A suite of nine applied Coursera project certificates completed across PyTorch and TensorFlow, spanning core deep-learning primitives (CNNs, generative models, metric learning) and applied problem domains (medical imaging, NLP, interpretability, class-imbalance).
Projects completed
| # | Project | Framework | Verify · PDF |
|---|---|---|---|
| 1 | Deep Learning with PyTorch: Siamese Network | PyTorch | verify · pdf |
| 2 | Deep Learning with PyTorch: Generative Adversarial Networks | PyTorch | verify · pdf |
| 3 | Deep Learning with PyTorch: Image Segmentation | PyTorch | verify · pdf |
| 4 | Detecting COVID-19 with Chest X-Ray using PyTorch | PyTorch | verify · pdf |
| 5 | Object Localization with TensorFlow | TensorFlow | verify · pdf |
| 6 | Visualizing Filters of a CNN using TensorFlow | TensorFlow | verify · pdf |
| 7 | Tweet Emotion Recognition with TensorFlow | TensorFlow | verify · pdf |
| 8 | Data Balancing with Generative AI: Credit Card Fraud Detection | PyTorch | verify · pdf |
| 9 | Breast Cancer Prediction Using Machine Learning | scikit-learn | verify · pdf |
What the set covers
- Core architectures — CNNs for classification, segmentation, and localization; GANs and generative models; Siamese networks for metric learning.
- Interpretability — filter visualization to unpack what convolutional layers learn at each depth.
- Applied medical imaging — COVID chest X-ray classification and breast cancer prediction.
- Applied NLP — emotion recognition from short-form text.
- Working with imbalance — generative approaches to handling skewed medical and financial datasets.
Each project was end-to-end: data loading, model construction, training, evaluation. Framework coverage across PyTorch and TensorFlow is intentional — both matter depending on whose code you’re reading.