MCPcopy Index your code
hub / github.com/tensorpack/tensorpack / _before_train

Method _before_train

tensorpack/callbacks/monitor.py:264–270  ·  view source on GitHub ↗
(self)

Source from the content-addressed store, hash-verified

262 self._writer.add_graph(self.graph)
263
264 def _before_train(self):
265 # Writing the graph is expensive (takes ~2min) when the graph is large.
266 # Therefore use a separate thread. It will then run in the
267 # background while TF is warming up in the first several iterations.
268 self._write_graph_thread = threading.Thread(target=self._write_graph)
269 self._write_graph_thread.daemon = True
270 self._write_graph_thread.start()
271
272 @HIDE_DOC
273 def process_summary(self, summary):

Callers

nothing calls this directly

Calls 1

startMethod · 0.45

Tested by

no test coverage detected