def before_run(self, run_context):
self._request_summary = (self._next_step is None
or self._timer.should_trigger_for_step(
self._next_step))
requests = {} #{"global_step": self._global_step_tensor}
opts = tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE)
return SessionRunArgs(requests, options=opts)
def before_run(self, _):
# cprnt(train="BEFORE Comet Histograms")
return SessionRunArgs(
fetches={
"global_step": tf.get_collection(tf.GraphKeys.GLOBAL_STEP),
"trainables": self.trainables,
})
def before_run(self, _):
# cprnt(**{self.mode: "BEFORE Console Logger"})
fetches = {
"global_step": tf.get_collection(tf.GraphKeys.GLOBAL_STEP),
**(self.tensors if self.mode == ModeKeys.TRAIN else {}),
}
return SessionRunArgs(fetches=fetches)
def before_run(self, _):
return SessionRunArgs( # noqa
fetches={
"global_step": tf.get_collection(tf.GraphKeys.GLOBAL_STEP),
"attention": tf.get_collection("ATTENTION"),
"targets": self.targets,
"labels": self.labels,
"predictions": self.predictions,
}
)
def before_run(self, _):
# cprnt(train="BEFORE Summary Saving ({})".format(self._global_step))
this_step = self._global_step + 1
fetches = {
"global_step":
tf.get_collection(tf.GraphKeys.GLOBAL_STEP),
**({
"summary": self.summary_op
} if this_step % self.summary_freq == 0 or this_step == 1 else {}),
}
return SessionRunArgs(fetches=fetches)
def before_run(self, _):
# cprnt(**{self.mode: "BEFORE Metadata"})
this_step = self.counter + 1
log_metadata = (self.mode == ModeKeys.EVAL and self.freq == "once"
if isinstance(self.freq, str) else this_step == 1
or this_step % self.freq == 0)
return SessionRunArgs(
fetches={
"global_step": tf.get_collection(tf.GraphKeys.GLOBAL_STEP)
},
options=(
tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE # pylint: disable=E1101
) if log_metadata else None),
)
def before_run(self, run_context):
return SessionRunArgs(self._global_step_tensor)
def before_run(self, _):
# cprnt(**{self.mode: "BEFORE Comet Progress"})
return SessionRunArgs(
fetches={
"global_step": tf.get_collection(tf.GraphKeys.GLOBAL_STEP)
})