <a href=\"https://www.kaggle.com/cdeotte\" target=\"_blank\">@cdeotte</a> </p>\n<p>Notebook <a href=\"https://www.kaggle.com/andradaolteanu/convolutional-neural-nets-cnns-explained\" target=\"_blank\">CNNs Explained</a> by <a href=\"https://www.kaggle.com/andradaolteanu\" target=\"_blank\">@andradaolteanu</a> </p>\n<p>I think you'll quickly find all you need to know in these two resources </p>\n<p>Good luck ! </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1087627,
      "author_name": "dimitreoliveira",
      "author_url": "",
      "post_date": "11/22/2020 23:14:46",
      "content": "<p>I have done this a couple of time using SHAP, here for <a href=\"https://www.kaggle.com/dimitreoliveira/diabetic-retinopathy-shap-model-explainability\" target=\"_blank\">Diabetic retinopathy</a> and here for <a href=\"https://www.kaggle.com/dimitreoliveira/melanoma-classification-shap-model-explained\" target=\"_blank\">Melanoma Classification</a>.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1087741,
      "author_name": "vanvalkenberg",
      "author_url": "",
      "post_date": "11/23/2020 03:48:15",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/mrinath\" target=\"_blank\">@mrinath</a>  this will help you<br>\n<a href=\"https://www.kaggle.com/vanvalkenberg/what-a-cnn-sees-plotting-feature-maps-of-cnn\" target=\"_blank\">https://www.kaggle.com/vanvalkenberg/what-a-cnn-sees-plotting-feature-maps-of-cnn</a></p>\n<p>it's a script for plotting Cnn's feature maps.<br>\nThanks.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1087458": "I was just curious as I wanted to know what my cnn model was learning, anyone figuring it out?\nIf you know any notebooks, resources  which make it easy to find out what the cnn's learning please give a link in the comments.\nThank you.",
    "1087476": "You can use Grad-Cam [https://arxiv.org/pdf/1610.02391.pdf](url)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2383059%2F090f47a02ec5ce1235d2f2afb8645a99%2F100.png?generation=1606070105173155&alt=media)",
    "1087547": "Topic [CNN Input Size Explained\n](https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/160147) by @cdeotte \n\nNotebook [CNNs Explained](https://www.kaggle.com/andradaolteanu/convolutional-neural-nets-cnns-explained) by @andradaolteanu \n\nI think you'll quickly find all you need to know in these two resources \n\nGood luck !",
    "1087627": "I have done this a couple of time using SHAP, here for [Diabetic retinopathy](https://www.kaggle.com/dimitreoliveira/diabetic-retinopathy-shap-model-explainability) and here for [Melanoma Classification](https://www.kaggle.com/dimitreoliveira/melanoma-classification-shap-model-explained).",
    "1087741": "Hi @mrinath  this will help you\nhttps://www.kaggle.com/vanvalkenberg/what-a-cnn-sees-plotting-feature-maps-of-cnn\n\nit's a script for plotting Cnn's feature maps.\nThanks."
  },
  "source": "meta"
}