{
  "id": 307169,
  "title": "Variational Autoencoder with Pytorch",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/307169",
  "author_name": "",
  "post_date": "2022-02-13T01:54:42.817185900Z",
  "votes": 6,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi, I'm a beginner and now learning Variational Autoencoder with Pytorch, <br>\nbut it seems that it's not working well.</p>\n<p>Here's a link to the code I've implemented.</p>\n<p><a href=\"https://www.kaggle.com/manatoyaguchi/h-m-variational-autoencoder-with-pytorch\" target=\"_blank\">https://www.kaggle.com/manatoyaguchi/h-m-variational-autoencoder-with-pytorch</a></p>\n<p>What improvements are needed for successful learning?</p>\n<p>for example,</p>\n<ol>\n<li>change loss function</li>\n<li>speed up and train with more epochs<br>\n(=&gt; How can we speed up the process?)</li>\n</ol>\n<p>If the implementation of vae is successful, I think we can incorporate innovations such as recommending closer vectors in the latent space!</p>\n<p>Any advice would be appreciated!</p>",
  "messages": [
    {
      "id": "1687599",
      "postDate": "02/13/2022 01:54:42",
      "content": "<p>Hi, I'm a beginner and now learning Variational Autoencoder with Pytorch, <br>\nbut it seems that it's not working well.</p>\n<p>Here's a link to the code I've implemented.</p>\n<p><a href=\"https://www.kaggle.com/manatoyaguchi/h-m-variational-autoencoder-with-pytorch\" target=\"_blank\">https://www.kaggle.com/manatoyaguchi/h-m-variational-autoencoder-with-pytorch</a></p>\n<p>What improvements are needed for successful learning?</p>\n<p>for example,</p>\n<ol>\n<li>change loss function</li>\n<li>speed up and train with more epochs<br>\n(=&gt; How can we speed up the process?)</li>\n</ol>\n<p>If the implementation of vae is successful, I think we can incorporate innovations such as recommending closer vectors in the latent space!</p>\n<p>Any advice would be appreciated!</p>",
      "rawMarkdown": "Hi, I'm a beginner and now learning Variational Autoencoder with Pytorch, \nbut it seems that it's not working well.\n\nHere's a link to the code I've implemented.\n\nhttps://www.kaggle.com/manatoyaguchi/h-m-variational-autoencoder-with-pytorch\n\nWhat improvements are needed for successful learning?\n\nfor example,\n1. change loss function\n2. speed up and train with more epochs\n(=> How can we speed up the process?)\n\nIf the implementation of vae is successful, I think we can incorporate innovations such as recommending closer vectors in the latent space!\n\nAny advice would be appreciated!",
      "votes": null
    },
    {
      "id": "1700825",
      "postDate": "02/22/2022 09:28:05",
      "content": "<p>Maybe this can help a little to understand the difference between parameters <a href=\"https://stackabuse.com/autoencoders-for-image-reconstruction-in-python-and-keras/\" target=\"_blank\">https://stackabuse.com/autoencoders-for-image-reconstruction-in-python-and-keras/</a></p>",
      "rawMarkdown": "Maybe this can help a little to understand the difference between parameters https://stackabuse.com/autoencoders-for-image-reconstruction-in-python-and-keras/",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1700825,
      "author_name": "adrianowalmeida",
      "author_url": "",
      "post_date": "02/22/2022 09:28:05",
      "content": "<p>Maybe this can help a little to understand the difference between parameters <a href=\"https://stackabuse.com/autoencoders-for-image-reconstruction-in-python-and-keras/\" target=\"_blank\">https://stackabuse.com/autoencoders-for-image-reconstruction-in-python-and-keras/</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1687599": "Hi, I'm a beginner and now learning Variational Autoencoder with Pytorch, \nbut it seems that it's not working well.\n\nHere's a link to the code I've implemented.\n\nhttps://www.kaggle.com/manatoyaguchi/h-m-variational-autoencoder-with-pytorch\n\nWhat improvements are needed for successful learning?\n\nfor example,\n1. change loss function\n2. speed up and train with more epochs\n(=> How can we speed up the process?)\n\nIf the implementation of vae is successful, I think we can incorporate innovations such as recommending closer vectors in the latent space!\n\nAny advice would be appreciated!",
    "1700825": "Maybe this can help a little to understand the difference between parameters https://stackabuse.com/autoencoders-for-image-reconstruction-in-python-and-keras/"
  },
  "source": "meta"
}