Course Objective & Scope
Off-the-shelf models rarely deliver enough accuracy on company images; you need to train or fine-tune on your own data. Starting from the logic of convolution and pooling, the programme examines architectures such as ResNet, EfficientNet and ConvNeXt in PyTorch.
Transfer learning, data augmentation (Albumentations), learning rate scheduling, mixed precision training and fighting overfitting are covered in hands-on labs. An end-to-end classification model is trained on a dataset similar to the organisation's data, errors are analysed and model decisions are explained with Grad-CAM.
Who Should Attend?
- Software and AI engineers who will build image models
- Data scientists applying deep learning to image data
- Technical leads managing computer vision teams
- Prerequisite: Python and basic machine learning concepts
Key Takeaways
- Write a CNN training and validation loop in PyTorch
- Reach high accuracy with little data using transfer learning
- Reduce overfitting with augmentation and regularisation
- Analyse models through confusion matrices and error samples
- Visualise model decisions with Grad-CAM
Curriculum
Convolution, Pooling and CNN Building Blocks
- Filters, stride, padding and receptive field
- Batch normalisation and activation functions
Modern CNN Architectures
- Comparing ResNet, EfficientNet and ConvNeXt
- Model selection with the timm library
Training Loops in PyTorch
- Dataset, DataLoader and choosing loss/optimiser
- Learning rate scheduling and mixed precision
Transfer Learning and Augmentation
- Freezing layers and gradual fine-tuning
- Augmentation strategies with Albumentations
Evaluation, Error Analysis and Explainability
- Confusion matrix, class imbalance and threshold selection
- Visualising decisions with Grad-CAM
Other Courses in This Category
-
Self-supervised learning (DINOv2), vision foundation models (SAM 2, Grounding DINO), synthetic data, active learning and large-scale training, for experienced computer vision engineers tackling hard production problems.
- Live Course
- 5 modules
-
Object detection and instance segmentation with YOLO, RT-DETR and Mask R-CNN: labelling, training, mAP-based evaluation and tracking integration, for teams building an end-to-end detection/segmentation prototype.
- Live Course
- 5 modules
-
Computer Vision Bootcamp
Comprehensive Computer Vision Bootcamp ranging from classical image processing techniques to PyTorch-based CNN, Vision Transformer (ViT), Swin Transformer, and ConvNeXt architectures. Hands-on projects cover object detection with YOLO and DETR, segmentation with SAM and U-Net, video analysis, 3D vision (NeRF, Gaussian Splatting), Diffusion models, and model deployment with ONNX/TensorRT.
- Live Course
- 13 modules
-
Semantic, instance and panoptic segmentation: pixel-level training with U-Net, DeepLabV3+, SegFormer, Mask2Former and SAM 2, Dice/mIoU evaluation and mask-based measurement, for industrial, medical and satellite imagery teams.
- Live Course
- 6 modules
-
ViT, DeiT, Swin and DINOv2 architectures: patch embedding, self-attention, comparison with CNNs and fine-tuning with Hugging Face, for engineers selecting and adapting transformer-based models for vision tasks.
- Live Course
- 5 modules
-
Document preprocessing, text recognition, layout analysis and table/form extraction with PaddleOCR, Surya, LayoutLMv3 and Donut, for technical teams turning invoices, contracts and operational documents into structured data.
- Live Course
- 5 modules
Get in touch
