Skip to content

feature: eval_rate -> eval_coef #329

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

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

feature: eval_rate -> eval_coef #329

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
e6c4e0d
feature: cnfig eval_rate -> eval_coef
YuriCat Aug 2, 2022
79243d2
fix: reset num_results and num_episodes after epoch 0
YuriCat Aug 2, 2022
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
2 changes: 1 addition & 1 deletion config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ train_args:
maximum_episodes: 100000
epochs: -1
num_batchers: 2
eval_rate: 0.1
eval_coef: 0.85
worker:
num_parallel: 6
lambda: 0.7
Expand Down
10 changes: 8 additions & 2 deletions handyrl/train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -403,8 +403,8 @@ def __init__(self, args, net=None, remote=False):
random.seed(args['seed'])

self.env = make_env(env_args)
eval_modify_rate = (args['update_episodes'] ** 0.85) / args['update_episodes']
self.eval_rate = max(args['eval_rate'], eval_modify_rate)
self.eval_rate_fn = lambda n: (n ** self.args['eval_coef']) / n
self.eval_rate = self.eval_rate_fn(self.args['minimum_episodes'] + self.args['update_episodes'])
self.shutdown_flag = False
self.flags = set()

Expand Down Expand Up @@ -520,11 +520,17 @@ def output_wp(name, results):
std = (r2 / (n + 1e-6) - mean ** 2) ** 0.5
print('generation stats = %.3f +- %.3f' % (mean, std))

if self.model_epoch == 0:
self.num_episodes = 0
self.num_results = 0

model, steps = self.trainer.update()
if model is None:
model = self.model
self.update_model(model, steps)

# update evaluation ratio
self.eval_rate = self.eval_rate_fn(self.args['update_episodes'])
# clear flags
self.flags = set()

Expand Down