FREE: YOLOv7 Custom Object Detection Course

Explore deep learning basics, neural networks, and perceptrons, with emphasis on activation functions and gradient issues, and how ResNet mitigates vanishing and exploding gradients.

Demonstrate how a convolutional network processes a cross image through convolution, activation, pooling, and flattening. See how the fully connected layer and softmax on two outputs determine cross class.

Install Anaconda by downloading the installer, running it, and selecting a user-only setup with default destination. Verify the install via the start menu, checking Anaconda Navigator and Anaconda prompt.

Install git on Windows via the standalone installer, accept defaults, and verify installation with git --version; git enables cloning the YOLO repository.

Learn how supervised learning uses labeled, relevant data for object detection with YOLO, and explore finding datasets via open sources like Kaggle, including usability, ratings, and licenses.

Understand the YOLO dataset annotation format for supervised object detection, with class ids and bounding boxes defined by normalized midpoint x, y and width, height relative to image dimensions.

Set up labelImg on Windows with anaconda navigator and python 3.9, then annotate a face mask dataset with bounding boxes for classes mask, no mask, bed mask in YOLO format.

Split your dataset into training and test sets to train and evaluate a YOLOv7 model, tune hyperparameters, and use a validation set to prevent overfitting.

Learn how to install Nvidia GeForce drivers on Windows 11 for a GeForce GTX 1650 Ti, selecting GeForce 16 series game-ready drivers version 517.48, then verify installation via command prompt.

Learn to install cuda on windows, check gpu compatibility with nvidia smi, download and extract the 11.7 installer, and verify installation with nvcc.

Install cuDNN on Windows by signing into Nvidia, choosing a cuDNN version compatible with your CUDA, downloading the installer, and copying bin, include, and lib files to CUDA toolkit directory.

Learn to install Yolov7 cpu on Windows by creating a conda env, cloning the repo, installing requirements, downloading weights, and running detect.py on sample images for object detection.

Learn to use Yolov7 for object detection across images, videos, and webcams, adjust confidence threshold and inference size, and save results to a file.

activate the YOLOv7 gpu environment and run the detect.py script to perform image object detection, using 0.5 threshold, 640 image size, and saving labeled results and bounding box data.

Prepare a face mask dataset for Yolov7 on Windows, extract it to the data folder, and split into train, validation, and test folders using the split_dataset.py 80/10/10.

Train yolov7 on custom objects using transfer learning with pre-trained weights, configure dataset and training options, and visualize the training progress with TensorBoard's MLP/mean average precision graph.

Detect face masks using the best yolov7 weights with the highest mlp value, applying inference on image, video, and webcam, with results saved to the inference folder.

Master object detection with yolov7, explore pose estimation and instance segmentation, and learn about newer versions like yolov8, v9, and v2 in this four-in-one course.