{
  "id": 127149,
  "title": "Visualize your results",
  "url": "/competitions/bengaliai-cv19/discussion/127149",
  "author_name": "",
  "post_date": "2020-01-22T17:04:13.794917400Z",
  "votes": 73,
  "comment_count": 17,
  "views": 0,
  "content": "<p>I've created this script just for fun, but it might be useful for analysis as well.</p>\n\n<p>It creates an animation about your model's predictions. You set a sample, and some transformation (for example, rotate from 0 to 90; step size 1), and it calculates your model's outputs (probabilities for the labels) for every step.</p>\n\n<h2>Implementation</h2>\n\n<p><a href=\"https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions\">https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions</a></p>\n\n<h2>Examples</h2>\n\n<p><img src=\"https://s5.gifyu.com/images/demo_trained_weights_misscls_2.gif\" alt=\"\"></p>\n\n<p><img src=\"https://s5.gifyu.com/images/demo_trained_weights_misscls_scale_2.gif\" alt=\"\"></p>",
  "messages": [
    {
      "id": "725955",
      "postDate": "01/22/2020 17:04:13",
      "content": "<p>I've created this script just for fun, but it might be useful for analysis as well.</p>\n\n<p>It creates an animation about your model's predictions. You set a sample, and some transformation (for example, rotate from 0 to 90; step size 1), and it calculates your model's outputs (probabilities for the labels) for every step.</p>\n\n<h2>Implementation</h2>\n\n<p><a href=\"https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions\">https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions</a></p>\n\n<h2>Examples</h2>\n\n<p><img src=\"https://s5.gifyu.com/images/demo_trained_weights_misscls_2.gif\" alt=\"\"></p>\n\n<p><img src=\"https://s5.gifyu.com/images/demo_trained_weights_misscls_scale_2.gif\" alt=\"\"></p>",
      "rawMarkdown": "I've created this script just for fun, but it might be useful for analysis as well.\n\nIt creates an animation about your model's predictions. You set a sample, and some transformation (for example, rotate from 0 to 90; step size 1), and it calculates your model's outputs (probabilities for the labels) for every step.\n\n## Implementation\n[https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions](https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions)\n\n\n## Examples\n\n![](https://s5.gifyu.com/images/demo_trained_weights_misscls_2.gif)\n\n![](https://s5.gifyu.com/images/demo_trained_weights_misscls_scale_2.gif)",
      "votes": null
    },
    {
      "id": "725960",
      "postDate": "01/22/2020 17:06:18",
      "content": "<p>Wow. Thank you for sharing! :)</p>",
      "rawMarkdown": "Wow. Thank you for sharing! :)",
      "votes": null
    },
    {
      "id": "725974",
      "postDate": "01/22/2020 17:18:49",
      "content": "<p>Nice..... Great job....!</p>",
      "rawMarkdown": "Nice..... Great job....!",
      "votes": null
    },
    {
      "id": "725986",
      "postDate": "01/22/2020 17:32:51",
      "content": "<p>Great\nAmazing!! <a href=\"/pestipeti\">@pestipeti</a> </p>",
      "rawMarkdown": "Great\nAmazing!! @pestipeti",
      "votes": null
    },
    {
      "id": "726367",
      "postDate": "01/23/2020 01:29:41",
      "content": "<p>Thanks for sharing!</p>",
      "rawMarkdown": "Thanks for sharing!",
      "votes": null
    },
    {
      "id": "726538",
      "postDate": "01/23/2020 04:05:13",
      "content": "<p>Wow, that's awesome! Thanks for sharing!</p>",
      "rawMarkdown": "Wow, that's awesome! Thanks for sharing!",
      "votes": null
    },
    {
      "id": "726667",
      "postDate": "01/23/2020 06:40:11",
      "content": "<p>Amazing</p>",
      "rawMarkdown": "Amazing",
      "votes": null
    },
    {
      "id": "727161",
      "postDate": "01/23/2020 14:00:31",
      "content": "<p><a href=\"/pestipeti\">@pestipeti</a> Amazing!!! one of the best (unique) visualization I have seen !!</p>",
      "rawMarkdown": "pestipeti Amazing!!! one of the best (unique) visualization I have seen !!",
      "votes": null
    },
    {
      "id": "728019",
      "postDate": "01/24/2020 10:39:48",
      "content": "<p>Amazing! Thanks for sharing !</p>",
      "rawMarkdown": "Amazing! Thanks for sharing !",
      "votes": null
    },
    {
      "id": "728090",
      "postDate": "01/24/2020 11:46:04",
      "content": "<p>really helpful, Thanks for sharing </p>",
      "rawMarkdown": "really helpful, Thanks for sharing",
      "votes": null
    },
    {
      "id": "728225",
      "postDate": "01/24/2020 14:15:14",
      "content": "<p>this is really helpful as an augmentation tips, thank you for sharing!</p>",
      "rawMarkdown": "this is really helpful as an augmentation tips, thank you for sharing!",
      "votes": null
    },
    {
      "id": "728728",
      "postDate": "01/25/2020 06:45:25",
      "content": "<p>666</p>",
      "rawMarkdown": "666",
      "votes": null
    },
    {
      "id": "729670",
      "postDate": "01/26/2020 14:05:40",
      "content": "<p>Nice.</p>",
      "rawMarkdown": "Nice.",
      "votes": null
    },
    {
      "id": "730424",
      "postDate": "01/27/2020 13:23:44",
      "content": "<p>Good Try Keep it up</p>",
      "rawMarkdown": "Good Try Keep it up",
      "votes": null
    },
    {
      "id": "731796",
      "postDate": "01/29/2020 04:03:08",
      "content": "<p>Awsome❗️ Thanks for sharing😄</p>",
      "rawMarkdown": "Awsome❗️ Thanks for sharing😄",
      "votes": null
    },
    {
      "id": "1092045",
      "postDate": "11/26/2020 13:54:34",
      "content": "<p>Great work</p>",
      "rawMarkdown": "Great work",
      "votes": null
    },
    {
      "id": "1138856",
      "postDate": "01/05/2021 02:46:40",
      "content": "<p>Looks great!</p>",
      "rawMarkdown": "Looks great!",
      "votes": null
    },
    {
      "id": "2852010",
      "postDate": "06/03/2024 04:45:25",
      "content": "<p>Nice work, Peter! Your script seems both fun and practical. Visualizing predictions can indeed offer deeper insights into model performance. Thanks for sharing the implementation and examples!</p>",
      "rawMarkdown": "Nice work, Peter! Your script seems both fun and practical. Visualizing predictions can indeed offer deeper insights into model performance. Thanks for sharing the implementation and examples!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1092045,
      "author_name": "amaanvora",
      "author_url": "",
      "post_date": "11/26/2020 13:54:34",
      "content": "<p>Great work</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1138856,
      "author_name": "macchitaka",
      "author_url": "",
      "post_date": "01/05/2021 02:46:40",
      "content": "<p>Looks great!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2852010,
      "author_name": "mubashirsidiki",
      "author_url": "",
      "post_date": "06/03/2024 04:45:25",
      "content": "<p>Nice work, Peter! Your script seems both fun and practical. Visualizing predictions can indeed offer deeper insights into model performance. Thanks for sharing the implementation and examples!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 725960,
      "author_name": "binaicrai",
      "author_url": "",
      "post_date": "01/22/2020 17:06:18",
      "content": "<p>Wow. Thank you for sharing! :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 725974,
      "author_name": "",
      "author_url": "",
      "post_date": "01/22/2020 17:18:49",
      "content": "<p>Nice..... Great job....!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 725986,
      "author_name": "veeralakrishna",
      "author_url": "",
      "post_date": "01/22/2020 17:32:51",
      "content": "<p>Great\nAmazing!! <a href=\"/pestipeti\">@pestipeti</a> </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 726367,
      "author_name": "saharseidi",
      "author_url": "",
      "post_date": "01/23/2020 01:29:41",
      "content": "<p>Thanks for sharing!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 726538,
      "author_name": "cdeotte",
      "author_url": "",
      "post_date": "01/23/2020 04:05:13",
      "content": "<p>Wow, that's awesome! Thanks for sharing!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 726667,
      "author_name": "kurianbenoy",
      "author_url": "",
      "post_date": "01/23/2020 06:40:11",
      "content": "<p>Amazing</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 727161,
      "author_name": "chekoduadarsh",
      "author_url": "",
      "post_date": "01/23/2020 14:00:31",
      "content": "<p><a href=\"/pestipeti\">@pestipeti</a> Amazing!!! one of the best (unique) visualization I have seen !!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 728019,
      "author_name": "",
      "author_url": "",
      "post_date": "01/24/2020 10:39:48",
      "content": "<p>Amazing! Thanks for sharing !</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 728090,
      "author_name": "negi009",
      "author_url": "",
      "post_date": "01/24/2020 11:46:04",
      "content": "<p>really helpful, Thanks for sharing </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 728225,
      "author_name": "markson14",
      "author_url": "",
      "post_date": "01/24/2020 14:15:14",
      "content": "<p>this is really helpful as an augmentation tips, thank you for sharing!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 728728,
      "author_name": "tiandaye",
      "author_url": "",
      "post_date": "01/25/2020 06:45:25",
      "content": "<p>666</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 729670,
      "author_name": "machinelp",
      "author_url": "",
      "post_date": "01/26/2020 14:05:40",
      "content": "<p>Nice.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 730424,
      "author_name": "tayyabali55",
      "author_url": "",
      "post_date": "01/27/2020 13:23:44",
      "content": "<p>Good Try Keep it up</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 731796,
      "author_name": "mashlyn",
      "author_url": "",
      "post_date": "01/29/2020 04:03:08",
      "content": "<p>Awsome❗️ Thanks for sharing😄</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "725955": "I've created this script just for fun, but it might be useful for analysis as well.\n\nIt creates an animation about your model's predictions. You set a sample, and some transformation (for example, rotate from 0 to 90; step size 1), and it calculates your model's outputs (probabilities for the labels) for every step.\n\n## Implementation\n[https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions](https://www.kaggle.com/pestipeti/visualize-your-model-s-augmented-predictions)\n\n\n## Examples\n\n![](https://s5.gifyu.com/images/demo_trained_weights_misscls_2.gif)\n\n![](https://s5.gifyu.com/images/demo_trained_weights_misscls_scale_2.gif)",
    "725960": "Wow. Thank you for sharing! :)",
    "725974": "Nice..... Great job....!",
    "725986": "Great\nAmazing!! @pestipeti",
    "726367": "Thanks for sharing!",
    "726538": "Wow, that's awesome! Thanks for sharing!",
    "726667": "Amazing",
    "727161": "pestipeti Amazing!!! one of the best (unique) visualization I have seen !!",
    "728019": "Amazing! Thanks for sharing !",
    "728090": "really helpful, Thanks for sharing",
    "728225": "this is really helpful as an augmentation tips, thank you for sharing!",
    "728728": "666",
    "729670": "Nice.",
    "730424": "Good Try Keep it up",
    "731796": "Awsome❗️ Thanks for sharing😄",
    "1092045": "Great work",
    "1138856": "Looks great!",
    "2852010": "Nice work, Peter! Your script seems both fun and practical. Visualizing predictions can indeed offer deeper insights into model performance. Thanks for sharing the implementation and examples!"
  },
  "source": "meta"
}