huangfu/InteractionImitation
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refactor LL buffer

Browse Source
This commit is contained in:
ebuehrle
2022-01-21 10:20:08 +01:00
parent 3b60c14319
commit 52d4b5a36c
2 changed files with 5 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
scratch/etienne/intersimple/data/generate.sh
View File

@@ -11,4 +11,5 @@
#python -m expert --env=NRasterizedRouteRandomAgentLocation --min_timesteps=100000 --env_args='{width:70,height:70,m_per_px:1,map_color:128}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N100000_NRasterizedRouteRandomAgentLocationw70h70mppx1mapc128.pkl' #python -m expert --env=NRasterizedRouteRandomAgentLocation --min_timesteps=100000 --env_args='{width:70,height:70,m_per_px:1,map_color:128}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N100000_NRasterizedRouteRandomAgentLocationw70h70mppx1mapc128.pkl'
#python -m expert --env=NRasterizedRouteSpeedRandomAgentLocation --min_timesteps=10000 --env_args='{width:70,height:70,m_per_px:1,map_color:128,mu:0.001}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N10000_NRasterizedRouteSpeedRandomAgentLocationw70h70mppx1mapc128mu.001.pkl' #python -m expert --env=NRasterizedRouteSpeedRandomAgentLocation --min_timesteps=10000 --env_args='{width:70,height:70,m_per_px:1,map_color:128,mu:0.001}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N10000_NRasterizedRouteSpeedRandomAgentLocationw70h70mppx1mapc128mu.001.pkl'
#python -m data.expert --env=NRasterizedRouteSpeedRandomAgentLocation --min_timesteps=10000 --env_args='{width:70,height:70,m_per_px:1,map_color:128,mu:0.001,skip_frames:5}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N10000_NRasterizedRouteSpeedRandomAgentLocationw70h70mppx1mapc128mu.001skip5.pkl' #python -m data.expert --env=NRasterizedRouteSpeedRandomAgentLocation --min_timesteps=10000 --env_args='{width:70,height:70,m_per_px:1,map_color:128,mu:0.001,skip_frames:5}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N10000_NRasterizedRouteSpeedRandomAgentLocationw70h70mppx1mapc128mu.001skip5.pkl'
python -m data.expert --env=TLNRasterizedRouteRandomAgentLocation --min_timesteps=100000 --env_args='{width:70,height:70,m_per_px:1,mu:0.001,random_skip:True,max_episode_steps:50}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N100000_TLNRasterizedRouteRandomAgentLocationw70h70mppx1mu.001rskips50.pkl' #python -m data.expert --env=TLNRasterizedRouteRandomAgentLocation --min_timesteps=100000 --env_args='{width:70,height:70,m_per_px:1,mu:0.001,random_skip:True,max_episode_steps:50}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N100000_TLNRasterizedRouteRandomAgentLocationw70h70mppx1mu.001rskips50.pkl'
python -m data.expert --env=TLNRasterizedRouteRandomAgentLocation --min_timesteps=50000 --env_args='{width:70,height:70,m_per_px:1,mu:0.001,random_skip:True,max_episode_steps:50}' --policy_args='{mu:0.001}' --path='NormalizedIntersimpleExpertMu.001N50000_TLNRasterizedRouteRandomAgentLocationw70h70mppx1mu.001rskips50.pkl'

13
scratch/etienne/intersimple/gail/options2.py
View File

@@ -14,7 +14,7 @@ def imitation_discriminator(discriminator):
class OptionsEnv(gym.Wrapper): class OptionsEnv(gym.Wrapper):
def __init__(self, env, options, discriminator, discount, ll_buffer_capacity, *args, **kwargs): def __init__(self, env, options, discriminator, discount, ll_buffer, *args, **kwargs):
super().__init__(env, *args, **kwargs) super().__init__(env, *args, **kwargs)
self.options = options self.options = options
@@ -27,8 +27,7 @@ class OptionsEnv(gym.Wrapper):
self.discriminator = discriminator self.discriminator = discriminator
self.discount = discount self.discount = discount
self.ll_buffer_capacity = ll_buffer_capacity self.ll_buffer = ll_buffer
self.ll_buffer = deque(maxlen=ll_buffer_capacity)
@staticmethod @staticmethod
def _hl_observation(obs, mask): def _hl_observation(obs, mask):
@@ -78,16 +77,10 @@ class OptionsEnv(gym.Wrapper):
return self._hl_observation(self.obs, self.m), reward, self.done, info return self._hl_observation(self.obs, self.m), reward, self.done, info
def sample_ll(self, n):
assert n <= self.ll_buffer_capacity, f'Sample size of {n} exceeds buffer capacity of {self.ll_buffer_capacity}'
assert n <= len(self.ll_buffer), f'Sample size of {n} exceeds buffer size of {len(self.ll_buffer)}'
ind = np.random.randint(len(self.ll_buffer), size=n)
return list(self.ll_buffer[i] for i in ind)
class RenderOptions(OptionsEnv): class RenderOptions(OptionsEnv):
def __init__(self, env, options, *args, **kwargs): def __init__(self, env, options, *args, **kwargs):
super().__init__(env, options, discriminator=lambda s, a, n, d: 0, discount=1, ll_buffer_capacity=0, *args, **kwargs) super().__init__(env, options, discriminator=lambda s, a, n, d: 0, discount=1, ll_buffer=deque(maxlen=0), *args, **kwargs)
def _ll_step(self, action): def _ll_step(self, action):
out = super()._ll_step(action) out = super()._ll_step(action)