Note: All images in this directory, unless specified otherwise, are licensed under CC BY-NC 4.0.
| Figure number | Description |
|---|---|
| 5-1 | TensorBoard default view showcasing real-time training metrics (the lightly shaded lines represent the accuracy from the previous run) |
| 5-2 | TensorFlow Embedding Projector showcasing data in clusters (can be run as a TensorBoard plugin) |
| 5-3 | What-If Tool’s datapoint editor makes it possible to filter and visualize data according to annotations of the dataset and labels from the classifier |
| 5-4 | PR curves in the Performance and Fairness section of the What-If Tool help to interactively select the optimal threshold to maximize precision and recall |
| 5-5 | Setup window for the What-If Tool |
| 5-6 | The What-If tool enables using multiple metrics, data visualization, and many more things under the sun |
| 5-7 | Choose the model to compare using the What-If Tool |
| 5-8 | Visualizations on images using MobileNet and tf-explain |
| 5-9 | Slicing and dividing the data based on predictions and real categories |
| 5-10 | Comparing transfer learning versus training a custom model on different datasets |
| 5-11 | Effect of % layers fine-tuned on model accuracy |
| 5-12 | Effect of the amount of data per category on model accuracy |
| 5-13 | Effect of learning rate on model accuracy and speed of convergence |
| 5-14 | Effect of different optimizers on the speed of convergence |
| 5-15 | Effect of batch size on accuracy and speed of convergence |
| 5-16 | Effect of image size on accuracy |
| 5-17 | Distribution of aspect ratio and corresponding accuracies in images |
| 5-18 | Output of augmentation strategies learned by reinforcement learning on the ImageNet dataset |
