Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
ModelZoo
ResNet50_tensorflow
Commits
6e070e77
"megatron/vscode:/vscode.git/clone" did not exist on "60ffce358957caa5a2c10d93ca379dd02e87d1a3"
Commit
6e070e77
authored
Mar 10, 2020
by
Chen Chen
Committed by
A. Unique TensorFlower
Mar 10, 2020
Browse files
Internal change
PiperOrigin-RevId: 300115297
parent
6e910b1f
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
9 additions
and
7 deletions
+9
-7
official/staging/training/controller.py
official/staging/training/controller.py
+9
-7
No files found.
official/staging/training/controller.py
View file @
6e070e77
...
...
@@ -94,9 +94,11 @@ class Controller(object):
# TODO(rxsang): Support training until exhaustion by passing
# `train_steps=-1`. Currently it cannot be supported with a host training
# loop because break statements are not supported with distributed dataset.
if
train_fn
is
not
None
and
train_steps
is
None
:
raise
ValueError
(
"`train_steps` is required when `train_fn` is provided."
)
if
train_fn
is
not
None
and
steps_per_loop
is
None
:
if
train_fn
is
not
None
:
if
train_steps
is
None
:
raise
ValueError
(
"`train_steps` is required when `train_fn` is "
"provided."
)
if
steps_per_loop
is
None
:
raise
ValueError
(
"`steps_per_loop` is required when `train_fn is "
"provided."
)
if
not
isinstance
(
steps_per_loop
,
int
)
or
steps_per_loop
<
1
:
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment