{
  "id": 95016,
  "title": "Since I don't find the post to disclose the pre-trained models...",
  "url": "/competitions/imaterialist-fashion-2019-FGVC6/discussion/95016",
  "author_name": "",
  "post_date": "2019-06-09T08:18:20.766176400Z",
  "votes": 2,
  "comment_count": 7,
  "views": 0,
  "content": "<p>Here I want to disclose that our team's pipeline is modified from Facebook's implementation and we also use its pre-trained weights:\n<a href=\"https://github.com/facebookresearch/maskrcnn-benchmark\">https://github.com/facebookresearch/maskrcnn-benchmark</a></p>",
  "messages": [
    {
      "id": "548347",
      "postDate": "06/09/2019 08:18:20",
      "content": "<p>Here I want to disclose that our team's pipeline is modified from Facebook's implementation and we also use its pre-trained weights:\n<a href=\"https://github.com/facebookresearch/maskrcnn-benchmark\">https://github.com/facebookresearch/maskrcnn-benchmark</a></p>",
      "rawMarkdown": "Here I want to disclose that our team's pipeline is modified from Facebook's implementation and we also use its pre-trained weights:\nhttps://github.com/facebookresearch/maskrcnn-benchmark",
      "votes": null
    },
    {
      "id": "548439",
      "postDate": "06/09/2019 11:42:16",
      "content": "<p>Thanks for starting the topic!</p>\n\n<p>We're using pre-trained models from <a href=\"https://github.com/facebookresearch/maskrcnn-benchmark\">https://github.com/facebookresearch/maskrcnn-benchmark</a> and <a href=\"https://github.com/open-mmlab/mmdetection/\">https://github.com/open-mmlab/mmdetection/</a></p>",
      "rawMarkdown": "Thanks for starting the topic!\n\nWe're using pre-trained models from https://github.com/facebookresearch/maskrcnn-benchmark and https://github.com/open-mmlab/mmdetection/",
      "votes": null
    },
    {
      "id": "548502",
      "postDate": "06/09/2019 13:44:50",
      "content": "<p>I used mmdetection repo with thier pretrained models too</p>",
      "rawMarkdown": "I used mmdetection repo with thier pretrained models too",
      "votes": null
    },
    {
      "id": "548910",
      "postDate": "06/10/2019 05:03:53",
      "content": "<p>I used mmdetection and maskrcnn-benchmark too</p>",
      "rawMarkdown": "I used mmdetection and maskrcnn-benchmark too",
      "votes": null
    },
    {
      "id": "549058",
      "postDate": "06/10/2019 08:57:28",
      "content": "<p>I used this pre-trained model.\n<a href=\"https://github.com/pytorch/vision/tree/master/torchvision/models\">https://github.com/pytorch/vision/tree/master/torchvision/models</a></p>\n\n<p>And I automatically generate bboxes from masks and use them for training.</p>",
      "rawMarkdown": "I used this pre-trained model.\nhttps://github.com/pytorch/vision/tree/master/torchvision/models\n\nAnd I automatically generate bboxes from masks and use them for training.",
      "votes": null
    },
    {
      "id": "549063",
      "postDate": "06/10/2019 09:04:54",
      "content": "<p>Except for the pretrained weights from Facebook <a href=\"https://github.com/facebookresearch/maskrcnn-benchmark\">https://github.com/facebookresearch/maskrcnn-benchmark</a>  and mmlab <a href=\"https://github.com/open-mmlab/mmcv\">https://github.com/open-mmlab/mmcv</a>,  we also use the model from <a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>.</p>",
      "rawMarkdown": "Except for the pretrained weights from Facebook https://github.com/facebookresearch/maskrcnn-benchmark  and mmlab https://github.com/open-mmlab/mmcv,  we also use the model from https://github.com/Cadene/pretrained-models.pytorch.",
      "votes": null
    },
    {
      "id": "549174",
      "postDate": "06/10/2019 11:31:55",
      "content": "<p>I used mmdetection repo with thier pretrained models too</p>",
      "rawMarkdown": "I used mmdetection repo with thier pretrained models too",
      "votes": null
    },
    {
      "id": "549245",
      "postDate": "06/10/2019 13:05:26",
      "content": "<p>mmdetection for me too</p>",
      "rawMarkdown": "mmdetection for me too",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 548439,
      "author_name": "lopuhin",
      "author_url": "",
      "post_date": "06/09/2019 11:42:16",
      "content": "<p>Thanks for starting the topic!</p>\n\n<p>We're using pre-trained models from <a href=\"https://github.com/facebookresearch/maskrcnn-benchmark\">https://github.com/facebookresearch/maskrcnn-benchmark</a> and <a href=\"https://github.com/open-mmlab/mmdetection/\">https://github.com/open-mmlab/mmdetection/</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 548502,
          "author_name": "stalkermustang",
          "author_url": "",
          "post_date": "06/09/2019 13:44:50",
          "content": "<p>I used mmdetection repo with thier pretrained models too</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 548910,
          "author_name": "hesene",
          "author_url": "",
          "post_date": "06/10/2019 05:03:53",
          "content": "<p>I used mmdetection and maskrcnn-benchmark too</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 549174,
          "author_name": "amiras",
          "author_url": "",
          "post_date": "06/10/2019 11:31:55",
          "content": "<p>I used mmdetection repo with thier pretrained models too</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 549245,
          "author_name": "ikibardin",
          "author_url": "",
          "post_date": "06/10/2019 13:05:26",
          "content": "<p>mmdetection for me too</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 549058,
      "author_name": "rskmoi",
      "author_url": "",
      "post_date": "06/10/2019 08:57:28",
      "content": "<p>I used this pre-trained model.\n<a href=\"https://github.com/pytorch/vision/tree/master/torchvision/models\">https://github.com/pytorch/vision/tree/master/torchvision/models</a></p>\n\n<p>And I automatically generate bboxes from masks and use them for training.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 549063,
      "author_name": "clannad",
      "author_url": "",
      "post_date": "06/10/2019 09:04:54",
      "content": "<p>Except for the pretrained weights from Facebook <a href=\"https://github.com/facebookresearch/maskrcnn-benchmark\">https://github.com/facebookresearch/maskrcnn-benchmark</a>  and mmlab <a href=\"https://github.com/open-mmlab/mmcv\">https://github.com/open-mmlab/mmcv</a>,  we also use the model from <a href=\"https://github.com/Cadene/pretrained-models.pytorch\">https://github.com/Cadene/pretrained-models.pytorch</a>.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "548347": "Here I want to disclose that our team's pipeline is modified from Facebook's implementation and we also use its pre-trained weights:\nhttps://github.com/facebookresearch/maskrcnn-benchmark",
    "548439": "Thanks for starting the topic!\n\nWe're using pre-trained models from https://github.com/facebookresearch/maskrcnn-benchmark and https://github.com/open-mmlab/mmdetection/",
    "548502": "I used mmdetection repo with thier pretrained models too",
    "548910": "I used mmdetection and maskrcnn-benchmark too",
    "549058": "I used this pre-trained model.\nhttps://github.com/pytorch/vision/tree/master/torchvision/models\n\nAnd I automatically generate bboxes from masks and use them for training.",
    "549063": "Except for the pretrained weights from Facebook https://github.com/facebookresearch/maskrcnn-benchmark  and mmlab https://github.com/open-mmlab/mmcv,  we also use the model from https://github.com/Cadene/pretrained-models.pytorch.",
    "549174": "I used mmdetection repo with thier pretrained models too",
    "549245": "mmdetection for me too"
  },
  "source": "meta"
}