Can I access machine learning test metrics to evaluate model performance

Robotics & Edge Computing Jetson & Embedded Systems Jetson Nano
1

After re-training a model (based on SSD-mobilenet) using transfer learning and my own images (train 70 : validate 15 : test 15) how can i access the training metrics produced when testing the model against the test images?

Is there a way to view precision and recall values of values for true positives, false positives, true negatives and false negatives?

Thanks for the help

2

Hi @jeheesom, assuming that you used the tutorial from Hello AI World, there is an eval_ssd.py script included here: jetson-inference/python/training/detection/ssd/eval_ssd.py

3

Just the information i needed, thank you.