Fix windows env setup bugs and other typo. (#11)

setup.py

@@ -16,7 +16,7 @@ setup(
     name='tianshou',
     version=version,
     description='A Library for Deep Reinforcement Learning',
-    long_description=open('README.md').read(),
+    long_description=open('README.md', encoding='utf8').read(),
     long_description_content_type='text/markdown',
     url='https://github.com/thu-ml/tianshou',
     author='TSAIL',

test/discrete/test_a2c.py

@@ -74,7 +74,7 @@ def test_a2c(args=get_args()):
         policy, train_envs, ReplayBuffer(args.buffer_size))
     test_collector = Collector(policy, test_envs)
     # log
-    writer = SummaryWriter(args.logdir + '/' + 'ppo')
+    writer = SummaryWriter(args.logdir + '/' + 'a2c')
 
     def stop_fn(x):
         return x >= env.spec.reward_threshold

test/discrete/test_dqn.py

@@ -73,7 +73,7 @@ def test_dqn(args=get_args()):
     # policy.set_eps(1)
     train_collector.collect(n_step=args.batch_size)
     # log
-    writer = SummaryWriter(args.logdir + '/' + 'ppo')
+    writer = SummaryWriter(args.logdir + '/' + 'dqn')
 
     def stop_fn(x):
         return x >= env.spec.reward_threshold

test/discrete/test_pg.py

@@ -122,7 +122,7 @@ def test_pg(args=get_args()):
         policy, train_envs, ReplayBuffer(args.buffer_size))
     test_collector = Collector(policy, test_envs)
     # log
-    writer = SummaryWriter(args.logdir + '/' + 'ppo')
+    writer = SummaryWriter(args.logdir + '/' + 'pg')
 
     def stop_fn(x):
         return x >= env.spec.reward_threshold

tianshou/__init__.py

@@ -1,7 +1,7 @@
 from tianshou import data, env, utils, policy, trainer, \
     exploration
 
-__version__ = '0.2.0post2'
+__version__ = '0.2.0.post2'
 __all__ = [
     'env',
     'data',