Add tutorial for Indian Currency Note Detection using YOLOv5

beginner_source/indian_notes_detection_tutorial.py

@@ -0,0 +1,75 @@
+"""
+Title: Indian Currency Note Detection using YOLOv5
+Author: Your Name
+Description:
+    This tutorial demonstrates how to detect Indian currency notes using YOLOv5 and Roboflow.
+    It covers dataset download, training, inference, and retraining for improved accuracy.
+
+    Requirements:
+        - Roboflow account with access to the dataset.
+        - Google Colab or local environment with GPU.
+        - Replace 'YOUR_API_KEY_HERE' with your Roboflow API key.
+"""
+
+import os
+
+# Step 1: Clone YOLOv5 repository and install requirements
+print("Cloning YOLOv5 and installing dependencies...")
+os.system("git clone https://github.com/ultralytics/yolov5")
+os.chdir("yolov5")
+os.system("pip install -r requirements.txt")
+os.system("pip install roboflow")
+
+# Step 2: Download Dataset from Roboflow
+from roboflow import Roboflow
+
+print("Downloading dataset from Roboflow...")
+rf = Roboflow(api_key="YOUR_API_KEY_HERE")  # Replace with your Roboflow API key
+project = rf.workspace("omkar-patkar-fes59").project("indian-currency-notes")
+version = project.version(4)
+dataset = version.download("yolov5")
+
+# Step 3: Explore dataset structure
+print("\n📂 Dataset location:", dataset.location)
+for root, dirs, files in os.walk(dataset.location):
+    print(f"📁 {root}")
+    for file in files[:5]:
+        print("   📄", file)
+    break
+
+# Step 4: Train the model (Initial training)
+print("\n🧠 Training YOLOv5 model...")
+os.system("""
+python train.py \
+  --img 640 \
+  --batch 16 \
+  --epochs 30 \
+  --data indian-currency/notes-4/data.yaml \
+  --weights yolov5s.pt \
+  --project currency-project \
+  --name yolo_currency \
+  --cache
+""")
+
+# Step 5: Run inference on test images
+print("\n🔍 Running detection on test images...")
+os.system("""
+python detect.py \
+  --weights currency-project/yolo_currency/weights/best.pt \
+  --img 640 \
+  --conf 0.25 \
+  --source indian-currency/notes-4/test/images
+""")
+
+# Step 6: Retrain with more epochs (optional)
+print("\n📈 Retraining with 50 epochs for improved accuracy...")
+os.system("""
+python train.py \
+  --img 640 \
+  --batch 16 \
+  --epochs 50 \
+  --data indian-currency/notes-4/data.yaml \
+  --weights yolov5s.pt \
+  --project currency-project/yolo_currency_v2 \
+  --name improved_run
+""")

index.rst

@@ -85,6 +85,14 @@ Welcome to PyTorch Tutorials
    :link: beginner/nn_tutorial.html
    :tags: Getting-Started
 
+.. customcarditem::
+   :header: Indian Currency Note Detection with YOLOv5
+   :card_description: Learn to train and run a YOLOv5 model to detect Indian currency notes using Roboflow and PyTorch.
+   :image: _static/img/thumbnails/cropped/indian-currency-yolo.png
+   :link: beginner/indian_notes_detection_tutorial.html
+   :tags: Image/Video,Object-Detection
+
+
 .. customcarditem::
    :header: Visualizing Models, Data, and Training with TensorBoard
    :card_description: Learn to use TensorBoard to visualize data and model training.
@@ -783,6 +791,9 @@ Welcome to PyTorch Tutorials
    :tags: TorchRec,Recommender
 
 
+
+
+
 .. End of tutorial card section
 
 .. raw:: html
@@ -832,6 +843,9 @@ Additional Resources
    :button_link: beginner/colab.html
    :button_text: Open
 
+
+
+
 .. End of callout section
 
 .. raw:: html