Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix decode training parser #124

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fix decode training parser #124

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
795ce7b
add expected_input_format to chunk parser
Apr 26, 2020
c505372
add docstring for expected_input_format, remove unused arg
Apr 26, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions tf/chunkparser.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,13 +80,13 @@ def __init__(self,
expected_input_format,
shuffle_size=1,
sample=1,
buffer_size=1,
batch_size=256,
workers=None):
"""
Read data and yield batches of raw tensors.

'chunks' list of chunk filenames.
'expected_input_format' is an int, one of [1, 2, 3]. Determines the middle planes in convert_v5_to_tuple
'shuffle_size' is the size of the shuffle buffer.
'sample' is the rate to down-sample.
'workers' is the number of child workers to use.
Expand All @@ -99,7 +99,7 @@ def __init__(self,
chunkdata: type Bytes. Multiple records of v5 format where each record
consists of (state, policy, result, q)

raw: A byte string holding raw tensors contenated together. This is
raw: A byte string holding raw tensors concatenated together. This is
used to pass data from the workers to the parent. Exists because
TensorFlow doesn't have a fast way to unpack bit vectors. 7950 bytes
long.
Expand Down
3 changes: 2 additions & 1 deletion tf/decode_training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -293,10 +293,11 @@ def describe(self):


class TrainingStep:
def __init__(self, version):
def __init__(self, version, input_format=1):
self.version = version
# Construct a fake parser just to get access to it's variables
self.parser = chunkparser.ChunkParser(chunkparser.ChunkDataSrc([]),
expected_input_format=input_format,
workers=1)
self.NUM_HIST = 8
self.NUM_PIECE_TYPES = 6
Expand Down