MCPcopy Create free account
hub / github.com/dqwang122/HeterSumGraph / run_eval

Function run_eval

train.py:180–258  ·  view source on GitHub ↗

Repeatedly runs eval iterations, logging to screen and writing summaries. Saves the model with the best loss seen so far. :param model: the model :param loader: valid dataset loader :param valset: valid dataset which includes text and summary :param hps

(model, loader, valset, hps, best_loss, best_F, non_descent_cnt, saveNo)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 1

run_trainingFunction · 0.85

Calls 3

evaluationMethod · 0.95
getMetricMethod · 0.95
SLTesterClass · 0.90

Tested by

no test coverage detected