def build_and_train(slot_affinity_code, log_dir, run_ID, config_key):
affinity = affinity_from_code(slot_affinity_code)
assert isinstance(affinity, list) # One for each GPU.
config = configs[config_key]
variant = load_variant(log_dir)
config = update_config(config, variant)
sampler = GpuSampler(
EnvCls=AtariEnv,
env_kwargs=config["env"],
CollectorCls=GpuWaitResetCollector,
TrajInfoCls=AtariTrajInfo,
**config["sampler"]
)
algo = A2C(optim_kwargs=config["optim"], **config["algo"])
agent = AtariFfAgent(model_kwargs=config["model"], **config["agent"])
runner = SyncRl(
algo=algo,
agent=agent,
sampler=sampler,
affinity=affinity,
**config["runner"]
)
name = config["env"]["game"]
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(slot_affinity_code, log_dir, run_ID, config_key):
affinity = get_affinity(slot_affinity_code)
config = configs[config_key]
variant = load_variant(log_dir)
config = update_config(config, variant)
sampler = GpuParallelSampler(
EnvCls=AtariEnv,
env_kwargs=config["env"],
CollectorCls=WaitResetCollector,
TrajInfoCls=AtariTrajInfo,
**config["sampler"]
)
algo = A2C(optim_kwargs=config["optim"], **config["algo"])
agent = AtariLstmAgent(model_kwargs=config["model"], **config["agent"])
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
affinity=affinity,
**config["runner"]
)
name = config["env"]["game"]
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(slot_affinity_code, log_dir, run_ID, config_key):
affinity = affinity_from_code(slot_affinity_code)
config = configs[config_key]
# variant = load_variant(log_dir)
# config = update_config(config, variant)
sampler = CpuSampler(
EnvCls=AtariEnv,
env_kwargs=config["env"],
CollectorCls=EpisodicLivesWaitResetCollector,
TrajInfoCls=AtariTrajInfo,
**config["sampler"]
)
algo = A2C(optim_kwargs=config["optim"], **config["algo"])
agent = AtariFfAgent(model_kwargs=config["model"], **config["agent"])
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
affinity=affinity,
**config["runner"]
)
name = config["env"]["game"]
with logger_context(log_dir, run_ID, name, config): # Might have to flatten config
runner.train()
def build_and_train(slot_affinity_code, log_dir, run_ID):
# (Or load from a central store of configs.)
config = dict(
env=dict(game="pong"),
algo=dict(learning_rate=7e-4),
sampler=dict(batch_B=16),
)
affinity = get_affinity(slot_affinity_code)
variant = load_variant(log_dir)
global config
config = update_config(config, variant)
sampler = GpuParallelSampler(
EnvCls=AtariEnv,
env_kwargs=config["env"],
CollectorCls=WaitResetCollector,
batch_T=5,
# batch_B=16, # Get from config.
max_decorrelation_steps=400,
**config["sampler"])
algo = A2C(**config["algo"]) # Run with defaults.
agent = AtariFfAgent()
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=50e6,
log_interval_steps=1e5,
affinity=affinity,
)
name = "a2c_" + config["env"]["game"]
log_dir = "example_6"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(game="pong",
run_ID=0,
cuda_idx=None,
mid_batch_reset=False,
n_parallel=2):
affinity = dict(cuda_idx=cuda_idx, workers_cpus=list(range(n_parallel)))
Collector = GpuResetCollector if mid_batch_reset else GpuWaitResetCollector
print(f"To satisfy mid_batch_reset=={mid_batch_reset}, using {Collector}.")
sampler = GpuParallelSampler(
EnvCls=AtariEnv,
env_kwargs=dict(game=game,
num_img_obs=1), # Learn on individual frames.
CollectorCls=Collector,
batch_T=20, # Longer sampling/optimization horizon for recurrence.
batch_B=16, # 16 parallel environments.
max_decorrelation_steps=400,
)
algo = A2C() # Run with defaults.
agent = AtariLstmAgent()
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=50e6,
log_interval_steps=1e5,
affinity=affinity,
)
config = dict(game=game)
name = "a2c_" + game
log_dir = "example_4"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(game="pong",
run_ID=0,
cuda_idx=None,
sample_mode="serial",
n_parallel=2):
affinity = dict(cuda_idx=cuda_idx, workers_cpus=list(range(n_parallel)))
gpu_cpu = "CPU" if cuda_idx is None else f"GPU {cuda_idx}"
if sample_mode == "serial":
Sampler = SerialSampler # (Ignores workers_cpus.)
print(f"Using serial sampler, {gpu_cpu} for sampling and optimizing.")
elif sample_mode == "cpu":
Sampler = CpuSampler
print(
f"Using CPU parallel sampler (agent in workers), {gpu_cpu} for optimizing."
)
elif sample_mode == "gpu":
Sampler = GpuSampler
print(
f"Using GPU parallel sampler (agent in master), {gpu_cpu} for sampling and optimizing."
)
elif sample_mode == "alternating":
Sampler = AlternatingSampler
affinity["workers_cpus"] += affinity["workers_cpus"] # (Double list)
affinity["alternating"] = True # Sampler will check for this.
print(
f"Using Alternating GPU parallel sampler, {gpu_cpu} for sampling and optimizing."
)
sampler = Sampler(
EnvCls=AtariEnv,
TrajInfoCls=AtariTrajInfo,
env_kwargs=dict(game=game),
batch_T=5, # 5 time-steps per sampler iteration.
batch_B=16, # 16 parallel environments.
max_decorrelation_steps=400,
)
algo = A2C() # Run with defaults.
agent = AtariFfAgent()
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=50e6,
log_interval_steps=1e5,
affinity=affinity,
)
config = dict(game=game)
name = "a2c_" + game
log_dir = "example_3"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(slot_affinity_code, log_dir, run_ID, config_key):
affinity = affinity_from_code(slot_affinity_code)
config = configs[config_key]
variant = load_variant(log_dir)
config = update_config(config, variant)
sampler = AlternatingSampler(EnvCls=gym_make,
env_kwargs=config["env"],
**config["sampler"])
algo = A2C(optim_kwargs=config["optim"], **config["algo"])
agent = MujocoFfAgent(model_kwargs=config["model"], **config["agent"])
runner = MinibatchRl(algo=algo,
agent=agent,
sampler=sampler,
affinity=affinity,
**config["runner"])
name = config["env"]["id"]
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(game="pong", run_ID=0):
# Seems like we should be able to skip the intermediate step of the code,
# but so far have just always run that way.
# Change these inputs to match local machine and desired parallelism.
affinity_code = encode_affinity(
n_cpu_cores=16, # Use 16 cores across all experiments.
n_gpu=8, # Use 8 gpus across all experiments.
hyperthread_offset=24, # If machine has 24 cores.
n_socket=2, # Presume CPU socket affinity to lower/upper half GPUs.
gpu_per_run=2, # How many GPUs to parallelize one run across.
# cpu_per_run=1,
)
slot_affinity_code = prepend_run_slot(run_slot=0, affinity_code=affinity_code)
affinity = get_affinity(slot_affinity_code)
breakpoint()
sampler = GpuParallelSampler(
EnvCls=AtariEnv,
env_kwargs=dict(game=game),
CollectorCls=WaitResetCollector,
batch_T=5,
batch_B=16,
max_decorrelation_steps=400,
)
algo = A2C() # Run with defaults.
agent = AtariFfAgent()
runner = MultiGpuRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=50e6,
log_interval_steps=1e5,
affinity=affinity,
)
config = dict(game=game)
name = "a2c_" + game
log_dir = "example_7"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(slot_affinity_code, log_dir, run_ID, config_key):
affinity = affinity_from_code(slot_affinity_code)
config = configs[config_key]
variant = load_variant(log_dir)
config = update_config(config, variant)
config["algo_name"] = 'A2C'
t_env = pomdp_interface(**config["env"])
config["algo"]["discount"] = t_env.discount
sampler = GpuSampler(EnvCls=pomdp_interface,
env_kwargs=config["env"],
**config["sampler"])
algo = A2C(optim_kwargs=config["optim"], **config["algo"])
agent = PomdpFfAgent(model_kwargs=config["model"], **config["agent"])
runner = MinibatchRl(algo=algo,
agent=agent,
sampler=sampler,
affinity=affinity,
**config["runner"])
name = config["env"]["id"]
with logger_context(log_dir, run_ID, name, config):
runner.train()
def findOptimalAgent(reward, run_ID=0):
"""
Find the optimal agent for the MDP (see Config for
specification) under a custom reward function
using rlpyt's implementation of A2C.
"""
cpus = list(range(C.N_PARALLEL))
affinity = dict(cuda_idx=C.CUDA_IDX, workers_cpus=cpus)
sampler = SerialSampler(EnvCls=rlpyt_make,
env_kwargs=dict(id=C.ENV, reward=reward),
batch_T=C.BATCH_T,
batch_B=C.BATCH_B,
max_decorrelation_steps=400,
eval_env_kwargs=dict(id=C.ENV),
eval_n_envs=5,
eval_max_steps=2500)
algo = A2C(discount=C.DISCOUNT,
learning_rate=C.LR,
value_loss_coeff=C.VALUE_LOSS_COEFF,
entropy_loss_coeff=C.ENTROPY_LOSS_COEFF)
agent = CategoricalPgAgent(AcrobotNet)
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=C.N_STEPS,
log_interval_steps=C.LOG_STEP,
affinity=affinity,
)
name = "a2c_" + C.ENV.lower()
log_dir = name
with logger_context(log_dir,
run_ID,
name,
snapshot_mode='last',
override_prefix=True):
runner.train()
return agent
def build_and_train(slot_affinity_code, log_dir, run_ID, config_key):
affinity = affinity_from_code(slot_affinity_code)
config = configs[config_key]
variant = load_variant(log_dir)
config = update_config(config, variant)
config["algo_name"] = 'A2C_RNN'
env = BatchPOMDPEnv(batch_B=config["sampler"]["batch_B"], **config["env"])
config["algo"]["discount"] = env.discount
sampler = BatchPOMDPSampler(env=env, **config["sampler"])
algo = A2C(optim_kwargs=config["optim"], **config["algo"])
agent = PomdpRnnAgent(model_kwargs=config["model"], **config["agent"])
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
affinity=affinity,
**config["runner"]
)
name = config["env"]["id"]
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(env_id="POMDP-hallway-episodic-v0",
run_ID=0,
cuda_idx=None,
n_parallel=6,
fomdp=False):
EnvCls = BatchPOMDPEnv
SamplerCls = BatchPOMDPSampler
batch_B = 30
batch_T = 100
env_args = dict(fomdp=fomdp, id=env_id, time_limit=100, batch_B=batch_B)
env = EnvCls(**env_args)
gamma = env.discount
affinity = dict(cuda_idx=cuda_idx,
workers_cpus=list(range(n_parallel)),
alternating=False)
lr = 1e-3
po = np.array([1, 0, 0, 1, 0], dtype=bool)
# Model kwargs
# model_kwargs = dict()
# model_kwargs = dict(hidden_sizes=[64, 64], shared_processor=False)
model_kwargs = dict(hidden_sizes=[64, 64],
rnn_type='gru',
rnn_size=256,
rnn_placement=1,
shared_processor=False,
layer_norm=True,
prev_action='All',
prev_reward='All')
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, shared_processor=False, use_interest=False, use_diversity=False, use_attention=False)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, use_interest=True, use_diversity=False,
# use_attention=False, rnn_type='gru', rnn_size=256, rnn_placement=1, shared_processor=False, layer_norm=True, prev_option=po)
sampler = SamplerCls(env, batch_T, max_decorrelation_steps=0)
# Samplers
# Algos (swapping out discount)
algo = A2C(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# algo = A2OC(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# algo = PPO(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# algo = PPOC(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# Agents
# agent = PomdpFfAgent(model_kwargs=model_kwargs)
agent = PomdpRnnAgent(model_kwargs=model_kwargs)
# agent = PomdpOcFfAgent(model_kwargs=model_kwargs)
# agent = PomdpOcRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpOcRnnAgent(model_kwargs=model_kwargs)
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=1e6,
log_interval_steps=1e3,
affinity=affinity,
)
config = dict(env_id=env_id,
fomdp=fomdp,
algo_name=algo.__class__.__name__,
learning_rate=lr,
sampler=sampler.__class__.__name__,
model=model_kwargs)
name = algo.NAME + '_' + env_id
log_dir = "pomdps"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(windx,
windy,
game="pong",
run_ID=0,
cuda_idx=None,
sample_mode="serial",
n_parallel=2,
num_envs=2,
eval=False,
train_mask=[True, True],
wandb_log=False,
save_models_to_wandb=False,
log_interval_steps=1e5,
alt_train=False,
n_steps=50e6,
max_episode_length=np.inf,
b_size=5,
max_decor_steps=10):
# def envs:
# player_model_kwargs = dict(hidden_sizes=[32], lstm_size=16, nonlinearity=torch.nn.ReLU, normalize_observation=False,
# norm_obs_clip=10, norm_obs_var_clip=1e-6)
# observer_model_kwargs = dict(hidden_sizes=[128], lstm_size=16, nonlinearity=torch.nn.ReLU,
# normalize_observation=False, norm_obs_clip=10, norm_obs_var_clip=1e-6)
player_reward_shaping = None
observer_reward_shaping = None
window_size = np.asarray([windx, windy])
affinity = dict(cuda_idx=cuda_idx, workers_cpus=list(range(n_parallel)))
gpu_cpu = "CPU" if cuda_idx is None else f"GPU {cuda_idx}"
if sample_mode == "serial":
alt = False
Sampler = SerialSampler # (Ignores workers_cpus.)
if eval:
eval_collector_cl = SerialEvalCollector
else:
eval_collector_cl = None
print(f"Using serial sampler, {gpu_cpu} for sampling and optimizing.")
elif sample_mode == "cpu":
alt = False
Sampler = CpuSampler
if eval:
eval_collector_cl = CpuEvalCollector
else:
eval_collector_cl = None
print(
f"Using CPU parallel sampler (agent in workers), {gpu_cpu} for optimizing."
)
env_kwargs = dict(env_name=game,
window_size=window_size,
player_reward_shaping=player_reward_shaping,
observer_reward_shaping=observer_reward_shaping,
max_episode_length=max_episode_length)
if eval:
eval_env_kwargs = env_kwargs
eval_max_steps = 1e4
num_eval_envs = num_envs
else:
eval_env_kwargs = None
eval_max_steps = None
num_eval_envs = 0
sampler = Sampler(
EnvCls=CWTO_EnvWrapperAtari,
env_kwargs=env_kwargs,
batch_T=b_size,
batch_B=num_envs,
max_decorrelation_steps=max_decor_steps,
eval_n_envs=num_eval_envs,
eval_CollectorCls=eval_collector_cl,
eval_env_kwargs=eval_env_kwargs,
eval_max_steps=eval_max_steps,
)
player_algo = A2C()
observer_algo = A2C()
player = AtariLstmAgent() #model_kwargs=player_model_kwargs)
observer = CWTO_AtariLstmAgent() #model_kwargs=observer_model_kwargs)
agent = CWTO_AgentWrapper(player, observer, alt=alt, train_mask=train_mask)
if eval:
RunnerCl = MinibatchRlEval
else:
RunnerCl = MinibatchRl
runner = RunnerCl(player_algo=player_algo,
observer_algo=observer_algo,
agent=agent,
sampler=sampler,
n_steps=n_steps,
log_interval_steps=log_interval_steps,
affinity=affinity,
wandb_log=wandb_log,
alt_train=alt_train)
config = dict(domain=game)
log_dir = os.getcwd() + "/cwto_logs/" + game
with logger_context(log_dir, run_ID, game, config):
runner.train()
if save_models_to_wandb:
agent.save_models_to_wandb()
def build_and_train(env_id="catch/0", run_ID=0, cuda_idx=None, n_parallel=6):
EnvCls = BSuiteEnv
n_episodes = 1e4
env_args = dict(id=env_id)
affinity = dict(cuda_idx=cuda_idx,
workers_cpus=list(range(n_parallel)),
alternating=True)
lr = 1e-3
# Model kwargs
# model_kwargs = dict()
# model_kwargs = dict(hidden_sizes=[64, 64])
model_kwargs = dict(hidden_sizes=[64, 64],
rnn_type='gru',
rnn_size=256,
rnn_placement=1,
shared_processor=True,
layer_norm=True)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, use_interest=False, use_diversity=False, use_attention=False)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, use_interest=False, use_diversity=False,
# use_attention=False, rnn_type='gru', rnn_size=128)
# Samplers
# sampler = AlternatingSampler(
# EnvCls=EnvCls,
# env_kwargs=env_args,
# eval_env_kwargs=env_args,
# batch_T=20, # One time-step per sampler iteration.
# batch_B=30, # One environment (i.e. sampler Batch dimension).
# max_decorrelation_steps=0,
# eval_n_envs=5,
# eval_max_steps=int(25e3),
# eval_max_trajectories=30
# )
#
sampler = SerialSampler(
EnvCls=EnvCls,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=32, # One time-step per sampler iteration.
batch_B=1, # One environment (i.e. sampler Batch dimension).
max_decorrelation_steps=0,
# eval_n_envs=2,
# eval_max_steps=int(51e2),
# eval_max_trajectories=5,
)
# Algos (swapping out discount)
algo = A2C(learning_rate=lr, clip_grad_norm=2.)
# algo = A2OC(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# Agents
# agent = BsuiteFfAgent(model_kwargs=model_kwargs)
agent = BsuiteRnnAgent(model_kwargs=model_kwargs)
# agent = BsuiteOcFfAgent(model_kwargs=model_kwargs)
# agent = BsuiteOcRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingBsuiteRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingBsuiteRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingBsuiteOcRnnAgent(model_kwargs=model_kwargs)
runner = EpisodicRlRunner(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=1e8,
n_episodes=1e4,
log_interval_steps=1e3,
affinity=affinity,
)
config = dict(env_id=env_id,
algo_name=algo.__class__.__name__,
learning_rate=lr)
name = algo.NAME + '_' + env_id
log_dir = "Bsuites"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(env_id="POMDP-hallway-episodic-v0", run_ID=0, cuda_idx=None, n_parallel=1, fomdp=False):
EnvCls = pomdp_interface
env_args = dict(fomdp=fomdp, id=env_id, time_limit=100)
test_instance = EnvCls(**env_args)
gamma = test_instance.discount
affinity = dict(cuda_idx=cuda_idx, workers_cpus=list(range(n_parallel)), alternating=True)
lr = 1e-3
po = np.array([1,0,0,1,0], dtype=bool)
# Model kwargs
# model_kwargs = dict()
# model_kwargs = dict(hidden_sizes=[64, 64], shared_processor=False)
model_kwargs = dict(hidden_sizes=[64, 64], rnn_type='gru', rnn_size=256, rnn_placement=1, shared_processor=False, layer_norm=True, prev_action=3, prev_reward=3)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, shared_processor=False, use_interest=False, use_diversity=False, use_attention=False)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, use_interest=True, use_diversity=False,
# use_attention=False, rnn_type='gru', rnn_size=256, rnn_placement=1, shared_processor=False, layer_norm=True, prev_option=po)
# Samplers
sampler = GpuSampler(
EnvCls=EnvCls,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=20, # One time-step per sampler iteration.
batch_B=30, # One environment (i.e. sampler Batch dimension).
max_decorrelation_steps=0,
eval_n_envs=5,
eval_max_steps=int(25e3),
eval_max_trajectories=30
)
# sampler = AlternatingSampler(
# EnvCls=EnvCls,
# env_kwargs=env_args,
# eval_env_kwargs=env_args,
# batch_T=20, # One time-step per sampler iteration.
# batch_B=30, # One environment (i.e. sampler Batch dimension).
# max_decorrelation_steps=0,
# eval_n_envs=5,
# eval_max_steps=int(25e3),
# eval_max_trajectories=30
# )
#
# sampler = SerialSampler(
# EnvCls=EnvCls,
# env_kwargs=env_args,
# eval_env_kwargs=env_args,
# batch_T=20, # One time-step per sampler iteration.
# batch_B=30, # One environment (i.e. sampler Batch dimension).
# max_decorrelation_steps=0,
# # eval_n_envs=2,
# # eval_max_steps=int(51e2),
# # eval_max_trajectories=5,
# )
# Algos (swapping out discount)
algo = A2C(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# algo = A2OC(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# algo = PPO(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# algo = PPOC(discount=gamma, learning_rate=lr, clip_grad_norm=2.)
# Agents
# agent = PomdpFfAgent(model_kwargs=model_kwargs)
agent = PomdpRnnAgent(model_kwargs=model_kwargs)
# agent = PomdpOcFfAgent(model_kwargs=model_kwargs)
# agent = PomdpOcRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpOcRnnAgent(model_kwargs=model_kwargs)
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=1e6,
log_interval_steps=1e3,
affinity=affinity,
)
config = dict(env_id=env_id, fomdp=fomdp, algo_name=algo.__class__.__name__, learning_rate=lr, sampler=sampler.__class__.__name__, model=model_kwargs)
name = algo.NAME + '_' + env_id
log_dir = "pomdps"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def build_and_train(env_id="Taxi-v3", run_ID=0, cuda_idx=None, n_parallel=6, fomdp=False):
EnvCls = gym_make if fomdp else make_po_taxi
env_args = dict(id=env_id)
affinity = dict(cuda_idx=cuda_idx, workers_cpus=list(range(n_parallel)), alternating=True)
lr = 1e-3
# Model kwargs
# model_kwargs = dict()
model_kwargs = dict(hidden_sizes=[64, 64])
# model_kwargs = dict(hidden_sizes=[64, 64], rnn_type='gru', rnn_size=128)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, use_interest=False, use_diversity=False, use_attention=False)
# model_kwargs = dict(hidden_sizes=[64, 64], option_size=4, use_interest=False, use_diversity=False,
# use_attention=False, rnn_type='gru', rnn_size=128)
# Samplers
# sampler = AlternatingSampler(
# EnvCls=EnvCls,
# env_kwargs=env_args,
# eval_env_kwargs=env_args,
# batch_T=20, # One time-step per sampler iteration.
# batch_B=30, # One environment (i.e. sampler Batch dimension).
# max_decorrelation_steps=0,
# eval_n_envs=5,
# eval_max_steps=int(25e3),
# eval_max_trajectories=30
# )
#
sampler = SerialSampler(
EnvCls=EnvCls,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=20, # One time-step per sampler iteration.
batch_B=30, # One environment (i.e. sampler Batch dimension).
max_decorrelation_steps=0,
# eval_n_envs=2,
# eval_max_steps=int(51e2),
# eval_max_trajectories=5,
)
# Algos (swapping out discount)
algo = A2C(discount=0.9, learning_rate=lr, entropy_loss_coeff=0.01, normalize_rewards='reward')
# algo = A2OC(discount=0.618, learning_rate=lr, clip_grad_norm=2.)
# Agents
agent = PomdpFfAgent(model_kwargs=model_kwargs)
# agent = PomdpRnnAgent(model_kwargs=model_kwargs)
# agent = PomdpOcFfAgent(model_kwargs=model_kwargs)
# agent = PomdpOcRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpRnnAgent(model_kwargs=model_kwargs)
# agent = AlternatingPomdpOcRnnAgent(model_kwargs=model_kwargs)
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=1e6,
log_interval_steps=1e3,
affinity=affinity,
)
config = dict(env_id=env_id, fomdp=fomdp, algo_name=algo.__class__.__name__, learning_rate=lr)
name = algo.NAME + '_' + env_id
log_dir = "pomdps"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def start_experiment(args):
args_json = json.dumps(vars(args), indent=4)
if not os.path.isdir(args.log_dir):
os.makedirs(args.log_dir)
with open(args.log_dir + '/arguments.json', 'w') as jsonfile:
jsonfile.write(args_json)
with open(args.log_dir + '/git.txt', 'w') as git_file:
branch = subprocess.check_output(['git', 'rev-parse', '--abbrev-ref', 'HEAD']).strip().decode('utf-8')
commit = subprocess.check_output(['git', 'rev-parse', 'HEAD']).strip().decode('utf-8')
git_file.write('{}/{}'.format(branch, commit))
config = dict(env_id=args.env)
if args.sample_mode == 'gpu':
# affinity = dict(num_gpus=args.num_gpus, workers_cpus=list(range(args.num_cpus)))
if args.num_gpus > 0:
# import ipdb; ipdb.set_trace()
affinity = make_affinity(
run_slot=0,
n_cpu_core=args.num_cpus, # Use 16 cores across all experiments.
n_gpu=args.num_gpus, # Use 8 gpus across all experiments.
# contexts_per_gpu=2,
# hyperthread_offset=72, # If machine has 24 cores.
# n_socket=2, # Presume CPU socket affinity to lower/upper half GPUs.
gpu_per_run=args.gpu_per_run, # How many GPUs to parallelize one run across.
# cpu_per_run=1,
)
print('Make multi-gpu affinity')
else:
affinity = dict(cuda_idx=0, workers_cpus=list(range(args.num_cpus)))
os.environ['CUDA_VISIBLE_DEVICES'] = str(0)
else:
affinity = dict(workers_cpus=list(range(args.num_cpus)))
# potentially reload models
initial_optim_state_dict = None
initial_model_state_dict = None
if args.pretrain != 'None':
os.system(f"find {args.log_dir} -name '*.json' -delete") # clean up json files for video recorder
checkpoint = torch.load(os.path.join(_RESULTS_DIR, args.pretrain, 'params.pkl'))
initial_optim_state_dict = checkpoint['optimizer_state_dict']
initial_model_state_dict = checkpoint['agent_state_dict']
# ----------------------------------------------------- POLICY ----------------------------------------------------- #
model_args = dict(curiosity_kwargs=dict(curiosity_alg=args.curiosity_alg), curiosity_step_kwargs=dict())
if args.curiosity_alg =='icm':
model_args['curiosity_kwargs']['feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs']['prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs']['forward_loss_wt'] = args.forward_loss_wt
model_args['curiosity_kwargs']['forward_model'] = args.forward_model
model_args['curiosity_kwargs']['feature_space'] = args.feature_space
elif args.curiosity_alg == 'micm':
model_args['curiosity_kwargs']['feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs']['prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs']['forward_loss_wt'] = args.forward_loss_wt
model_args['curiosity_kwargs']['forward_model'] = args.forward_model
model_args['curiosity_kwargs']['ensemble_mode'] = args.ensemble_mode
model_args['curiosity_kwargs']['device'] = args.sample_mode
elif args.curiosity_alg == 'disagreement':
model_args['curiosity_kwargs']['feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['ensemble_size'] = args.ensemble_size
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs']['prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs']['forward_loss_wt'] = args.forward_loss_wt
model_args['curiosity_kwargs']['device'] = args.sample_mode
model_args['curiosity_kwargs']['forward_model'] = args.forward_model
elif args.curiosity_alg == 'ndigo':
model_args['curiosity_kwargs']['feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['pred_horizon'] = args.pred_horizon
model_args['curiosity_kwargs']['prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs']['device'] = args.sample_mode
elif args.curiosity_alg == 'rnd':
model_args['curiosity_kwargs']['feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs']['drop_probability'] = args.drop_probability
model_args['curiosity_kwargs']['gamma'] = args.discount
model_args['curiosity_kwargs']['device'] = args.sample_mode
if args.curiosity_alg != 'none':
model_args['curiosity_step_kwargs']['curiosity_step_minibatches'] = args.curiosity_step_minibatches
if args.env in _MUJOCO_ENVS:
if args.lstm:
agent = MujocoLstmAgent(initial_model_state_dict=initial_model_state_dict)
else:
agent = MujocoFfAgent(initial_model_state_dict=initial_model_state_dict)
else:
if args.lstm:
agent = AtariLstmAgent(
initial_model_state_dict=initial_model_state_dict,
model_kwargs=model_args,
no_extrinsic=args.no_extrinsic,
dual_model=args.dual_model,
)
else:
agent = AtariFfAgent(initial_model_state_dict=initial_model_state_dict,
model_kwargs=model_args,
no_extrinsic=args.no_extrinsic,
dual_model=args.dual_model)
# ----------------------------------------------------- LEARNING ALG ----------------------------------------------------- #
if args.alg == 'ppo':
algo = PPO(
discount=args.discount,
learning_rate=args.lr,
value_loss_coeff=args.v_loss_coeff,
entropy_loss_coeff=args.entropy_loss_coeff,
OptimCls=torch.optim.Adam,
optim_kwargs=None,
clip_grad_norm=args.grad_norm_bound,
initial_optim_state_dict=initial_optim_state_dict, # is None is not reloading a checkpoint
gae_lambda=args.gae_lambda,
minibatches=args.minibatches, # if recurrent: batch_B needs to be at least equal, if not recurrent: batch_B*batch_T needs to be at least equal to this
epochs=args.epochs,
ratio_clip=args.ratio_clip,
linear_lr_schedule=args.linear_lr,
normalize_advantage=args.normalize_advantage,
normalize_reward=args.normalize_reward,
curiosity_type=args.curiosity_alg,
policy_loss_type=args.policy_loss_type
)
elif args.alg == 'a2c':
algo = A2C(
discount=args.discount,
learning_rate=args.lr,
value_loss_coeff=args.v_loss_coeff,
entropy_loss_coeff=args.entropy_loss_coeff,
OptimCls=torch.optim.Adam,
optim_kwargs=None,
clip_grad_norm=args.grad_norm_bound,
initial_optim_state_dict=initial_optim_state_dict,
gae_lambda=args.gae_lambda,
normalize_advantage=args.normalize_advantage
)
# ----------------------------------------------------- SAMPLER ----------------------------------------------------- #
# environment setup
traj_info_cl = TrajInfo # environment specific - potentially overriden below
if 'mario' in args.env.lower():
env_cl = mario_make
env_args = dict(
game=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=args.normalize_obs,
normalize_obs_steps=10000
)
elif args.env in _PYCOLAB_ENVS:
env_cl = deepmind_make
traj_info_cl = PycolabTrajInfo
env_args = dict(
game=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=args.normalize_obs,
normalize_obs_steps=10000,
log_heatmaps=args.log_heatmaps,
logdir=args.log_dir,
obs_type=args.obs_type,
grayscale=args.grayscale,
max_steps_per_episode=args.max_episode_steps
)
elif args.env in _MUJOCO_ENVS:
env_cl = gym_make
env_args = dict(
id=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=False,
normalize_obs_steps=10000
)
elif args.env in _ATARI_ENVS:
env_cl = AtariEnv
traj_info_cl = AtariTrajInfo
env_args = dict(
game=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=args.normalize_obs,
normalize_obs_steps=10000,
downsampling_scheme='classical',
record_freq=args.record_freq,
record_dir=args.log_dir,
horizon=args.max_episode_steps,
score_multiplier=args.score_multiplier,
repeat_action_probability=args.repeat_action_probability,
fire_on_reset=args.fire_on_reset
)
if args.sample_mode == 'gpu':
if args.lstm:
collector_class = GpuWaitResetCollector
else:
collector_class = GpuResetCollector
sampler = GpuSampler(
EnvCls=env_cl,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=args.timestep_limit,
batch_B=args.num_envs,
max_decorrelation_steps=0,
TrajInfoCls=traj_info_cl,
eval_n_envs=args.eval_envs,
eval_max_steps=args.eval_max_steps,
eval_max_trajectories=args.eval_max_traj,
record_freq=args.record_freq,
log_dir=args.log_dir,
CollectorCls=collector_class
)
else:
if args.lstm:
collector_class = CpuWaitResetCollector
else:
collector_class = CpuResetCollector
sampler = CpuSampler(
EnvCls=env_cl,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=args.timestep_limit, # timesteps in a trajectory episode
batch_B=args.num_envs, # environments distributed across workers
max_decorrelation_steps=0,
TrajInfoCls=traj_info_cl,
eval_n_envs=args.eval_envs,
eval_max_steps=args.eval_max_steps,
eval_max_trajectories=args.eval_max_traj,
record_freq=args.record_freq,
log_dir=args.log_dir,
CollectorCls=collector_class
)
# ----------------------------------------------------- RUNNER ----------------------------------------------------- #
if args.eval_envs > 0:
runner = (MinibatchRlEval if args.num_gpus <= 1 else SyncRlEval)(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=args.iterations,
affinity=affinity,
log_interval_steps=args.log_interval,
log_dir=args.log_dir,
pretrain=args.pretrain
)
else:
runner = (MinibatchRl if args.num_gpus <= 1 else SyncRl)(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=args.iterations,
affinity=affinity,
log_interval_steps=args.log_interval,
log_dir=args.log_dir,
pretrain=args.pretrain
)
with logger_context(args.log_dir, config, snapshot_mode="last", use_summary_writer=True):
runner.train()
def build_and_train(env_id="GridEnv-v1",
run_ID=0,
cuda_idx=None,
sample_mode="serial",
n_parallel=2,
args={}):
affinity = dict(cuda_idx=cuda_idx, workers_cpus=list(range(n_parallel)))
gpu_cpu = "CPU" if cuda_idx is None else f"GPU {cuda_idx}"
if sample_mode == "serial":
Sampler = SerialSampler # (Ignores workers_cpus.)
print(f"Using serial sampler, {gpu_cpu} for sampling and optimizing.")
elif sample_mode == "cpu":
Sampler = CpuSampler
print(
f"Using CPU parallel sampler (agent in workers), {gpu_cpu} for optimizing."
)
elif sample_mode == "gpu":
Sampler = GpuSampler
print(
f"Using GPU parallel sampler (agent in master), {gpu_cpu} for sampling and optimizing."
)
elif sample_mode == "alternating":
Sampler = AlternatingSampler
affinity["workers_cpus"] += affinity["workers_cpus"] # (Double list)
affinity["alternating"] = True # Sampler will check for this.
print(
f"Using Alternating GPU parallel sampler, {gpu_cpu} for sampling and optimizing."
)
sampler = Sampler(
EnvCls=gym_make,
env_kwargs=dict(id=env_id, stochastic=True, p=0.15, size=(7, 7)),
batch_T=5, # 5 time-steps per sampler iteration.
batch_B=16,
max_decorrelation_steps=1000,
eval_n_envs=0,
)
algo = A2C(learning_rate=args.lr)
agentCls, agent_basis = get_agent_cls_grid(args.network)
agent = agentCls(
model_kwargs={
'basis': agent_basis,
'channels': args.channels,
'kernel_sizes': args.filters,
'paddings': args.paddings,
'fc_sizes': args.fcs,
'strides': args.strides
})
runner = MinibatchRl(
algo=algo,
agent=agent,
sampler=sampler,
n_steps=2e6,
log_interval_steps=10e3,
affinity=affinity,
)
config = dict(env_id=env_id,
lr=args.lr,
network=args.network,
fcs=str(args.fcs),
channels=args.channels,
strides=args.strides,
paddings=args.paddings)
name = f"{args.folder}_{args.network}"
log_dir = f"{args.folder}_{args.network}"
with logger_context(log_dir, run_ID, name, config):
runner.train()
def start_experiment(args):
args_json = json.dumps(vars(args), indent=4)
if not os.path.isdir(args.log_dir):
os.makedirs(args.log_dir)
with open(args.log_dir + '/arguments.json', 'w') as jsonfile:
jsonfile.write(args_json)
config = dict(env_id=args.env)
if args.sample_mode == 'gpu':
assert args.num_gpus > 0
affinity = dict(cuda_idx=0, workers_cpus=list(range(args.num_cpus)))
os.environ['CUDA_VISIBLE_DEVICES'] = str(0)
else:
affinity = dict(workers_cpus=list(range(args.num_cpus)))
# potentially reload models
initial_optim_state_dict = None
initial_model_state_dict = None
if args.pretrain != 'None':
os.system(f"find {args.log_dir} -name '*.json' -delete"
) # clean up json files for video recorder
checkpoint = torch.load(
os.path.join(_RESULTS_DIR, args.pretrain, 'params.pkl'))
initial_optim_state_dict = checkpoint['optimizer_state_dict']
initial_model_state_dict = checkpoint['agent_state_dict']
# ----------------------------------------------------- POLICY ----------------------------------------------------- #
model_args = dict(curiosity_kwargs=dict(curiosity_alg=args.curiosity_alg))
if args.curiosity_alg == 'icm':
model_args['curiosity_kwargs'][
'feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs'][
'prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs'][
'forward_loss_wt'] = args.forward_loss_wt
elif args.curiosity_alg == 'disagreement':
model_args['curiosity_kwargs'][
'feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['ensemble_size'] = args.ensemble_size
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs'][
'prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs'][
'forward_loss_wt'] = args.forward_loss_wt
model_args['curiosity_kwargs']['device'] = args.sample_mode
elif args.curiosity_alg == 'ndigo':
model_args['curiosity_kwargs'][
'feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs']['pred_horizon'] = args.pred_horizon
model_args['curiosity_kwargs']['batch_norm'] = args.batch_norm
model_args['curiosity_kwargs']['num_predictors'] = args.num_predictors
model_args['curiosity_kwargs']['device'] = args.sample_mode
elif args.curiosity_alg == 'rnd':
model_args['curiosity_kwargs'][
'feature_encoding'] = args.feature_encoding
model_args['curiosity_kwargs'][
'prediction_beta'] = args.prediction_beta
model_args['curiosity_kwargs'][
'drop_probability'] = args.drop_probability
model_args['curiosity_kwargs']['gamma'] = args.discount
model_args['curiosity_kwargs']['device'] = args.sample_mode
if args.env in _MUJOCO_ENVS:
if args.lstm:
agent = MujocoLstmAgent(
initial_model_state_dict=initial_model_state_dict)
else:
agent = MujocoFfAgent(
initial_model_state_dict=initial_model_state_dict)
else:
if args.lstm:
agent = AtariLstmAgent(
initial_model_state_dict=initial_model_state_dict,
model_kwargs=model_args,
no_extrinsic=args.no_extrinsic)
else:
agent = AtariFfAgent(
initial_model_state_dict=initial_model_state_dict)
# ----------------------------------------------------- LEARNING ALG ----------------------------------------------------- #
if args.alg == 'ppo':
if args.kernel_mu == 0.:
kernel_params = None
else:
kernel_params = (args.kernel_mu, args.kernel_sigma)
algo = PPO(
discount=args.discount,
learning_rate=args.lr,
value_loss_coeff=args.v_loss_coeff,
entropy_loss_coeff=args.entropy_loss_coeff,
OptimCls=torch.optim.Adam,
optim_kwargs=None,
clip_grad_norm=args.grad_norm_bound,
initial_optim_state_dict=
initial_optim_state_dict, # is None is not reloading a checkpoint
gae_lambda=args.gae_lambda,
minibatches=args.
minibatches, # if recurrent: batch_B needs to be at least equal, if not recurrent: batch_B*batch_T needs to be at least equal to this
epochs=args.epochs,
ratio_clip=args.ratio_clip,
linear_lr_schedule=args.linear_lr,
normalize_advantage=args.normalize_advantage,
normalize_reward=args.normalize_reward,
kernel_params=kernel_params,
curiosity_type=args.curiosity_alg)
elif args.alg == 'a2c':
algo = A2C(discount=args.discount,
learning_rate=args.lr,
value_loss_coeff=args.v_loss_coeff,
entropy_loss_coeff=args.entropy_loss_coeff,
OptimCls=torch.optim.Adam,
optim_kwargs=None,
clip_grad_norm=args.grad_norm_bound,
initial_optim_state_dict=initial_optim_state_dict,
gae_lambda=args.gae_lambda,
normalize_advantage=args.normalize_advantage)
# ----------------------------------------------------- SAMPLER ----------------------------------------------------- #
# environment setup
traj_info_cl = TrajInfo # environment specific - potentially overriden below
if 'mario' in args.env.lower():
env_cl = mario_make
env_args = dict(game=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=args.normalize_obs,
normalize_obs_steps=10000)
elif 'deepmind' in args.env.lower(): # pycolab deepmind environments
env_cl = deepmind_make
traj_info_cl = PycolabTrajInfo
env_args = dict(game=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=args.normalize_obs,
normalize_obs_steps=10000,
log_heatmaps=args.log_heatmaps,
logdir=args.log_dir,
obs_type=args.obs_type,
max_steps_per_episode=args.max_episode_steps)
elif args.env in _MUJOCO_ENVS:
env_cl = gym_make
env_args = dict(id=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=False,
normalize_obs_steps=10000)
elif args.env in _ATARI_ENVS:
env_cl = AtariEnv
traj_info_cl = AtariTrajInfo
env_args = dict(
game=args.env,
no_extrinsic=args.no_extrinsic,
no_negative_reward=args.no_negative_reward,
normalize_obs=args.normalize_obs,
normalize_obs_steps=10000,
downsampling_scheme='classical',
record_freq=args.record_freq,
record_dir=args.log_dir,
horizon=args.max_episode_steps,
)
if args.sample_mode == 'gpu':
if args.lstm:
collector_class = GpuWaitResetCollector
else:
collector_class = GpuResetCollector
sampler = GpuSampler(EnvCls=env_cl,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=args.timestep_limit,
batch_B=args.num_envs,
max_decorrelation_steps=0,
TrajInfoCls=traj_info_cl,
eval_n_envs=args.eval_envs,
eval_max_steps=args.eval_max_steps,
eval_max_trajectories=args.eval_max_traj,
record_freq=args.record_freq,
log_dir=args.log_dir,
CollectorCls=collector_class)
else:
if args.lstm:
collector_class = CpuWaitResetCollector
else:
collector_class = CpuResetCollector
sampler = CpuSampler(
EnvCls=env_cl,
env_kwargs=env_args,
eval_env_kwargs=env_args,
batch_T=args.timestep_limit, # timesteps in a trajectory episode
batch_B=args.num_envs, # environments distributed across workers
max_decorrelation_steps=0,
TrajInfoCls=traj_info_cl,
eval_n_envs=args.eval_envs,
eval_max_steps=args.eval_max_steps,
eval_max_trajectories=args.eval_max_traj,
record_freq=args.record_freq,
log_dir=args.log_dir,
CollectorCls=collector_class)
# ----------------------------------------------------- RUNNER ----------------------------------------------------- #
if args.eval_envs > 0:
runner = MinibatchRlEval(algo=algo,
agent=agent,
sampler=sampler,
n_steps=args.iterations,
affinity=affinity,
log_interval_steps=args.log_interval,
log_dir=args.log_dir,
pretrain=args.pretrain)
else:
runner = MinibatchRl(algo=algo,
agent=agent,
sampler=sampler,
n_steps=args.iterations,
affinity=affinity,
log_interval_steps=args.log_interval,
log_dir=args.log_dir,
pretrain=args.pretrain)
with logger_context(args.log_dir,
config,
snapshot_mode="last",
use_summary_writer=True):
runner.train()
