245 lines
758 KiB
Plaintext
245 lines
758 KiB
Plaintext
|
{
|
||
|
|
"cells": [
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 1,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"import ultralytics\n",
|
||
|
|
"from ultralytics import YOLO\n"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 39,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"from PIL import Image\n",
|
||
|
|
"from IPython.display import display\n",
|
||
|
|
"import cv2\n",
|
||
|
|
"import numpy as np"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 2,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"name": "stdout",
|
||
|
|
"output_type": "stream",
|
||
|
|
"text": [
|
||
|
|
"Ultralytics YOLOv8.0.215 🚀 Python-3.9.18 torch-2.1.1+cpu CPU (11th Gen Intel Core(TM) i7-11800H 2.30GHz)\n",
|
||
|
|
"Setup complete ✅ (16 CPUs, 63.7 GB RAM, 14.9/1907.7 GB disk)\n"
|
||
|
|
]
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"ultralytics.checks()"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 3,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"model = YOLO(\"best.pt\")"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"Rozwiązanie bazowe na fragmencie ze zbioru testowego"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 4,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"file_in = \"IMG_0983\""
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 5,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"name": "stdout",
|
||
|
|
"output_type": "stream",
|
||
|
|
"text": [
|
||
|
|
"\n",
|
||
|
|
"image 1/1 e:\\gonito_tomato\\dev-0\\in\\IMG_0983.jpg: 640x480 4 b_greens, 539.1ms\n",
|
||
|
|
"Speed: 3.0ms preprocess, 539.1ms inference, 1.0ms postprocess per image at shape (1, 3, 640, 480)\n"
|
||
|
|
]
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"img = \"dev-0/in/\" + file_in + \".jpg\"\n",
|
||
|
|
"pred_image = model.predict(conf=0.5, source=img)\n",
|
||
|
|
"result = pred_image[0]"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 6,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"image = cv2.imread(img)\n",
|
||
|
|
"# Convert image to uint8 data type\n",
|
||
|
|
"image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
|
||
|
|
"image = np.array(image, dtype=np.uint8)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 7,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"for box in result.boxes:\n",
|
||
|
|
" cords =box.xyxy[0].tolist()\n",
|
||
|
|
" class_id =box.cls[0].item()\n",
|
||
|
|
" conf = box.conf[0].item()\n",
|
||
|
|
" start = (int(cords[0]),int(cords[1])) # x0, y0\n",
|
||
|
|
" end = (int(cords[2]),int(cords[3])) # x1, y1\n",
|
||
|
|
" text = result.names[class_id]\n",
|
||
|
|
" text_size = cv2.getTextSize(text, cv2.FONT_HERSHEY_DUPLEX, 1.9, 2)[0]\n",
|
||
|
|
" rectangle_position = (start[0], start[1] - text_size[1] - 10)\n",
|
||
|
|
" rectangle_size = (text_size[0] + 30, text_size[1] + 20)\n",
|
||
|
|
" cv2.rectangle(image, rectangle_position, (rectangle_position[0] + rectangle_size[0], rectangle_position[1] + rectangle_size[1]), (255, 255, 255), -1)\n",
|
||
|
|
"\n",
|
||
|
|
" # Draw the bounding box\n",
|
||
|
|
" cv2.rectangle(image, start, end, (0, 200, 0), thickness=2)\n",
|
||
|
|
"\n",
|
||
|
|
" # Put the text on top of the highlighted background\n",
|
||
|
|
" cv2.putText(image, text, (start[0] + 15, start[1] - 5), cv2.FONT_HERSHEY_DUPLEX, 1.9, (10, 0, 10), 2) # show detection objects on image# show detection objects on image"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 8,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"tensor([2., 2., 2., 2.])"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 8,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"result.boxes.cls"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 14,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/jpeg": "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
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<PIL.Image.Image image mode=RGB size=450x600>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"\n",
|
||
|
|
"picture = Image.fromarray(image)\n",
|
||
|
|
"desired_size = (700, 600)\n",
|
||
|
|
"picture.thumbnail(desired_size)\n",
|
||
|
|
"# Display the image\n",
|
||
|
|
"display(picture)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"Metric for GONITO - mean average precision"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 40,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"name": "stdout",
|
||
|
|
"output_type": "stream",
|
||
|
|
"text": [
|
||
|
|
"Ultralytics YOLOv8.0.215 🚀 Python-3.9.18 torch-2.1.1+cpu CPU (11th Gen Intel Core(TM) i7-11800H 2.30GHz)\n"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"name": "stderr",
|
||
|
|
"output_type": "stream",
|
||
|
|
"text": [
|
||
|
|
"\u001b[34m\u001b[1mval: \u001b[0mScanning E:\\tomAIto\\model\\datasets\\val\\labels.cache... 161 images, 0 backgrounds, 0 corrupt: 100%|██████████| 161/161 [00:00<?, ?it/s]\n",
|
||
|
|
" Class Images Instances Box(P R mAP50 mAP50-95): 100%|██████████| 11/11 [02:00<00:00, 10.93s/it]\n"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"name": "stdout",
|
||
|
|
"output_type": "stream",
|
||
|
|
"text": [
|
||
|
|
" all 161 1996 0.814 0.835 0.88 0.786\n",
|
||
|
|
" b_fully_ripened 161 72 0.799 0.778 0.854 0.788\n",
|
||
|
|
" b_half_ripened 161 116 0.78 0.819 0.844 0.765\n",
|
||
|
|
" b_green 161 387 0.861 0.928 0.956 0.847\n",
|
||
|
|
" l_fully_ripened 161 269 0.803 0.835 0.878 0.79\n",
|
||
|
|
" l_half_ripened 161 223 0.758 0.776 0.819 0.735\n",
|
||
|
|
" l_green 161 929 0.881 0.873 0.928 0.791\n",
|
||
|
|
"Speed: 2.0ms preprocess, 629.8ms inference, 0.0ms loss, 0.4ms postprocess per image\n",
|
||
|
|
"Results saved to \u001b[1mruns\\detect\\val6\u001b[0m\n",
|
||
|
|
"0.786025264281105\n"
|
||
|
|
]
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"metrics = model.val(data='dataset.yaml')\n",
|
||
|
|
"mAP = metrics.box.ap.mean()\n",
|
||
|
|
"print(mAP)"
|
||
|
|
]
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"metadata": {
|
||
|
|
"kernelspec": {
|
||
|
|
"display_name": "base",
|
||
|
|
"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": "3.9.18"
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"nbformat": 4,
|
||
|
|
"nbformat_minor": 2
|
||
|
|
}
|