{
  "id": 257335,
  "title": " Useful Optimizers",
  "url": "/competitions/g2net-gravitational-wave-detection/discussion/257335",
  "author_name": "",
  "post_date": "2021-08-03T01:12:09.023792100Z",
  "votes": 6,
  "comment_count": 4,
  "views": 0,
  "content": "<p><a href=\"https://github.com/takfarine/Optimizers\" target=\"_blank\">https://github.com/takfarine/Optimizers</a></p>",
  "messages": [
    {
      "id": "1418238",
      "postDate": "08/03/2021 01:12:09",
      "content": "<p><a href=\"https://github.com/takfarine/Optimizers\" target=\"_blank\">https://github.com/takfarine/Optimizers</a></p>",
      "rawMarkdown": "https://github.com/takfarine/Optimizers",
      "votes": null
    },
    {
      "id": "1418735",
      "postDate": "08/03/2021 01:45:14",
      "content": "<p>Also you can explore <a href=\"https://github.com/jettify/pytorch-optimizer\" target=\"_blank\">torch_optimizer library</a>. Despite almost any of us uses the same optimizer nearly all the time, this repo provides a really nice comparison and visualization of a variety of optimizers (though, the authors could run their experiments several times to make the results more statistically significant).<br>\n<img src=\"https://raw.githubusercontent.com/jettify/pytorch-optimizer/master/docs/rastrigin_Adam.png\" alt=\"\"></p>",
      "rawMarkdown": "Also you can explore [torch_optimizer library](https://github.com/jettify/pytorch-optimizer). Despite almost any of us uses the same optimizer nearly all the time, this repo provides a really nice comparison and visualization of a variety of optimizers (though, the authors could run their experiments several times to make the results more statistically significant).\n![](https://raw.githubusercontent.com/jettify/pytorch-optimizer/master/docs/rastrigin_Adam.png)",
      "votes": null
    },
    {
      "id": "1426331",
      "postDate": "08/03/2021 07:28:26",
      "content": "<p>great resource, thanks for sharing.</p>",
      "rawMarkdown": "great resource, thanks for sharing.",
      "votes": null
    },
    {
      "id": "1448249",
      "postDate": "08/04/2021 17:03:26",
      "content": "<p>Thanks for sharing 😊👏. Although I've attained my current local leaderboard using <code>Adam</code>, I'll give <strong><code>AdamW</code></strong> a try as well.</p>",
      "rawMarkdown": "Thanks for sharing 😊👏. Although I've attained my current local leaderboard using `Adam`, I'll give **`AdamW`** a try as well.",
      "votes": null
    },
    {
      "id": "1561299",
      "postDate": "10/27/2021 13:13:10",
      "content": "<p>Hey All,</p>\n<p>Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey <a href=\"https://forms.gle/QP9L16niPexozyhu5\" target=\"_blank\">https://forms.gle/QP9L16niPexozyhu5</a>.</p>\n<p>Thank you all,</p>\n<p>Regards,<br>\nChris</p>",
      "rawMarkdown": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1418735,
      "author_name": "iafoss",
      "author_url": "",
      "post_date": "08/03/2021 01:45:14",
      "content": "<p>Also you can explore <a href=\"https://github.com/jettify/pytorch-optimizer\" target=\"_blank\">torch_optimizer library</a>. Despite almost any of us uses the same optimizer nearly all the time, this repo provides a really nice comparison and visualization of a variety of optimizers (though, the authors could run their experiments several times to make the results more statistically significant).<br>\n<img src=\"https://raw.githubusercontent.com/jettify/pytorch-optimizer/master/docs/rastrigin_Adam.png\" alt=\"\"></p>",
      "votes": null,
      "replies": [
        {
          "id": 1426331,
          "author_name": "vonneumann",
          "author_url": "",
          "post_date": "08/03/2021 07:28:26",
          "content": "<p>great resource, thanks for sharing.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1448249,
      "author_name": "sauravmaheshkar",
      "author_url": "",
      "post_date": "08/04/2021 17:03:26",
      "content": "<p>Thanks for sharing 😊👏. Although I've attained my current local leaderboard using <code>Adam</code>, I'll give <strong><code>AdamW</code></strong> a try as well.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1561299,
      "author_name": "zerafachris",
      "author_url": "",
      "post_date": "10/27/2021 13:13:10",
      "content": "<p>Hey All,</p>\n<p>Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey <a href=\"https://forms.gle/QP9L16niPexozyhu5\" target=\"_blank\">https://forms.gle/QP9L16niPexozyhu5</a>.</p>\n<p>Thank you all,</p>\n<p>Regards,<br>\nChris</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1418238": "https://github.com/takfarine/Optimizers",
    "1418735": "Also you can explore [torch_optimizer library](https://github.com/jettify/pytorch-optimizer). Despite almost any of us uses the same optimizer nearly all the time, this repo provides a really nice comparison and visualization of a variety of optimizers (though, the authors could run their experiments several times to make the results more statistically significant).\n![](https://raw.githubusercontent.com/jettify/pytorch-optimizer/master/docs/rastrigin_Adam.png)",
    "1426331": "great resource, thanks for sharing.",
    "1448249": "Thanks for sharing 😊👏. Although I've attained my current local leaderboard using `Adam`, I'll give **`AdamW`** a try as well.",
    "1561299": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris"
  },
  "source": "meta"
}