def sync(root, force_all=False):
last_ts = last_timestamp()
if last_ts and not force_all:
# Build a list of days from the day before our last timestamp to today
num_days = (dt.datetime.utcnow() -
dt.datetime.utcfromtimestamp(last_ts) +
dt.timedelta(days=1)).days
ds = [
(dt.datetime.utcnow() - dt.timedelta(days=d)).strftime("%Y-%m-%d")
for d in range(num_days + 1)
]
for d in ds:
if not os.path.isdir(os.path.join(root, d)):
os.mkdir(os.path.join(root, d))
cmd = [
"gsutil", "-m", "rsync", "-r",
os.path.join(fsdb.eval_dir(), d),
os.path.join(root, d)
]
print(" ".join(cmd))
subprocess.call(cmd)
ingest_dirs(root, ds)
else:
cmd = ["gsutil", "-m", "rsync", "-r", fsdb.eval_dir(), root]
print(" ".join(cmd))
subprocess.call(cmd)
dirs = os.listdir(root)
ingest_dirs(root, dirs)
def main(unused_argv):
logging.getLogger('mlperf_compliance').propagate = False
sgf_dir = os.path.join(fsdb.eval_dir(), 'target')
target = 'tf,' + os.path.join(fsdb.models_dir(), 'target.pb.og')
models = load_train_times()
timestamp_to_log = 0
iter_evaluated = 0
for i, (timestamp, name, path) in enumerate(models):
minigo_print(key=constants.EVAL_START, metadata={'epoch_num': i + 1})
iter_evaluated += 1
winrate = wait(evaluate_model(path + '.og', target, sgf_dir, i + 1))
minigo_print(key=constants.EVAL_ACCURACY,
value=winrate,
metadata={'epoch_num': i + 1})
minigo_print(key=constants.EVAL_STOP, metadata={'epoch_num': i + 1})
if winrate >= 0.50:
timestamp_to_log = timestamp
print('Model {} beat target after {}s'.format(name, timestamp))
break
minigo_print(key='eval_result',
metadata={
'iteration': iter_evaluated,
'timestamp': timestamp_to_log
})
def main(unused_argv):
"""Run the reinforcement learning loop."""
print('Wiping dir %s' % FLAGS.base_dir, flush=True)
shutil.rmtree(FLAGS.base_dir, ignore_errors=True)
utils.ensure_dir_exists(fsdb.models_dir())
utils.ensure_dir_exists(fsdb.selfplay_dir())
utils.ensure_dir_exists(fsdb.holdout_dir())
utils.ensure_dir_exists(fsdb.eval_dir())
utils.ensure_dir_exists(fsdb.golden_chunk_dir())
utils.ensure_dir_exists(fsdb.working_dir())
# Copy the target model to the models directory so we can find it easily.
shutil.copy('ml_perf/target.pb', fsdb.models_dir())
logging.getLogger().addHandler(
logging.FileHandler(os.path.join(FLAGS.base_dir, 'reinforcement.log')))
formatter = logging.Formatter('[%(asctime)s] %(message)s',
'%Y-%m-%d %H:%M:%S')
for handler in logging.getLogger().handlers:
handler.setFormatter(formatter)
with utils.logged_timer('Total time'):
for target_win_rate in rl_loop():
if target_win_rate > 0.5:
return logging.info('Passed exit criteria.')
logging.info('Failed to converge.')
def main(unused_argv):
"""Run the reinforcement learning loop."""
print('Wiping dir %s' % FLAGS.base_dir, flush=True)
shutil.rmtree(FLAGS.base_dir, ignore_errors=True)
utils.ensure_dir_exists(fsdb.models_dir())
utils.ensure_dir_exists(fsdb.selfplay_dir())
utils.ensure_dir_exists(fsdb.holdout_dir())
utils.ensure_dir_exists(fsdb.eval_dir())
utils.ensure_dir_exists(fsdb.golden_chunk_dir())
utils.ensure_dir_exists(fsdb.working_dir())
# Copy the flag files so there's no chance of them getting accidentally
# overwritten while the RL loop is running.
flags_dir = os.path.join(FLAGS.base_dir, 'flags')
shutil.copytree(FLAGS.flags_dir, flags_dir)
FLAGS.flags_dir = flags_dir
# Copy the target model to the models directory so we can find it easily.
shutil.copy('ml_perf/target.pb', fsdb.models_dir())
logging.getLogger().addHandler(
logging.FileHandler(os.path.join(FLAGS.base_dir, 'rl_loop.log')))
formatter = logging.Formatter('[%(asctime)s] %(message)s',
'%Y-%m-%d %H:%M:%S')
for handler in logging.getLogger().handlers:
handler.setFormatter(formatter)
with utils.logged_timer('Total time'):
try:
rl_loop()
finally:
asyncio.get_event_loop().close()
async def evaluate_trained_model_parallel(state):
"""Evaluate the most recently trained model against the current best model.
Args:
state: the RL loop State instance.
"""
all_tasks = []
loop = asyncio.get_event_loop()
for i in range(FLAGS.num_parallel_eval):
all_tasks.append(
loop.create_task(
evaluate_model_parallel(
state.train_model_path_eval, state.best_model_path_eval,
os.path.join(fsdb.eval_dir(), state.train_model_name),
state.seed, i, 'parallel_eval')))
all_lines = await asyncio.gather(*all_tasks, return_exceptions=True)
total_games = 0
total_wins = 0
for lines in all_lines:
if type(lines) == RuntimeError or type(lines) == OSError:
raise lines
total_games = total_games + lines[0]
total_wins = total_wins + lines[1]
print('Iter = {} Eval of {} against best={}, games={}, wins={}'.format(
state.iter_num, state.train_model_name, state.best_model_name,
total_games, total_wins))
win_rate = total_wins / total_games
return win_rate
def main(unused_argv):
"""Run the reinforcement learning loop."""
print('Wiping dir %s' % FLAGS.base_dir, flush=True)
shutil.rmtree(FLAGS.base_dir, ignore_errors=True)
dirs = [fsdb.models_dir(), fsdb.selfplay_dir(), fsdb.holdout_dir(),
fsdb.eval_dir(), fsdb.golden_chunk_dir(), fsdb.working_dir()]
for d in dirs:
ensure_dir_exists(d);
# Copy the flag files so there's no chance of them getting accidentally
# overwritten while the RL loop is running.
flags_dir = os.path.join(FLAGS.base_dir, 'flags')
shutil.copytree(FLAGS.flags_dir, flags_dir)
FLAGS.flags_dir = flags_dir
# Copy the target model to the models directory so we can find it easily.
for file_name in [
"target.pb", "target_raw.ckpt.data-00000-of-00001",
"target_raw.ckpt.index", "target_raw.ckpt.meta"]:
shutil.copy(FLAGS.target_path[:-len("target.pb")] + file_name,
os.path.join(fsdb.models_dir(), file_name))
logging.getLogger().addHandler(
logging.FileHandler(os.path.join(FLAGS.base_dir, 'rl_loop.log')))
formatter = logging.Formatter('[%(asctime)s] %(message)s',
'%Y-%m-%d %H:%M:%S')
for handler in logging.getLogger().handlers:
handler.setFormatter(formatter)
with logged_timer('Total time'):
try:
rl_loop()
finally:
asyncio.get_event_loop().close()
def main(unused_argv):
sgf_dir = os.path.join(fsdb.eval_dir(), 'target')
target = 'tf,' + os.path.join(fsdb.models_dir(), 'target.pb')
models = load_train_times()
for i, (timestamp, name, path) in enumerate(models):
winrate = wait(evaluate_model(path, name, target, sgf_dir))
if winrate >= 0.50:
break
def main(unused_argv):
sgf_dir = os.path.join(fsdb.eval_dir(), 'target')
target = 'tf,' + os.path.join(fsdb.models_dir(), 'target.pb')
models = load_train_times()
for i, (timestamp, name, path) in enumerate(models):
winrate = wait(evaluate_model(path, target, sgf_dir, i + 1))
if winrate >= 0.50:
print('Model {} beat target after {}s'.format(name, timestamp))
break
async def evaluate_trained_model(state):
"""Evaluate the most recently trained model against the current best model.
Args:
state: the RL loop State instance.
"""
return await evaluate_model(
state.train_model_path, state.best_model_path,
os.path.join(fsdb.eval_dir(), state.train_model_name), state.seed)
def evaluate(state, args, name, slice):
sgf_dir = os.path.join(fsdb.eval_dir(), state.train_model_name)
result = checked_run([
'external/minigo/cc/main', '--mode=eval', '--parallel_games=100',
'--model={}'.format(
state.train_model_path), '--sgf_dir={}'.format(sgf_dir)
] + args, name)
result = get_lines(result, slice)
logging.info(result)
pattern = '{}\s+\d+\s+(\d+\.\d+)%'.format(state.train_model_name)
return float(re.search(pattern, result).group(1)) * 0.01
def evaluate_trained_model_noasync(state):
"""Evaluate the most recently trained model against the current best model.
Args:
state: the RL loop State instance.
"""
return evaluate_model_noasync(
state.train_model_path_eval, state.best_model_path_eval,
os.path.join(fsdb.eval_dir(), state.train_model_name), state.seed,
'parallel_eval')
async def eval_vs_target(state):
# If we're bootstrapping a checkpoint, evaluate the newly trained model
# against the target.
# TODO(tommadams): evaluate the previously trained model against the
# target in parallel with training the next model.
if FLAGS.bootstrap and state.iter_num > 2:
target_model_path = os.path.join(fsdb.models_dir(), 'target.pb')
sgf_dir = os.path.join(fsdb.eval_dir(),
'{}-vs-target'.format(state.train_model_name))
win_rate_vs_target = await evaluate_model(state.selfplay_model_path,
target_model_path, sgf_dir)
return win_rate_vs_target
return None
def main(unused_argv):
"""Run the reinforcement learning loop."""
logging.getLogger('mlperf_compliance').propagate = False
##-->multi-node setup
if FLAGS.use_multinode:
mpi_comm = MPI.COMM_WORLD
mpi_rank = mpi_comm.Get_rank()
mpi_size = mpi_comm.Get_size()
print('[MPI Init] MPI rank {}, mpi size is {} host is {}'.format(
mpi_rank, mpi_size, socket.gethostname()))
else:
mpi_comm = None
mpi_rank = 0
mpi_size = 1
print('Wiping dir %s' % FLAGS.base_dir, flush=True)
shutil.rmtree(FLAGS.base_dir, ignore_errors=True)
dirs = [
fsdb.models_dir(),
fsdb.selfplay_dir(),
fsdb.holdout_dir(),
fsdb.eval_dir(),
fsdb.golden_chunk_dir(),
fsdb.working_dir()
]
##-->sharedFS for dataExchange. tmp solution 5/6/2019
if FLAGS.use_multinode:
ensure_dir_exists(FLAGS.shared_dir_exchange)
for d in dirs:
ensure_dir_exists(d)
# Copy the flag files so there's no chance of them getting accidentally
# overwritten while the RL loop is running.
flags_dir = os.path.join(FLAGS.base_dir, 'flags')
shutil.copytree(FLAGS.flags_dir, flags_dir)
FLAGS.flags_dir = flags_dir
# Copy the target model to the models directory so we can find it easily.
shutil.copy(FLAGS.target_path, os.path.join(fsdb.models_dir(),
'target.pb'))
shutil.copy(FLAGS.target_path + '.og',
os.path.join(fsdb.models_dir(), 'target.pb.og'))
with logged_timer('Total time from mpi_rank={}'.format(mpi_rank)):
try:
rl_loop(mpi_comm, mpi_rank, mpi_size)
finally:
asyncio.get_event_loop().close()
def rl_loop():
"""The main reinforcement learning (RL) loop."""
state = State()
prev_win_rate_vs_target = 0
if FLAGS.bootstrap:
wait(bootstrap_selfplay(state))
else:
initialize_from_checkpoint(state)
# Start the selfplay workers. They will wait for a model to become available
# in the training directory before starting to play.
selfplay_processes, selfplay_logs = wait(start_selfplay())
try:
# Now start the full training loop.
while state.iter_num < FLAGS.iterations:
state.iter_num += 1
wait_for_training_examples(state, FLAGS.min_games_per_iteration)
tf_records = wait(sample_training_examples(state))
wait(train(state, tf_records))
# If we're bootstrapping a checkpoint, evaluate the newly trained model
# against the target.
# TODO(tommadams): evaluate the previously trained model against the
# target in parallel with training the next model.
if FLAGS.bootstrap and state.iter_num > 15:
target_model_path = os.path.join(fsdb.models_dir(),
'target.pb')
sgf_dir = os.path.join(
fsdb.eval_dir(),
'{}-vs-target'.format(state.train_model_name))
win_rate_vs_target = wait(
evaluate_model(state.train_model_path, target_model_path,
sgf_dir))
if (win_rate_vs_target >= FLAGS.bootstrap_target_win_rate
and prev_win_rate_vs_target > 0):
# The tranined model won a sufficient number of games against
# the target. Create the checkpoint that will be used to start
# the real benchmark and exit.
create_checkpoint()
break
prev_win_rate_vs_target = win_rate_vs_target
finally:
wait(end_selfplay(selfplay_processes, selfplay_logs))
def main(unused_argv):
sgf_dir = os.path.join(fsdb.eval_dir(), 'target')
target = 'tf,' + os.path.join(fsdb.models_dir(), 'target.pb')
models = load_train_times()
for i, (timestamp, name, path) in enumerate(models):
mll.eval_start(i)
winrate = wait(evaluate_model(path, target, sgf_dir, i + 1))
mll.eval_stop(i)
mll.eval_accuracy(i, winrate)
if winrate >= 0.50:
print('Model {} beat target after {}s'.format(name, timestamp))
mll.eval_result(i, timestamp)
mll.run_stop('success')
return
mll.eval_result(i, 0)
mll.run_stop('aborted')
def evaluate(state, against_model):
eval_model = state.train_model_name
eval_model_path = os.path.join(fsdb.models_dir(), eval_model)
against_model_path = os.path.join(fsdb.models_dir(), against_model)
sgf_dir = os.path.join(fsdb.eval_dir(), eval_model)
result = checked_run([
'bazel-bin/cc/eval', '--num_readouts=100', '--parallel_games=100',
'--model={}.pb'.format(eval_model_path),
'--model_two={}.pb'.format(against_model_path),
'--sgf_dir={}'.format(sgf_dir)
] + cc_flags(state), 'evaluation against ' + against_model)
result = get_lines(result, make_slice[-7:])
logging.info(result)
pattern = '{}\s+\d+\s+(\d+\.\d+)%'.format(eval_model)
win_rate = float(re.search(pattern, result).group(1)) * 0.01
logging.info('Win rate %s vs %s: %.3f', eval_model, against_model,
win_rate)
return win_rate
def evaluate(state):
eval_model = state.train_model_name
best_model = state.best_model_name
eval_model_path = os.path.join(fsdb.models_dir(), eval_model)
best_model_path = os.path.join(fsdb.models_dir(), best_model)
sgf_dir = os.path.join(fsdb.eval_dir(), eval_model)
result = checked_run(
'evaluation', 'bazel-bin/cc/eval',
'--flagfile={}'.format(os.path.join(FLAGS.flags_dir, 'eval.flags')),
'--model={}.pb'.format(eval_model_path),
'--model_two={}.pb'.format(best_model_path),
'--sgf_dir={}'.format(sgf_dir), '--seed={}'.format(state.seed))
result = get_lines(result, make_slice[-7:])
logging.info(result)
pattern = '{}\s+\d+\s+(\d+\.\d+)%'.format(eval_model)
win_rate = float(re.search(pattern, result).group(1)) * 0.01
logging.info('Win rate %s vs %s: %.3f', eval_model, best_model, win_rate)
return win_rate
def main(unused_argv):
"""Run the reinforcement learning loop."""
mll.init_start()
print('Wiping dir %s' % FLAGS.base_dir, flush=True)
shutil.rmtree(FLAGS.base_dir, ignore_errors=True)
dirs = [fsdb.models_dir(), fsdb.selfplay_dir(), fsdb.holdout_dir(),
fsdb.eval_dir(), fsdb.golden_chunk_dir(), fsdb.working_dir(),
fsdb.mpi_log_dir()]
for d in dirs:
ensure_dir_exists(d);
# Copy the flag files so there's no chance of them getting accidentally
# overwritten while the RL loop is running.
flags_dir = os.path.join(FLAGS.base_dir, 'flags')
shutil.copytree(FLAGS.flags_dir, flags_dir)
FLAGS.flags_dir = flags_dir
# Copy the target model to the models directory so we can find it easily.
shutil.copy(FLAGS.target_path, os.path.join(fsdb.models_dir(), 'target.pb'))
logging.getLogger().addHandler(
logging.FileHandler(os.path.join(FLAGS.base_dir, 'rl_loop.log')))
formatter = logging.Formatter('[%(asctime)s] %(message)s',
'%Y-%m-%d %H:%M:%S')
for handler in logging.getLogger().handlers:
handler.setFormatter(formatter)
logging.info('Selfplay nodes = {}'.format(FLAGS.selfplay_node))
logging.info('Train nodes = {}'.format(FLAGS.train_node))
logging.info('Eval nodes = {}'.format(FLAGS.eval_node))
with logged_timer('Total time'):
try:
mll.init_stop()
mll.run_start()
rl_loop()
finally:
asyncio.get_event_loop().close()
def main(unused_argv):
for i in range(0, NUM_LOOP):
if i == 0:
src_model_name = shipname.generate(0)
fsdb.switch_base(os.path.join(base_dir, src_model_name))
src_model_path = os.path.join(fsdb.models_dir(), src_model_name)
bootstrap_model_path = os.path.join(fsdb.models_dir(),
src_model_name)
mask_flags.checked_run([
'python3', 'bootstrap.py',
'--export_path={}'.format(bootstrap_model_path),
'--work_dir={}'.format(fsdb.working_dir()),
'--flagfile=rl_loop/local_flags'
])
dst_model_name = shipname.generate(1)
fsdb.switch_base(os.path.join(base_dir, dst_model_name))
else:
src_model_name = dst_model_name
src_model_path = os.path.join(fsdb.models_dir(), src_model_name)
dst_model_name = shipname.generate(i + 1)
fsdb.switch_base(os.path.join(base_dir, dst_model_name))
utils.ensure_dir_exists(fsdb.models_dir())
utils.ensure_dir_exists(fsdb.selfplay_dir())
utils.ensure_dir_exists(fsdb.holdout_dir())
utils.ensure_dir_exists(fsdb.sgf_dir())
utils.ensure_dir_exists(fsdb.eval_dir())
utils.ensure_dir_exists(fsdb.golden_chunk_dir())
utils.ensure_dir_exists(fsdb.working_dir())
#bootstrap_name = shipname.generate(0)
#bootstrap_model_path = os.path.join(fsdb.models_dir(), bootstrap_name)
print(src_model_name)
print(src_model_path)
selfplay_cmd = [
'python3', 'selfplay.py', '--load_file={}'.format(src_model_path),
'--selfplay_dir={}'.format(
os.path.join(fsdb.selfplay_dir(),
dst_model_name)), '--holdout_dir={}'.format(
os.path.join(fsdb.holdout_dir(),
dst_model_name)),
'--sgf_dir={}'.format(fsdb.sgf_dir()), '--holdout_pct=0',
'--flagfile=rl_loop/local_flags'
]
# Selfplay twice
mask_flags.checked_run(selfplay_cmd)
mask_flags.checked_run(selfplay_cmd)
# and once more to generate a held out game for validation
# exploits flags behavior where if you pass flag twice, second one wins.
mask_flags.checked_run(selfplay_cmd + ['--holdout_pct=100'])
# Double check that at least one sgf has been generated.
assert os.listdir(os.path.join(fsdb.sgf_dir(), 'full'))
print("Making shuffled golden chunk from selfplay data...")
# TODO(amj): refactor example_buffer so it can be called the same way
# as everything else.
eb.make_chunk_for(output_dir=fsdb.golden_chunk_dir(),
local_dir=fsdb.working_dir(),
game_dir=fsdb.selfplay_dir(),
model_num=1,
positions=64,
threads=8,
sampling_frac=1)
tf_records = sorted(
gfile.Glob(os.path.join(fsdb.golden_chunk_dir(), '*.tfrecord.zz')))
#trained_model_name = shipname.generate(1)
trained_model_name = dst_model_name
trained_model_path = os.path.join(fsdb.models_dir(),
trained_model_name)
# Train on shuffled game data
mask_flags.checked_run([
'python3', 'train.py', *tf_records,
'--work_dir={}'.format(fsdb.working_dir()),
'--export_path={}'.format(trained_model_path),
'--flagfile=rl_loop/local_flags'
])
print("Finished!")
async def evaluate_target_model(state):
sgf_dir = os.path.join(fsdb.eval_dir(), 'target')
target = 'tf,' + os.path.join(fsdb.models_dir(), 'target.pb')
return await evaluate_model(state.train_model_path, target, sgf_dir,
state.iter_num)
def rl_loop():
"""The main reinforcement learning (RL) loop."""
state = State()
if FLAGS.bootstrap:
# Play the first round of selfplay games with a fake model that returns
# random noise. We do this instead of playing multiple games using a
# single model bootstrapped with random noise to avoid any initial bias.
wait(bootstrap_selfplay(state))
# Train a real model from the random selfplay games.
tf_records = wait(sample_training_examples(state))
state.iter_num += 1
wait(train(state, tf_records))
# Select the newly trained model as the best.
state.best_model_name = state.train_model_name
state.gen_num += 1
# Run selfplay using the new model.
wait(selfplay(state))
else:
# Start from a partially trained model.
initialize_from_checkpoint(state)
prev_win_rate_vs_target = 0
# Now start the full training loop.
while state.iter_num <= FLAGS.iterations:
tf_records = wait(sample_training_examples(state))
state.iter_num += 1
# Run selfplay in parallel with sequential (train, eval).
model_win_rate, _ = wait(
[train_eval(state, tf_records),
selfplay(state)])
# If we're bootstrapping a checkpoint, evaluate the newly trained model
# against the target.
if FLAGS.bootstrap:
target_model_path = os.path.join(fsdb.models_dir(), 'target.pb')
sgf_dir = os.path.join(
fsdb.eval_dir(), '{}-vs-target'.format(state.train_model_name))
win_rate_vs_target = wait(
evaluate_model(state.train_model_path, target_model_path,
sgf_dir))
if (win_rate_vs_target >= FLAGS.bootstrap_target_win_rate
and prev_win_rate_vs_target > 0):
# The tranined model won a sufficient number of games against
# the target. Create the checkpoint that will be used to start
# the real benchmark and exit.
create_checkpoint()
break
prev_win_rate_vs_target = win_rate_vs_target
if model_win_rate >= FLAGS.gating_win_rate:
# Promote the trained model to the best model and increment the
# generation number.
state.best_model_name = state.train_model_name
state.gen_num += 1
def main(unused_argv):
"""Run the reinforcement learning loop."""
utils.ensure_dir_exists(fsdb.models_dir())
utils.ensure_dir_exists(fsdb.selfplay_dir())
utils.ensure_dir_exists(fsdb.holdout_dir())
utils.ensure_dir_exists(fsdb.sgf_dir())
utils.ensure_dir_exists(fsdb.eval_dir())
utils.ensure_dir_exists(fsdb.golden_chunk_dir())
utils.ensure_dir_exists(fsdb.working_dir())
bootstrap_name = shipname.generate(0)
bootstrap_model_path = os.path.join(fsdb.models_dir(), bootstrap_name)
mask_flags.checked_run([
'python3', 'bootstrap.py',
'--export_path={}'.format(bootstrap_model_path),
'--work_dir={}'.format(fsdb.working_dir()),
'--flagfile=rl_loop/local_flags'
])
selfplay_cmd = [
'python3', 'selfplay.py',
'--load_file={}'.format(bootstrap_model_path),
'--selfplay_dir={}'.format(
os.path.join(fsdb.selfplay_dir(),
bootstrap_name)), '--holdout_dir={}'.format(
os.path.join(fsdb.holdout_dir(), bootstrap_name)),
'--sgf_dir={}'.format(fsdb.sgf_dir()), '--holdout_pct=0',
'--flagfile=rl_loop/local_flags'
]
# Selfplay twice
mask_flags.checked_run(selfplay_cmd)
mask_flags.checked_run(selfplay_cmd)
# and once more to generate a held out game for validation
# exploits flags behavior where if you pass flag twice, second one wins.
mask_flags.checked_run(selfplay_cmd + ['--holdout_pct=100'])
# Double check that at least one sgf has been generated.
assert os.listdir(os.path.join(fsdb.sgf_dir(), 'full'))
print("Making shuffled golden chunk from selfplay data...")
# TODO(amj): refactor example_buffer so it can be called the same way
# as everything else.
eb.make_chunk_for(output_dir=fsdb.golden_chunk_dir(),
local_dir=fsdb.working_dir(),
game_dir=fsdb.selfplay_dir(),
model_num=1,
positions=64,
threads=8,
sampling_frac=1)
tf_records = sorted(
gfile.Glob(os.path.join(fsdb.golden_chunk_dir(), '*.tfrecord.zz')))
trained_model_name = shipname.generate(1)
trained_model_path = os.path.join(fsdb.models_dir(), trained_model_name)
# Train on shuffled game data
mask_flags.checked_run([
'python3', 'train.py', *tf_records,
'--work_dir={}'.format(fsdb.working_dir()),
'--export_path={}'.format(trained_model_path),
'--flagfile=rl_loop/local_flags'
])
# Validate the trained model on held out game
mask_flags.checked_run([
'python3', 'validate.py',
os.path.join(fsdb.holdout_dir(), bootstrap_name),
'--work_dir={}'.format(fsdb.working_dir()),
'--flagfile=rl_loop/local_flags'
])
# Verify that trained model works for selfplay
# exploits flags behavior where if you pass flag twice, second one wins.
mask_flags.checked_run(selfplay_cmd +
['--load_file={}'.format(trained_model_path)])
mask_flags.checked_run([
'python3', 'evaluate.py', bootstrap_model_path, trained_model_path,
'--games=1', '--eval_sgf_dir={}'.format(fsdb.eval_dir()),
'--flagfile=rl_loop/local_flags'
])
print("Completed integration test!")