From 5faf51b4f8c6e32567bc225ca33a0ab180efadcf Mon Sep 17 00:00:00 2001
From: Hsien-Chin Lin <linh@hhu.de>
Date: Thu, 19 Jan 2023 23:51:07 +0100
Subject: [PATCH] wip
---
convlab/policy/ppo/train.py | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/convlab/policy/ppo/train.py b/convlab/policy/ppo/train.py
index e4971abc..6a3f6564 100755
--- a/convlab/policy/ppo/train.py
+++ b/convlab/policy/ppo/train.py
@@ -239,17 +239,17 @@ if __name__ == '__main__':
policy_sys.log_dir = config_save_path.replace('configs', 'logs')
policy_sys.save_dir = save_path
- # logging.info(f"Evaluating at start - {time_now}" + '-'*60)
- # time_now = time.time()
- # eval_dict = eval_policy(conf, policy_sys, env, sess,
- # save_eval, log_save_path)
- # logging.info(f"Finished evaluating, time spent: {time.time() - time_now}")
-
- # for key in eval_dict:
- # tb_writer.add_scalar(key, eval_dict[key], 0)
- # best_complete_rate = eval_dict['complete_rate']
- # best_success_rate = eval_dict['success_rate_strict']
- # best_return = eval_dict['avg_return']
+ logging.info(f"Evaluating at start - {time_now}" + '-'*60)
+ time_now = time.time()
+ eval_dict = eval_policy(conf, policy_sys, env, sess,
+ save_eval, log_save_path)
+ logging.info(f"Finished evaluating, time spent: {time.time() - time_now}")
+
+ for key in eval_dict:
+ tb_writer.add_scalar(key, eval_dict[key], 0)
+ best_complete_rate = eval_dict['complete_rate']
+ best_success_rate = eval_dict['success_rate_strict']
+ best_return = eval_dict['avg_return']
logging.info("Start of Training: " +
time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime()))
--
GitLab