{
  "id": 130032,
  "title": "Getting “HTTP Error 403: Forbidden” in google colab",
  "url": "/competitions/bengaliai-cv19/discussion/130032",
  "author_name": "",
  "post_date": "2020-02-11T19:32:51.886924700Z",
  "votes": 8,
  "comment_count": 17,
  "views": 0,
  "content": "<p>in google colab today i was trying to load pretrained weights for model  training but i am \"Getting “HTTP Error 403: Forbidden” \" error!\nanyone faced similar problem before?</p>",
  "messages": [
    {
      "id": "743093",
      "postDate": "02/11/2020 19:32:51",
      "content": "<p>in google colab today i was trying to load pretrained weights for model  training but i am \"Getting “HTTP Error 403: Forbidden” \" error!\nanyone faced similar problem before?</p>",
      "rawMarkdown": "in google colab today i was trying to load pretrained weights for model  training but i am \"Getting “HTTP Error 403: Forbidden” \" error!\nanyone faced similar problem before?",
      "votes": null
    },
    {
      "id": "743483",
      "postDate": "02/12/2020 05:26:49",
      "content": "<p>I'm, but I've got this error when trying to load pre-trained weights in kaggle's notebook.</p>",
      "rawMarkdown": "I'm, but I've got this error when trying to load pre-trained weights in kaggle's notebook.",
      "votes": null
    },
    {
      "id": "743520",
      "postDate": "02/12/2020 05:59:26",
      "content": "<p>Same problem in Colab. Two days ago it wasn't an issue.</p>",
      "rawMarkdown": "Same problem in Colab. Two days ago it wasn't an issue.",
      "votes": null
    },
    {
      "id": "743621",
      "postDate": "02/12/2020 06:28:24",
      "content": "<p>I have also started facing this issue recently, in kaggle kernels.</p>\n\n<p><a href=\"/sohier\">@sohier</a> <a href=\"/addisonhoward\">@addisonhoward</a> <a href=\"/reasat\">@reasat</a> <a href=\"/imtiazprio\">@imtiazprio</a> </p>",
      "rawMarkdown": "I have also started facing this issue recently, in kaggle kernels.\n\n@sohier @addisonhoward @reasat @imtiazprio",
      "votes": null
    },
    {
      "id": "743627",
      "postDate": "02/12/2020 06:33:10",
      "content": "<p>same issue 👎  I guess it is torchvision problem, not kaggle or colab.</p>",
      "rawMarkdown": "same issue 👎  I guess it is torchvision problem, not kaggle or colab.",
      "votes": null
    },
    {
      "id": "743737",
      "postDate": "02/12/2020 08:48:22",
      "content": "<p>A Github issue has been created on the same:</p>\n\n<p><code>https://github.com/pytorch/vision/issues/1876</code></p>",
      "rawMarkdown": "A Github issue has been created on the same:\n\n`https://github.com/pytorch/vision/issues/1876`",
      "votes": null
    },
    {
      "id": "743762",
      "postDate": "02/12/2020 09:26:11",
      "content": "<p>I am also having the same issue since today. Please someone provide a solution fast. I have to submit a project. Thanks in advance. </p>",
      "rawMarkdown": "I am also having the same issue since today. Please someone provide a solution fast. I have to submit a project. Thanks in advance.",
      "votes": null
    },
    {
      "id": "743777",
      "postDate": "02/12/2020 09:51:05",
      "content": "<p>Yes, although I saved pretrained weights on google drive and just copy the weights from there to the cache folder.  </p>",
      "rawMarkdown": "Yes, although I saved pretrained weights on google drive and just copy the weights from there to the cache folder.",
      "votes": null
    },
    {
      "id": "743800",
      "postDate": "02/12/2020 10:14:34",
      "content": "<p>As said here : <a href=\"https://github.com/pytorch/vision/issues/1876#issuecomment-585046536\">https://github.com/pytorch/vision/issues/1876#issuecomment-585046536</a>\nA temporary workaround is to to update the <code>model_urls</code>prefixes from <code>https://download.pytorch.org/models/</code> to <code>https://s3.amazonaws.com/pytorch/models/</code></p>",
      "rawMarkdown": "As said here : https://github.com/pytorch/vision/issues/1876#issuecomment-585046536\nA temporary workaround is to to update the `model_urls `prefixes from `https://download.pytorch.org/models/` to `https://s3.amazonaws.com/pytorch/models/`",
      "votes": null
    },
    {
      "id": "743861",
      "postDate": "02/12/2020 10:34:34",
      "content": "<p>Why is this a temporary workaround?</p>",
      "rawMarkdown": "Why is this a temporary workaround?",
      "votes": null
    },
    {
      "id": "743866",
      "postDate": "02/12/2020 10:39:53",
      "content": "<p>Because it seems retrieving .pth models from amazon aws servers is working. However, this solution need to change some code of pytorch source, thus it is only a temporary workaround</p>",
      "rawMarkdown": "Because it seems retrieving .pth models from amazon aws servers is working. However, this solution need to change some code of pytorch source, thus it is only a temporary workaround",
      "votes": null
    },
    {
      "id": "743881",
      "postDate": "02/12/2020 11:04:55",
      "content": "<p>Is that solution worked for anyone??</p>",
      "rawMarkdown": "Is that solution worked for anyone??",
      "votes": null
    },
    {
      "id": "743887",
      "postDate": "02/12/2020 11:15:27",
      "content": "<p>You can use the following code right at the start of your notebook,</p>\n\n<p>!wget <a href=\"https://s3.amazonaws.com/pytorch/models/vgg11-bbd30ac9.pth\">https://s3.amazonaws.com/pytorch/models/vgg11-bbd30ac9.pth</a>\n!mkdir -p /root/.cache/torch/checkpoints/\n!mv vgg11-bbd30ac9.pth /root/.cache/torch/checkpoints/vgg11-bbd30ac9.pth</p>\n\n<p>Just update the link with the failed one : <a href=\"https://s3.amazonaws.com/pytorch/models/YOUR_MODEL_HERE\">https://s3.amazonaws.com/pytorch/models/YOUR_MODEL_HERE</a></p>",
      "rawMarkdown": "You can use the following code right at the start of your notebook,\n\n!wget https://s3.amazonaws.com/pytorch/models/vgg11-bbd30ac9.pth\n!mkdir -p /root/.cache/torch/checkpoints/\n!mv vgg11-bbd30ac9.pth /root/.cache/torch/checkpoints/vgg11-bbd30ac9.pth\n\n\nJust update the link with the failed one : https://s3.amazonaws.com/pytorch/models/YOUR_MODEL_HERE",
      "votes": null
    },
    {
      "id": "743907",
      "postDate": "02/12/2020 11:35:08",
      "content": "<p>Did you find any solution???</p>",
      "rawMarkdown": "Did you find any solution???",
      "votes": null
    },
    {
      "id": "744098",
      "postDate": "02/12/2020 14:56:57",
      "content": "<p>download pth file with weights worked for me</p>",
      "rawMarkdown": "download pth file with weights worked for me",
      "votes": null
    },
    {
      "id": "744100",
      "postDate": "02/12/2020 14:58:11",
      "content": "<p>Looks like it started working again.</p>",
      "rawMarkdown": "Looks like it started working again.",
      "votes": null
    },
    {
      "id": "744240",
      "postDate": "02/12/2020 17:17:11",
      "content": "<p>Really? When?</p>",
      "rawMarkdown": "Really? When?",
      "votes": null
    },
    {
      "id": "744242",
      "postDate": "02/12/2020 17:17:27",
      "content": "<p>Thanks</p>",
      "rawMarkdown": "Thanks",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 743483,
      "author_name": "phunghieu",
      "author_url": "",
      "post_date": "02/12/2020 05:26:49",
      "content": "<p>I'm, but I've got this error when trying to load pre-trained weights in kaggle's notebook.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 743520,
      "author_name": "arturdatascientist",
      "author_url": "",
      "post_date": "02/12/2020 05:59:26",
      "content": "<p>Same problem in Colab. Two days ago it wasn't an issue.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 743621,
      "author_name": "rohitagarwal",
      "author_url": "",
      "post_date": "02/12/2020 06:28:24",
      "content": "<p>I have also started facing this issue recently, in kaggle kernels.</p>\n\n<p><a href=\"/sohier\">@sohier</a> <a href=\"/addisonhoward\">@addisonhoward</a> <a href=\"/reasat\">@reasat</a> <a href=\"/imtiazprio\">@imtiazprio</a> </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 743627,
      "author_name": "kupchanski",
      "author_url": "",
      "post_date": "02/12/2020 06:33:10",
      "content": "<p>same issue 👎  I guess it is torchvision problem, not kaggle or colab.</p>",
      "votes": null,
      "replies": [
        {
          "id": 743907,
          "author_name": "ashkatkrowser",
          "author_url": "",
          "post_date": "02/12/2020 11:35:08",
          "content": "<p>Did you find any solution???</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 744098,
          "author_name": "kupchanski",
          "author_url": "",
          "post_date": "02/12/2020 14:56:57",
          "content": "<p>download pth file with weights worked for me</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 743737,
      "author_name": "rosewambui",
      "author_url": "",
      "post_date": "02/12/2020 08:48:22",
      "content": "<p>A Github issue has been created on the same:</p>\n\n<p><code>https://github.com/pytorch/vision/issues/1876</code></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 743762,
      "author_name": "ashkatkrowser",
      "author_url": "",
      "post_date": "02/12/2020 09:26:11",
      "content": "<p>I am also having the same issue since today. Please someone provide a solution fast. I have to submit a project. Thanks in advance. </p>",
      "votes": null,
      "replies": [
        {
          "id": 743887,
          "author_name": "tamirmal",
          "author_url": "",
          "post_date": "02/12/2020 11:15:27",
          "content": "<p>You can use the following code right at the start of your notebook,</p>\n\n<p>!wget <a href=\"https://s3.amazonaws.com/pytorch/models/vgg11-bbd30ac9.pth\">https://s3.amazonaws.com/pytorch/models/vgg11-bbd30ac9.pth</a>\n!mkdir -p /root/.cache/torch/checkpoints/\n!mv vgg11-bbd30ac9.pth /root/.cache/torch/checkpoints/vgg11-bbd30ac9.pth</p>\n\n<p>Just update the link with the failed one : <a href=\"https://s3.amazonaws.com/pytorch/models/YOUR_MODEL_HERE\">https://s3.amazonaws.com/pytorch/models/YOUR_MODEL_HERE</a></p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 743777,
      "author_name": "timetraveller98",
      "author_url": "",
      "post_date": "02/12/2020 09:51:05",
      "content": "<p>Yes, although I saved pretrained weights on google drive and just copy the weights from there to the cache folder.  </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 743800,
      "author_name": "yohann84l",
      "author_url": "",
      "post_date": "02/12/2020 10:14:34",
      "content": "<p>As said here : <a href=\"https://github.com/pytorch/vision/issues/1876#issuecomment-585046536\">https://github.com/pytorch/vision/issues/1876#issuecomment-585046536</a>\nA temporary workaround is to to update the <code>model_urls</code>prefixes from <code>https://download.pytorch.org/models/</code> to <code>https://s3.amazonaws.com/pytorch/models/</code></p>",
      "votes": null,
      "replies": [
        {
          "id": 743861,
          "author_name": "rohitagarwal",
          "author_url": "",
          "post_date": "02/12/2020 10:34:34",
          "content": "<p>Why is this a temporary workaround?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 743866,
          "author_name": "yohann84l",
          "author_url": "",
          "post_date": "02/12/2020 10:39:53",
          "content": "<p>Because it seems retrieving .pth models from amazon aws servers is working. However, this solution need to change some code of pytorch source, thus it is only a temporary workaround</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 743881,
          "author_name": "ashkatkrowser",
          "author_url": "",
          "post_date": "02/12/2020 11:04:55",
          "content": "<p>Is that solution worked for anyone??</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 744242,
          "author_name": "mobassir",
          "author_url": "",
          "post_date": "02/12/2020 17:17:27",
          "content": "<p>Thanks</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 744100,
      "author_name": "kupchanski",
      "author_url": "",
      "post_date": "02/12/2020 14:58:11",
      "content": "<p>Looks like it started working again.</p>",
      "votes": null,
      "replies": [
        {
          "id": 744240,
          "author_name": "mobassir",
          "author_url": "",
          "post_date": "02/12/2020 17:17:11",
          "content": "<p>Really? When?</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "743093": "in google colab today i was trying to load pretrained weights for model  training but i am \"Getting “HTTP Error 403: Forbidden” \" error!\nanyone faced similar problem before?",
    "743483": "I'm, but I've got this error when trying to load pre-trained weights in kaggle's notebook.",
    "743520": "Same problem in Colab. Two days ago it wasn't an issue.",
    "743621": "I have also started facing this issue recently, in kaggle kernels.\n\n@sohier @addisonhoward @reasat @imtiazprio",
    "743627": "same issue 👎  I guess it is torchvision problem, not kaggle or colab.",
    "743737": "A Github issue has been created on the same:\n\n`https://github.com/pytorch/vision/issues/1876`",
    "743762": "I am also having the same issue since today. Please someone provide a solution fast. I have to submit a project. Thanks in advance.",
    "743777": "Yes, although I saved pretrained weights on google drive and just copy the weights from there to the cache folder.",
    "743800": "As said here : https://github.com/pytorch/vision/issues/1876#issuecomment-585046536\nA temporary workaround is to to update the `model_urls `prefixes from `https://download.pytorch.org/models/` to `https://s3.amazonaws.com/pytorch/models/`",
    "743861": "Why is this a temporary workaround?",
    "743866": "Because it seems retrieving .pth models from amazon aws servers is working. However, this solution need to change some code of pytorch source, thus it is only a temporary workaround",
    "743881": "Is that solution worked for anyone??",
    "743887": "You can use the following code right at the start of your notebook,\n\n!wget https://s3.amazonaws.com/pytorch/models/vgg11-bbd30ac9.pth\n!mkdir -p /root/.cache/torch/checkpoints/\n!mv vgg11-bbd30ac9.pth /root/.cache/torch/checkpoints/vgg11-bbd30ac9.pth\n\n\nJust update the link with the failed one : https://s3.amazonaws.com/pytorch/models/YOUR_MODEL_HERE",
    "743907": "Did you find any solution???",
    "744098": "download pth file with weights worked for me",
    "744100": "Looks like it started working again.",
    "744240": "Really? When?",
    "744242": "Thanks"
  },
  "source": "meta"
}