{
  "id": 481397,
  "title": "Input_shape for multiple convolutional neural layer ",
  "url": "/competitions/digit-recognizer/discussion/481397",
  "author_name": "Abubacker S",
  "post_date": "2024-03-03T13:02:54.498000",
  "votes": 1,
  "comment_count": 3,
  "views": null,
  "content": "<p>In CNN , after convolutional , pooling layer what will be the input_shape for the next layer?</p>",
  "messages": [
    {
      "id": 2689522,
      "postDate": "2024-03-09T22:26:06.947Z",
      "content": "<p>In a CNN, after applying a convolutional layer or a pooling layer, the input shape for the next layer depends on the output shape of the previous layer. The output shape is determined by the hyperparameters used in the convolutional or pooling layer.</p>",
      "rawMarkdown": "In a CNN, after applying a convolutional layer or a pooling layer, the input shape for the next layer depends on the output shape of the previous layer. The output shape is determined by the hyperparameters used in the convolutional or pooling layer.",
      "votes": 2
    },
    {
      "id": 2689277,
      "postDate": "2024-03-09T19:20:42.270Z",
      "content": "<p><a href=\"https://www.kaggle.com/abubacker\" target=\"_blank\">@abubacker</a> my understanding is that you would want a fully connected layer after the pooling layer. The input shape would be the same shape as the output shape from the previous pooling layer. In Tensorflow, my understanding is that you would want to use a <code>Linear</code> layer for the fully connected layer. Hope this helps!</p>",
      "rawMarkdown": "@abubacker my understanding is that you would want a fully connected layer after the pooling layer. The input shape would be the same shape as the output shape from the previous pooling layer. In Tensorflow, my understanding is that you would want to use a `Linear` layer for the fully connected layer. Hope this helps!",
      "votes": 2
    },
    {
      "id": 2694082,
      "postDate": "2024-03-12T20:23:56.813Z",
      "content": "<p>the same as the ouput</p>",
      "rawMarkdown": "the same as the ouput",
      "votes": 1
    },
    {
      "id": 2679354,
      "postDate": "2024-03-03T13:02:54.500Z",
      "content": "<p>In CNN , after convolutional , pooling layer what will be the input_shape for the next layer?</p>",
      "rawMarkdown": "In CNN , after convolutional , pooling layer what will be the input_shape for the next layer?",
      "votes": 1
    }
  ],
  "comments": [
    {
      "id": 2689522,
      "author_name": "JverMeer",
      "author_url": "",
      "post_date": "2024-03-09T22:26:06.947000",
      "content": "<p>In a CNN, after applying a convolutional layer or a pooling layer, the input shape for the next layer depends on the output shape of the previous layer. The output shape is determined by the hyperparameters used in the convolutional or pooling layer.</p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 2689277,
      "author_name": "Craig Thomas",
      "author_url": "",
      "post_date": "2024-03-09T19:20:42.270000",
      "content": "<p><a href=\"https://www.kaggle.com/abubacker\" target=\"_blank\">@abubacker</a> my understanding is that you would want a fully connected layer after the pooling layer. The input shape would be the same shape as the output shape from the previous pooling layer. In Tensorflow, my understanding is that you would want to use a <code>Linear</code> layer for the fully connected layer. Hope this helps!</p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 2694082,
      "author_name": "Volodymyr Hotsiy",
      "author_url": "",
      "post_date": "2024-03-12T20:23:56.813000",
      "content": "<p>the same as the ouput</p>",
      "votes": 1,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2689522": "In a CNN, after applying a convolutional layer or a pooling layer, the input shape for the next layer depends on the output shape of the previous layer. The output shape is determined by the hyperparameters used in the convolutional or pooling layer.",
    "2689277": "@abubacker my understanding is that you would want a fully connected layer after the pooling layer. The input shape would be the same shape as the output shape from the previous pooling layer. In Tensorflow, my understanding is that you would want to use a `Linear` layer for the fully connected layer. Hope this helps!",
    "2694082": "the same as the ouput",
    "2679354": "In CNN , after convolutional , pooling layer what will be the input_shape for the next layer?"
  }
}