search

mks0601 / TF-SimpleHumanPose

TensorFlow implementation of "Simple Baselines for Human Pose Estimation and Tracking", ECCV 2018
335 stars 90 forks source link

Validation Loss #63

Open Abby263 opened 4 years ago

Abby263 commented 4 years ago

Hi, I was trying to plot Validation loss ( to decide when to stop training on custom data ) but was facing issues under "make_network" function in model.py.

Can you please suggest a better way for plotting Validation loss, as it will add up as an extra metrics over ( Val Precision & Val Recall) to decide best performing model.

mks0601 commented 4 years ago

I think making a shell script for testing snapshot from each epoch is good.