{ "id": 253019, "title": "Constant Q Transformed Dataset", "url": "/competitions/g2net-gravitational-wave-detection/discussion/253019", "author_name": "Ayush Thakur", "post_date": "2021-07-14T17:21:32.432000", "votes": 3, "comment_count": 3, "views": 0, "content": "

Hey all, I have created a dataset that might be useful for training computer vision-based models.

\n

Find the Dataset here

\n
    \n

  1. The kernel, Create Constant Q Transformed Dataset was used to create the dataset. It uses multiprocessing to utilize all the available CPU cores.

    2. \n

  2. You can use this kernel, Visualize Dataset Interactively using W&B Tables to log a sample of the dataset to W&B Tables and visualize it.

    3. \n
\n

Visualize the dataset interactively using W&B Tables

\n

\"img\"

\n

I have logged only 0.01% of the training data. It was randomly sampled. Given that the dataset is balanced the resulting sampled dataset is balanced as well. Using W&B Tables you can:

\n", "messages": [ { "id": 1388153, "postDate": "2021-07-14T17:21:32.433Z", "content": "

Hey all, I have created a dataset that might be useful for training computer vision-based models.

\n

Find the Dataset here

\n
    \n

  1. The kernel, Create Constant Q Transformed Dataset was used to create the dataset. It uses multiprocessing to utilize all the available CPU cores.

    2. \n

  2. You can use this kernel, Visualize Dataset Interactively using W&B Tables to log a sample of the dataset to W&B Tables and visualize it.

    3. \n
\n

Visualize the dataset interactively using W&B Tables

\n

\"img\"

\n

I have logged only 0.01% of the training data. It was randomly sampled. Given that the dataset is balanced the resulting sampled dataset is balanced as well. Using W&B Tables you can:

\n", "rawMarkdown": "Hey all, I have created a dataset that might be useful for training computer vision-based models. \n\n### [Find the Dataset here](http://wandb.me/g2net-cqt)\n\n1. The kernel, [Create Constant Q Transformed Dataset](http://wandb.me/cqt-create-dataset) was used to create the dataset. It uses multiprocessing to utilize all the available CPU cores.\n\n2. You can use this kernel, [Visualize Dataset Interactively using W&B Tables](http://wandb.me/g2net-log-cqt) to log a sample of the dataset to W&B Tables and visualize it. \n\n\n### [Visualize the dataset interactively using W&B Tables](http://wandb.me/cqt-wandb-table-viz)\n\n![img](https://i.imgur.com/329cY0c.gif)\n\nI have logged only 0.01% of the training data. It was randomly sampled. Given that the dataset is balanced the resulting sampled dataset is balanced as well. Using W&B Tables you can:\n\n* Visualize individual examples interactively.\n* Group by some column - here I have grouped by `ground-truth` and that gives the distribution of labels. \n* There are a lot of interesting insights you can derive from this tool. You can check out this [kernel](http://wandb.me/better-tables) to learn more. \n", "votes": 3 }, { "id": 1561275, "postDate": "2021-10-27T13:11:39.823Z", "content": "

Hey All,

\n

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 https://forms.gle/QP9L16niPexozyhu5.

\n

Thank you all,

\n

Regards,
\nChris

", "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" }, { "id": 1392897, "postDate": "2021-07-19T07:38:26.870Z", "content": "

Quite an interesting piece of work. Thanks for sharing!!!👍

", "rawMarkdown": "Quite an interesting piece of work. Thanks for sharing!!!👍" }, { "id": 1394365, "postDate": "2021-07-20T10:15:58.933Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 1561275, "author_name": "ChristopherZerafa", "author_url": "", "post_date": "2021-10-27T13:11:39.823000", "content": "

Hey All,

\n

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 https://forms.gle/QP9L16niPexozyhu5.

\n

Thank you all,

\n

Regards,
\nChris

", "votes": 0, "replies": [] }, { "id": 1392897, "author_name": "Ankit", "author_url": "", "post_date": "2021-07-19T07:38:26.870000", "content": "

Quite an interesting piece of work. Thanks for sharing!!!👍

", "votes": 0, "replies": [] }, { "id": 1394365, "author_name": "", "author_url": "", "post_date": "2021-07-20T10:15:58.933000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1388153": "Hey all, I have created a dataset that might be useful for training computer vision-based models. \n\n### [Find the Dataset here](http://wandb.me/g2net-cqt)\n\n1. The kernel, [Create Constant Q Transformed Dataset](http://wandb.me/cqt-create-dataset) was used to create the dataset. It uses multiprocessing to utilize all the available CPU cores.\n\n2. You can use this kernel, [Visualize Dataset Interactively using W&B Tables](http://wandb.me/g2net-log-cqt) to log a sample of the dataset to W&B Tables and visualize it. \n\n\n### [Visualize the dataset interactively using W&B Tables](http://wandb.me/cqt-wandb-table-viz)\n\n![img](https://i.imgur.com/329cY0c.gif)\n\nI have logged only 0.01% of the training data. It was randomly sampled. Given that the dataset is balanced the resulting sampled dataset is balanced as well. Using W&B Tables you can:\n\n* Visualize individual examples interactively.\n* Group by some column - here I have grouped by `ground-truth` and that gives the distribution of labels. \n* There are a lot of interesting insights you can derive from this tool. You can check out this [kernel](http://wandb.me/better-tables) to learn more. \n", "1561275": "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", "1392897": "Quite an interesting piece of work. Thanks for sharing!!!👍", "1394365": "" } }