{
  "id": 290956,
  "title": "TensorFlow Object Detection API",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/290956",
  "author_name": "",
  "post_date": "2021-11-27T01:48:05.421396Z",
  "votes": 15,
  "comment_count": 1,
  "views": 0,
  "content": "<p>If you’re trying to get started with object detection, the tensorflow object detection API is a great place to start, especially since tensorflow is the sponsor of this competition.</p>\n<p><strong>How to use it:</strong></p>\n<ol>\n<li>Install the tensorflow object detection api</li>\n<li>Create a dataset folder</li>\n<li>Prepare the train and valid image sets in tfrecord format and put them in the dataset folder</li>\n<li>Create a label file in the dataset folder</li>\n<li>Select and download a model from the tensorflow 2 detection model zoo</li>\n<li>Create a config file in the dataset folder using a template</li>\n<li>Train and evaluate the model with the model_main_tf2.py file</li>\n<li>Save the model with the exporter_main_v2.py file</li>\n<li>Make your predictions</li>\n</ol>\n<p>To understand how to perform any of these steps, check out the resources below</p>\n<p><strong>Resources:</strong><br>\n<a href=\"https://github.com/tensorflow/models/tree/master/research/object_detection\" target=\"_blank\">Tensorflow object detection GitHub repo</a><br>\n<a href=\"https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md\" target=\"_blank\">Model Zoo</a><br>\n<a href=\"https://github.com/tensorflow/models/tree/master/research/object_detection/configs/tf2\" target=\"_blank\">Config starters</a><br>\n<a href=\"https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html\" target=\"_blank\">Documentation Tutorial</a><br>\n<a href=\"https://www.kaggle.com/khanhlvg/cots-detection-w-tensorflow-object-detection-api\" target=\"_blank\">Notebook example (training) by Khanh</a><br>\n<a href=\"https://www.kaggle.com/khanhlvg/inference-using-efficientdet-d0-model-tensorflow\" target=\"_blank\">Notebook example (inference) by Khanh</a></p>",
  "messages": [
    {
      "id": "1596911",
      "postDate": "11/27/2021 01:48:05",
      "content": "<p>If you’re trying to get started with object detection, the tensorflow object detection API is a great place to start, especially since tensorflow is the sponsor of this competition.</p>\n<p><strong>How to use it:</strong></p>\n<ol>\n<li>Install the tensorflow object detection api</li>\n<li>Create a dataset folder</li>\n<li>Prepare the train and valid image sets in tfrecord format and put them in the dataset folder</li>\n<li>Create a label file in the dataset folder</li>\n<li>Select and download a model from the tensorflow 2 detection model zoo</li>\n<li>Create a config file in the dataset folder using a template</li>\n<li>Train and evaluate the model with the model_main_tf2.py file</li>\n<li>Save the model with the exporter_main_v2.py file</li>\n<li>Make your predictions</li>\n</ol>\n<p>To understand how to perform any of these steps, check out the resources below</p>\n<p><strong>Resources:</strong><br>\n<a href=\"https://github.com/tensorflow/models/tree/master/research/object_detection\" target=\"_blank\">Tensorflow object detection GitHub repo</a><br>\n<a href=\"https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md\" target=\"_blank\">Model Zoo</a><br>\n<a href=\"https://github.com/tensorflow/models/tree/master/research/object_detection/configs/tf2\" target=\"_blank\">Config starters</a><br>\n<a href=\"https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html\" target=\"_blank\">Documentation Tutorial</a><br>\n<a href=\"https://www.kaggle.com/khanhlvg/cots-detection-w-tensorflow-object-detection-api\" target=\"_blank\">Notebook example (training) by Khanh</a><br>\n<a href=\"https://www.kaggle.com/khanhlvg/inference-using-efficientdet-d0-model-tensorflow\" target=\"_blank\">Notebook example (inference) by Khanh</a></p>",
      "rawMarkdown": "If you’re trying to get started with object detection, the tensorflow object detection API is a great place to start, especially since tensorflow is the sponsor of this competition.\n\n**How to use it:**\n1. Install the tensorflow object detection api\n2. Create a dataset folder\n3. Prepare the train and valid image sets in tfrecord format and put them in the dataset folder\n4. Create a label file in the dataset folder\n5. Select and download a model from the tensorflow 2 detection model zoo\n6. Create a config file in the dataset folder using a template\n7. Train and evaluate the model with the model_main_tf2.py file\n8. Save the model with the exporter_main_v2.py file\n9. Make your predictions\n\nTo understand how to perform any of these steps, check out the resources below\n\n**Resources:**\n[Tensorflow object detection GitHub repo](https://github.com/tensorflow/models/tree/master/research/object_detection)\n[Model Zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md)\n[Config starters](https://github.com/tensorflow/models/tree/master/research/object_detection/configs/tf2)\n[Documentation Tutorial](https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html)\n[Notebook example (training) by Khanh](https://www.kaggle.com/khanhlvg/cots-detection-w-tensorflow-object-detection-api)\n[Notebook example (inference) by Khanh](https://www.kaggle.com/khanhlvg/inference-using-efficientdet-d0-model-tensorflow)",
      "votes": null
    },
    {
      "id": "2087121",
      "postDate": "01/05/2023 10:48:18",
      "content": "<p>Install the TensorFlow object detection API?<br>\ndid you do that?//<br>\nany reference notebook who has done that???</p>",
      "rawMarkdown": "Install the TensorFlow object detection API?\ndid you do that?//\nany reference notebook who has done that???",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2087121,
      "author_name": "prashanthsheri",
      "author_url": "",
      "post_date": "01/05/2023 10:48:18",
      "content": "<p>Install the TensorFlow object detection API?<br>\ndid you do that?//<br>\nany reference notebook who has done that???</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1596911": "If you’re trying to get started with object detection, the tensorflow object detection API is a great place to start, especially since tensorflow is the sponsor of this competition.\n\n**How to use it:**\n1. Install the tensorflow object detection api\n2. Create a dataset folder\n3. Prepare the train and valid image sets in tfrecord format and put them in the dataset folder\n4. Create a label file in the dataset folder\n5. Select and download a model from the tensorflow 2 detection model zoo\n6. Create a config file in the dataset folder using a template\n7. Train and evaluate the model with the model_main_tf2.py file\n8. Save the model with the exporter_main_v2.py file\n9. Make your predictions\n\nTo understand how to perform any of these steps, check out the resources below\n\n**Resources:**\n[Tensorflow object detection GitHub repo](https://github.com/tensorflow/models/tree/master/research/object_detection)\n[Model Zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md)\n[Config starters](https://github.com/tensorflow/models/tree/master/research/object_detection/configs/tf2)\n[Documentation Tutorial](https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html)\n[Notebook example (training) by Khanh](https://www.kaggle.com/khanhlvg/cots-detection-w-tensorflow-object-detection-api)\n[Notebook example (inference) by Khanh](https://www.kaggle.com/khanhlvg/inference-using-efficientdet-d0-model-tensorflow)",
    "2087121": "Install the TensorFlow object detection API?\ndid you do that?//\nany reference notebook who has done that???"
  },
  "source": "meta"
}