{
  "fetched_at": "2026-10-19T08:56:56Z",
  "source_url": "https://www.kaggle.com/competitions/counting-cells-in-microscopy-images-2023/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The training data consists of 2000 microscopy images of cells, stored in a numpy array X, and corresponding segmentation masks stored in a numpy array y. Each image (and each corresponding mask) has shape (128, 128). The segmentation mask y[i] for image X[i] contains \"soft labels\" (that is, numbers between 0 and 1) which tell you which pixels in the image X[i] belong to a cell nucleus.\n\nThe test data consists of 2000 more images of cells. No masks are provided for the test images. Your goal is to count how many cells appear in each test image.\n\nThe training data and test images are stored in numpy's .npz file format. You can load the training data using the following Python commands:\n\n`import numpy as np`\n\n`npzfile = np.load('train_data.npz')`\n\n`X = npzfile['X']`\n\n`y = npzfile['y']`\n\nThen, you can display an image with superimposed labels using the following Python commands:\n\n`import matplotlib.pyplot as plt`\n\n`i = 0 # change this value to view different images`\n\n`plt.imshow(X[i]/255 + .5*y[i], cmap='gray')`\n\n## Files\n\n*   **test_images.npz** - Contains the test images, described above\n*   **train_data.npz** - Contains the training data, described above\n*   **sample_submission.csv** - a sample submission file in the correct format\n\n## Columns\nYour submission file should be in .csv format with the following two columns:\n\n*   `index` - An integer between 0 and 1999. Index i corresponds to the test image X[i].\n*   `count` - Your predicted cell counts for each test image.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "How many cells appear in a given microscopy image?\nYour goal is to develop an algorithm which is capable of answering that question.\n\n## Acknowledgements\n\nThe data for this competition was collected in the Schroeder lab at the University of San Francisco. The creation of this lab was funded by a donation from Kirk Schroeder: https://www.usfca.edu/news/gift-expands-biotech",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The evaluation metric for this competition is mean absolute error.\n\n\n## Submission Format\n\nYour submission file should be in .csv format with two columns: \"index\" and \"count\". Each index is an integer between 0 and 1999 which specifies a particular test image.\n\nThe \"count\" column contains your predictions for the number of cells which appear in each test image.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "You are allowed to discuss the contest with classmates, but each student must write their own code and participate individually in this contest.\n\nSimply copying a u-net implementation from the internet is considered academic dishonesty. You can get help, but you must write your own code based on your own understanding.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}