def handle(self, tdp: PreprocessedTrainingBatch) -> None:
assert isinstance(tdp.training_input, PreprocessedMemoryNetworkInput)
batch_size, _, _ = tdp.training_input.next_state.float_features.size()
tdp = PreprocessedTrainingBatch(
training_input=PreprocessedMemoryNetworkInput(
state=tdp.training_input.state,
action=tdp.training_input.action, # type: ignore
time_diff=torch.ones_like(
tdp.training_input.reward[torch.randperm(batch_size)]
).float(),
# shuffle the data
next_state=tdp.training_input.next_state._replace(
float_features=tdp.training_input.next_state.float_features[
torch.randperm(batch_size)
]
),
reward=tdp.training_input.reward[torch.randperm(batch_size)],
not_terminal=tdp.training_input.not_terminal[ # type: ignore
torch.randperm(batch_size)
],
step=None,
),
extras=ExtraData(),
)
losses = self.trainer_or_evaluator.train(tdp, batch_first=True)
self.results.append(losses)
def handle(self, tdp: TrainingBatch) -> None:
batch_size, _, _ = tdp.training_input.next_state.size()
tdp = TrainingBatch(
training_input=MemoryNetworkInput(
state=tdp.training_input.state,
action=tdp.training_input.action,
# shuffle the data
next_state=tdp.training_input.next_state[torch.randperm(
batch_size)],
reward=tdp.training_input.reward[torch.randperm(batch_size)],
not_terminal=tdp.training_input.not_terminal[torch.randperm(
batch_size)],
),
extras=ExtraData(),
)
losses = self.trainer_or_evaluator.train(tdp, batch_first=True)
self.results.append(losses)
def preprocess(self, batch) -> TrainingBatch:
state_features_dense, state_features_dense_presence = self.sparse_to_dense_processor(
batch["state_features"])
next_state_features_dense, next_state_features_dense_presence = self.sparse_to_dense_processor(
batch["next_state_features"])
mdp_ids = np.array(batch["mdp_id"]).reshape(-1, 1)
sequence_numbers = torch.tensor(batch["sequence_number"],
dtype=torch.int32).reshape(-1, 1)
rewards = torch.tensor(batch["reward"],
dtype=torch.float32).reshape(-1, 1)
time_diffs = torch.tensor(batch["time_diff"],
dtype=torch.int32).reshape(-1, 1)
if "action_probability" in batch:
propensities = torch.tensor(batch["action_probability"],
dtype=torch.float32).reshape(-1, 1)
else:
propensities = torch.ones(rewards.shape, dtype=torch.float32)
return TrainingBatch(
training_input=BaseInput(
state=FeatureVector(float_features=ValuePresence(
value=state_features_dense,
presence=state_features_dense_presence,
)),
next_state=FeatureVector(float_features=ValuePresence(
value=next_state_features_dense,
presence=next_state_features_dense_presence,
)),
reward=rewards,
time_diff=time_diffs,
),
extras=ExtraData(
mdp_id=mdp_ids,
sequence_number=sequence_numbers,
action_probability=propensities,
),
)
def evaluate(self, tdp: TrainingBatch):
""" Calculate feature importance: setting each state/action feature to
the mean value and observe loss increase. """
self.trainer.mdnrnn.mdnrnn.eval()
state_features = tdp.training_input.state.float_features
action_features = tdp.training_input.action.float_features # type: ignore
batch_size, seq_len, state_dim = state_features.size() # type: ignore
action_dim = action_features.size()[2] # type: ignore
action_feature_num = self.action_feature_num
state_feature_num = self.state_feature_num
feature_importance = torch.zeros(action_feature_num + state_feature_num)
orig_losses = self.trainer.get_loss(tdp, state_dim=state_dim, batch_first=True)
orig_loss = orig_losses["loss"].cpu().detach().item()
del orig_losses
action_feature_boundaries = self.sorted_action_feature_start_indices + [
action_dim
]
state_feature_boundaries = self.sorted_state_feature_start_indices + [state_dim]
for i in range(action_feature_num):
action_features = tdp.training_input.action.float_features.reshape( # type: ignore
(batch_size * seq_len, action_dim)
).data.clone()
# if actions are discrete, an action's feature importance is the loss
# increase due to setting all actions to this action
if self.discrete_action:
assert action_dim == action_feature_num
action_vec = torch.zeros(action_dim)
action_vec[i] = 1
action_features[:] = action_vec # type: ignore
# if actions are continuous, an action's feature importance is the loss
# increase due to masking this action feature to its mean value
else:
boundary_start, boundary_end = (
action_feature_boundaries[i],
action_feature_boundaries[i + 1],
)
action_features[ # type: ignore
:, boundary_start:boundary_end
] = self.compute_median_feature_value( # type: ignore
action_features[:, boundary_start:boundary_end] # type: ignore
)
action_features = action_features.reshape( # type: ignore
(batch_size, seq_len, action_dim)
) # type: ignore
new_tdp = TrainingBatch(
training_input=MemoryNetworkInput( # type: ignore
state=tdp.training_input.state,
action=FeatureVector( # type: ignore
float_features=action_features
), # type: ignore
next_state=tdp.training_input.next_state,
reward=tdp.training_input.reward,
not_terminal=tdp.training_input.not_terminal,
),
extras=ExtraData(),
)
losses = self.trainer.get_loss(
new_tdp, state_dim=state_dim, batch_first=True
)
feature_importance[i] = losses["loss"].cpu().detach().item() - orig_loss
del losses
for i in range(state_feature_num):
state_features = tdp.training_input.state.float_features.reshape( # type: ignore
(batch_size * seq_len, state_dim)
).data.clone()
boundary_start, boundary_end = (
state_feature_boundaries[i],
state_feature_boundaries[i + 1],
)
state_features[ # type: ignore
:, boundary_start:boundary_end
] = self.compute_median_feature_value(
state_features[:, boundary_start:boundary_end] # type: ignore
)
state_features = state_features.reshape( # type: ignore
(batch_size, seq_len, state_dim)
) # type: ignore
new_tdp = TrainingBatch(
training_input=MemoryNetworkInput( # type: ignore
state=FeatureVector(float_features=state_features), # type: ignore
action=tdp.training_input.action,
next_state=tdp.training_input.next_state,
reward=tdp.training_input.reward,
not_terminal=tdp.training_input.not_terminal,
),
extras=ExtraData(),
)
losses = self.trainer.get_loss(
new_tdp, state_dim=state_dim, batch_first=True
)
feature_importance[i + action_feature_num] = (
losses["loss"].cpu().detach().item() - orig_loss
)
del losses
self.trainer.mdnrnn.mdnrnn.train()
logger.info(
"**** Debug tool feature importance ****: {}".format(feature_importance)
)
return {"feature_loss_increase": feature_importance.numpy()}
def evaluate(self, tdp: TrainingBatch):
""" Calculate feature importance: setting each state/action feature to
the mean value and observe loss increase. """
self.trainer.mdnrnn.mdnrnn.eval()
state_features = tdp.training_input.state.float_features
action_features = tdp.training_input.action.float_features
batch_size, seq_len, state_dim = state_features.size()
action_dim = action_features.size()[2]
action_feature_num = self.feature_extractor.action_feature_num
state_feature_num = self.feature_extractor.state_feature_num
feature_importance = torch.zeros(action_feature_num +
state_feature_num)
orig_losses = self.trainer.get_loss(tdp,
state_dim=state_dim,
batch_first=True)
orig_loss = orig_losses["loss"].cpu().detach().item()
del orig_losses
action_feature_boundaries = (
self.feature_extractor.sorted_action_feature_start_indices +
[action_dim])
state_feature_boundaries = (
self.feature_extractor.sorted_state_feature_start_indices +
[state_dim])
for i in range(action_feature_num):
action_features = tdp.training_input.action.float_features.reshape(
(batch_size * seq_len, action_dim)).data.clone()
boundary_start, boundary_end = (
action_feature_boundaries[i],
action_feature_boundaries[i + 1],
)
action_features[:, boundary_start:
boundary_end] = action_features[:, boundary_start:
boundary_end].mean(
dim=0)
action_features = action_features.reshape(
(batch_size, seq_len, action_dim))
new_tdp = TrainingBatch(
training_input=MemoryNetworkInput(
state=tdp.training_input.state,
action=FeatureVector(float_features=action_features),
next_state=tdp.training_input.next_state,
reward=tdp.training_input.reward,
not_terminal=tdp.training_input.not_terminal,
),
extras=ExtraData(),
)
losses = self.trainer.get_loss(new_tdp,
state_dim=state_dim,
batch_first=True)
feature_importance[i] = losses["loss"].cpu().detach().item(
) - orig_loss
del losses
for i in range(state_feature_num):
state_features = tdp.training_input.state.float_features.reshape(
(batch_size * seq_len, state_dim)).data.clone()
boundary_start, boundary_end = (
state_feature_boundaries[i],
state_feature_boundaries[i + 1],
)
state_features[:, boundary_start:
boundary_end] = state_features[:, boundary_start:
boundary_end].mean(
dim=0)
state_features = state_features.reshape(
(batch_size, seq_len, state_dim))
new_tdp = TrainingBatch(
training_input=MemoryNetworkInput(
state=FeatureVector(float_features=state_features),
action=tdp.training_input.action,
next_state=tdp.training_input.next_state,
reward=tdp.training_input.reward,
not_terminal=tdp.training_input.not_terminal,
),
extras=ExtraData(),
)
losses = self.trainer.get_loss(new_tdp,
state_dim=state_dim,
batch_first=True)
feature_importance[i + action_feature_num] = (
losses["loss"].cpu().detach().item() - orig_loss)
del losses
self.trainer.mdnrnn.mdnrnn.train()
logger.info("**** Debug tool feature importance ****: {}".format(
feature_importance))
return {"feature_loss_increase": feature_importance.numpy()}