May 1 Creating midpoints and visualisation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keys in the new HDF5 file: ['boxes', 'center_coordinates', 'images']\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import h5py\n",
    "\n",
    "# Open your original HDF5 file\n",
    "with h5py.File(\"TrainingData5zeroes midpoints.h5\", 'r') as hdf:\n",
    "    # Read the 'boxes' dataset\n",
    "    boxes = np.array(hdf['boxes'])\n",
    "\n",
    "# Calculate the center coordinates\n",
    "center_coordinates = np.zeros((boxes.shape[0], boxes.shape[1], 3))  # Adjust dimensions as needed\n",
    "center_coordinates[:, :, 0] = boxes[:, :, 0]  # Copy the probabilities\n",
    "center_coordinates[:, :, 1] = (boxes[:, :, 1] + boxes[:, :, 3]) / 2  # Center X\n",
    "center_coordinates[:, :, 2] = (boxes[:, :, 2] + boxes[:, :, 4]) / 2  # Center Y\n",
    "\n",
    "# Optionally, save the center coordinates back to the HDF5 file or a new file\n",
    "with h5py.File(\"TrainingData5zeroes midpoints.h5\", 'a') as hdf:  # 'a' for append, or 'w' to overwrite/create new\n",
    "    # If 'center_coordinates' key already exists, we should remove it first\n",
    "    if 'center_coordinates' in hdf.keys():\n",
    "        del hdf['center_coordinates']\n",
    "    \n",
    "    # Create a new dataset in the file for the center coordinates\n",
    "    hdf.create_dataset('center_coordinates', data=center_coordinates)\n",
    "\n",
    "    # Verify the keys in the file\n",
    "    print(\"Keys in the new HDF5 file:\", list(hdf.keys()))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1., 58., 33., 59., 34.],\n",
       "       [ 0.,  0.,  0.,  0.,  0.],\n",
       "       [ 0.,  0.,  0.,  0.,  0.],\n",
       "       ...,\n",
       "       [ 0.,  0.,  0.,  0.,  0.],\n",
       "       [ 0.,  0.,  0.,  0.,  0.],\n",
       "       [ 0.,  0.,  0.,  0.,  0.]])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " boxes[16]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Images loaded, shape: (10000, 64, 64)\n",
      "Boxes loaded, shape: (10000, 280, 5)\n",
      "Center coordinates loaded, shape: (10000, 280, 3)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import h5py\n",
    "\n",
    "# Path to your HDF5 file\n",
    "file_path = \"TrainingData5zeroes midpoints.h5\"\n",
    "\n",
    "# Open the HDF5 file\n",
    "with h5py.File(file_path, 'r') as hdf:\n",
    "    # Read the 'images' dataset\n",
    "    images = np.array(hdf['images'])\n",
    "    print(\"Images loaded, shape:\", images.shape)\n",
    "    \n",
    "    # Read the 'boxes' dataset\n",
    "    boxes = np.array(hdf['boxes'])\n",
    "    print(\"Boxes loaded, shape:\", boxes.shape)\n",
    "    \n",
    "    # Read the 'center_coordinates' dataset\n",
    "    center_coordinates = np.array(hdf['center_coordinates'])\n",
    "    print(\"Center coordinates loaded, shape:\", center_coordinates.shape)\n",
    "\n",
    "# Now you have the arrays loaded:\n",
    "# 'images', 'boxes', and 'center_coordinates'\n",
    "# You can process or analyze these as needed.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1. , 26. , 48.5],\n",
       "       [ 1. , 53.5, 57. ],\n",
       "       [ 1. , 45.5, 62. ],\n",
       "       [ 1. , 45. , 49.5],\n",
       "       [ 1. ,  9.5, 55.5],\n",
       "       [ 1. , 53.5, 41.5],\n",
       "       [ 1. ,  1. , 15.5],\n",
       "       [ 1. , 60. , 49.5],\n",
       "       [ 1. , 26.5, 23. ],\n",
       "       [ 1. , 22.5, 14. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ],\n",
       "       [ 0. ,  0. ,  0. ]])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "center_coordinates[334]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import h5py\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as patches\n",
    "\n",
    "# Open your HDF5 file to read the data\n",
    "file_path = \"TrainingData5zeroes midpoints.h5\"\n",
    "with h5py.File(file_path, 'r') as hdf:\n",
    "    images = np.array(hdf['images'])  # Assuming images are stored properly and can be visualized directly\n",
    "    boxes = np.array(hdf['boxes'])  # Original bounding boxes\n",
    "\n",
    "# Function to display a single image and its bounding boxes\n",
    "def show_image_with_boxes(image, boxes, ax=None):\n",
    "    if ax is None:\n",
    "        fig, ax = plt.subplots(1)\n",
    "    # Show image\n",
    "    ax.imshow(image)\n",
    "    # Add bounding boxes\n",
    "    for i in range(boxes.shape[0]):\n",
    "        # Only draw if there's a valid box (assuming the first column is the class or probability > 0)\n",
    "        if boxes[i, 0] > 0:\n",
    "            rect = patches.Rectangle((boxes[i, 1], boxes[i, 2]), boxes[i, 3] - boxes[i, 1], boxes[i, 4] - boxes[i, 2],\n",
    "                                     linewidth=1, edgecolor='r', facecolor='none')\n",
    "            ax.add_patch(rect)\n",
    "    plt.show()\n",
    "\n",
    "# Visualize the first few images\n",
    "# for i in range(min(len(images), 15)):  # Change 5 to however many images you want to show\n",
    "show_image_with_boxes(images[334], boxes[334])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import h5py\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as patches\n",
    "\n",
    "# Open your HDF5 file to read the data\n",
    "file_path = \"TrainingData5zeroes midpoints.h5\"\n",
    "with h5py.File(file_path, 'r') as hdf:\n",
    "    images = np.array(hdf['images'])  # Assuming images are stored properly and can be visualized directly\n",
    "    center_coordinates = np.array(hdf['center_coordinates'])  # Center coordinates\n",
    "\n",
    "# Function to display a single image and its bounding box centers\n",
    "def show_image_with_centers(image, centers, ax=None):\n",
    "    if ax is None:\n",
    "        fig, ax = plt.subplots(1)\n",
    "    # Show image\n",
    "    ax.imshow(image)\n",
    "    # Add center points\n",
    "    for i in range(centers.shape[0]):\n",
    "        # Only plot if the probability (or class) is greater than 0\n",
    "        if centers[i, 0] > 0:\n",
    "            ax.plot(centers[i, 1], centers[i, 2], 'rd')  # 'ro' stands for red circle\n",
    "    plt.show()\n",
    "\n",
    "# # Visualize the first few images\n",
    "# for i in range(min(len(images), 15)):  # Change 5 to however many images you want to show\n",
    "#     show_image_with_centers(images[i], center_coordinates[i])\n",
    "show_image_with_centers(images[334], center_coordinates[334])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import h5py\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Open your HDF5 file to read the data\n",
    "file_path = \"TrainingData5zeroes midpoints.h5\"\n",
    "with h5py.File(file_path, 'r') as hdf:\n",
    "    images = np.array(hdf['images'])  # Assuming images are stored properly and can be visualized directly\n",
    "    center_coordinates = np.array(hdf['center_coordinates'])  # Center coordinates\n",
    "\n",
    "# Function to display a single image and its bounding box centers\n",
    "def show_image_with_centers(image, centers, ax=None):\n",
    "    if ax is None:\n",
    "        fig, ax = plt.subplots(1)\n",
    "    # Show image\n",
    "    ax.imshow(image)\n",
    "    # Add center points\n",
    "    for i in range(centers.shape[0]):\n",
    "        # Only plot if the probability (or class) is greater than 0\n",
    "        if centers[i, 0] > 0:\n",
    "            ax.plot(centers[i, 1], centers[i, 2], 'r.', markersize=2)  # 'r.' stands for red dot, markersize controls the size\n",
    "    plt.show()\n",
    "\n",
    "# Visualize the first few images\n",
    "show_image_with_centers(images[334], center_coordinates[334])\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tiny_yolo",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}