snsd0805/vlnbert
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat: add NOT_FOUND action in rollout

Browse Source
This commit is contained in:
Ting-Jun Wang 2023-11-06 18:31:14 +08:00
parent 4936098b5e
commit 03a3e5b489
Signed by: snsd0805
GPG Key ID: 48D331A3D6160354
2 changed files with 70 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

74
r2r_src/agent.py
View File

@ -147,7 +147,7 @@ class Seq2SeqAgent(BaseAgent):
return Variable(torch.from_numpy(features), requires_grad=False).cuda() return Variable(torch.from_numpy(features), requires_grad=False).cuda()
def _candidate_variable(self, obs): def _candidate_variable(self, obs):
candidate_leng = [len(ob['candidate']) + 1 for ob in obs] # +1 is for the end candidate_leng = [len(ob['candidate']) + 2 for ob in obs] # +1 is for the end
candidate_feat = np.zeros((len(obs), max(candidate_leng), self.feature_size + args.angle_feat_size), dtype=np.float32) candidate_feat = np.zeros((len(obs), max(candidate_leng), self.feature_size + args.angle_feat_size), dtype=np.float32)
# Note: The candidate_feat at len(ob['candidate']) is the feature for the END # Note: The candidate_feat at len(ob['candidate']) is the feature for the END
@ -155,6 +155,8 @@ class Seq2SeqAgent(BaseAgent):
for i, ob in enumerate(obs): for i, ob in enumerate(obs):
for j, cc in enumerate(ob['candidate']): for j, cc in enumerate(ob['candidate']):
candidate_feat[i, j, :] = cc['feature'] candidate_feat[i, j, :] = cc['feature']
candidate_feat[i, len(ob['candidate']), :] = np.zeros(self.feature_size+args.angle_feat_size, dtype=np.float32) # <STOP>
candidate_feat[i, len(ob['candidate'])+1, :] = np.ones(self.feature_size+args.angle_feat_size, dtype=np.float32) # <NOT_FOUND>
return torch.from_numpy(candidate_feat).cuda(), candidate_leng return torch.from_numpy(candidate_feat).cuda(), candidate_leng
@ -186,10 +188,13 @@ class Seq2SeqAgent(BaseAgent):
break break
else: # Stop here else: # Stop here
assert ob['teacher'] == ob['viewpoint'] # The teacher action should be "STAY HERE" assert ob['teacher'] == ob['viewpoint'] # The teacher action should be "STAY HERE"
a[i] = len(ob['candidate']) if ob['found']:
a[i] = len(ob['candidate'])
else:
a[i] = len(ob['candidate'])+1
return torch.from_numpy(a).cuda() return torch.from_numpy(a).cuda()
def make_equiv_action(self, a_t, perm_obs, perm_idx=None, traj=None): def make_equiv_action(self, a_t, perm_obs, perm_idx=None, traj=None, found=None):
""" """
Interface between Panoramic view and Egocentric view Interface between Panoramic view and Egocentric view
It will convert the action panoramic view action a_t to equivalent egocentric view actions for the simulator It will convert the action panoramic view action a_t to equivalent egocentric view actions for the simulator
@ -205,7 +210,7 @@ class Seq2SeqAgent(BaseAgent):
for i, idx in enumerate(perm_idx): for i, idx in enumerate(perm_idx):
action = a_t[i] action = a_t[i]
if action != -1: # -1 is the <stop> action if action != -1 and action != -2: # -1 is the <stop> action
select_candidate = perm_obs[i]['candidate'][action] select_candidate = perm_obs[i]['candidate'][action]
src_point = perm_obs[i]['viewIndex'] src_point = perm_obs[i]['viewIndex']
trg_point = select_candidate['pointId'] trg_point = select_candidate['pointId']
@ -228,6 +233,10 @@ class Seq2SeqAgent(BaseAgent):
# print("action: {} view_index: {}".format(action, state.viewIndex)) # print("action: {} view_index: {}".format(action, state.viewIndex))
if traj is not None: if traj is not None:
traj[i]['path'].append((state.location.viewpointId, state.heading, state.elevation)) traj[i]['path'].append((state.location.viewpointId, state.heading, state.elevation))
elif action == -1 or action == -2:
if found is not None:
found[i] = action
def rollout(self, train_ml=None, train_rl=True, reset=True): def rollout(self, train_ml=None, train_rl=True, reset=True):
""" """
@ -246,7 +255,7 @@ class Seq2SeqAgent(BaseAgent):
obs = np.array(self.env.reset()) obs = np.array(self.env.reset())
else: else:
obs = np.array(self.env._get_obs()) obs = np.array(self.env._get_obs())
batch_size = len(obs) batch_size = len(obs)
# Language input # Language input
@ -270,6 +279,8 @@ class Seq2SeqAgent(BaseAgent):
'instr_id': ob['instr_id'], 'instr_id': ob['instr_id'],
'path': [(ob['viewpoint'], ob['heading'], ob['elevation'])], 'path': [(ob['viewpoint'], ob['heading'], ob['elevation'])],
} for ob in perm_obs] } for ob in perm_obs]
found = [ None for _ in range(len(perm_obs)) ]
# Init the reward shaping # Init the reward shaping
last_dist = np.zeros(batch_size, np.float32) last_dist = np.zeros(batch_size, np.float32)
@ -293,6 +304,15 @@ class Seq2SeqAgent(BaseAgent):
for t in range(self.episode_len): for t in range(self.episode_len):
input_a_t, candidate_feat, candidate_leng = self.get_input_feat(perm_obs) input_a_t, candidate_feat, candidate_leng = self.get_input_feat(perm_obs)
'''
# show feature
for index, feat in enumerate(candidate_feat):
for ff in feat:
print(ff)
print(candidate_leng[index])
print()
'''
# the first [CLS] token, initialized by the language BERT, serves # the first [CLS] token, initialized by the language BERT, serves
@ -324,9 +344,22 @@ class Seq2SeqAgent(BaseAgent):
# Supervised training # Supervised training
target = self._teacher_action(perm_obs, ended) target = self._teacher_action(perm_obs, ended)
for i in perm_obs:
print(i['found'], end=' ')
ml_loss += self.criterion(logit, target) ml_loss += self.criterion(logit, target)
'''
for index, mask in enumerate(candidate_mask):
print(mask)
print(candidate_leng[index])
print(logit[index])
print(target[index])
print("\n\n")
'''
# Determine next model inputs # Determine next model inputs
if self.feedback == 'teacher': if self.feedback == 'teacher':
a_t = target # teacher forcing a_t = target # teacher forcing
elif self.feedback == 'argmax': elif self.feedback == 'argmax':
@ -344,15 +377,24 @@ class Seq2SeqAgent(BaseAgent):
else: else:
print(self.feedback) print(self.feedback)
sys.exit('Invalid feedback option') sys.exit('Invalid feedback option')
# Prepare environment action # Prepare environment action
# NOTE: Env action is in the perm_obs space # NOTE: Env action is in the perm_obs space
cpu_a_t = a_t.cpu().numpy() cpu_a_t = a_t.cpu().numpy()
for i, next_id in enumerate(cpu_a_t): for i, next_id in enumerate(cpu_a_t):
if next_id == (candidate_leng[i]-1) or next_id == args.ignoreid or ended[i]: # The last action is <end> if next_id == (args.ignoreid) or ended[i]:
cpu_a_t[i] = -1 # Change the <end> and ignore action to -1 cpu_a_t[i] = found[i]
elif next_id == (candidate_leng[i]-2):
cpu_a_t[i] = -1
elif next_id == (candidate_leng[i]-1):
cpu_a_t[i] = -2
print(cpu_a_t)
# Make action and get the new state # Make action and get the new state
self.make_equiv_action(cpu_a_t, perm_obs, perm_idx, traj) self.make_equiv_action(cpu_a_t, perm_obs, perm_idx, traj, found)
print(self.feedback, found)
obs = np.array(self.env._get_obs()) obs = np.array(self.env._get_obs())
perm_obs = obs[perm_idx] # Perm the obs for the resu perm_obs = obs[perm_idx] # Perm the obs for the resu
@ -376,6 +418,20 @@ class Seq2SeqAgent(BaseAgent):
if action_idx == -1: # If the action now is end if action_idx == -1: # If the action now is end
if dist[i] < 3.0: # Correct if dist[i] < 3.0: # Correct
reward[i] = 2.0 + ndtw_score[i] * 2.0 reward[i] = 2.0 + ndtw_score[i] * 2.0
if ob['found']:
reward[i] += 1
else:
reward[i] -= 2
else: # Incorrect
reward[i] = -2.0
elif action_idx == -2:
if dist[i] < 3.0:
reward[i] = 2.0 + ndtw_score[i] * 2.0
if ob['found']:
reward[i] -= 2
else:
reward[i] += 1
else: # Incorrect else: # Incorrect
reward[i] = -2.0 reward[i] = -2.0
else: # The action is not end else: # The action is not end
@ -399,6 +455,7 @@ class Seq2SeqAgent(BaseAgent):
# Update the finished actions # Update the finished actions
# -1 means ended or ignored (already ended) # -1 means ended or ignored (already ended)
ended[:] = np.logical_or(ended, (cpu_a_t == -1)) ended[:] = np.logical_or(ended, (cpu_a_t == -1))
ended[:] = np.logical_or(ended, (cpu_a_t == -2))
# Early exit if all ended # Early exit if all ended
if ended.all(): if ended.all():
@ -476,6 +533,7 @@ class Seq2SeqAgent(BaseAgent):
else: else:
self.losses.append(self.loss.item() / self.episode_len) # This argument is useless. self.losses.append(self.loss.item() / self.episode_len) # This argument is useless.
print('\n')
return traj return traj
def test(self, use_dropout=False, feedback='argmax', allow_cheat=False, iters=None): def test(self, use_dropout=False, feedback='argmax', allow_cheat=False, iters=None):

5
r2r_src/env.py
View File

@ -127,6 +127,7 @@ class R2RBatch():
new_item = dict(item) new_item = dict(item)
new_item['instr_id'] = '%s_%d' % (item['path_id'], j) new_item['instr_id'] = '%s_%d' % (item['path_id'], j)
new_item['instructions'] = instr new_item['instructions'] = instr
new_item['found'] = item['found'][j]
''' BERT tokenizer ''' ''' BERT tokenizer '''
instr_tokens = tokenizer.tokenize(instr) instr_tokens = tokenizer.tokenize(instr)
@ -328,6 +329,7 @@ class R2RBatch():
# [visual_feature, angle_feature] for views # [visual_feature, angle_feature] for views
feature = np.concatenate((feature, self.angle_feature[base_view_id]), -1) feature = np.concatenate((feature, self.angle_feature[base_view_id]), -1)
obs.append({ obs.append({
'instr_id' : item['instr_id'], 'instr_id' : item['instr_id'],
'scan' : state.scanId, 'scan' : state.scanId,
@ -341,7 +343,8 @@ class R2RBatch():
'instructions' : item['instructions'], 'instructions' : item['instructions'],
'teacher' : self._shortest_path_action(state, item['path'][-1]), 'teacher' : self._shortest_path_action(state, item['path'][-1]),
'gt_path' : item['path'], 'gt_path' : item['path'],
'path_id' : item['path_id'] 'path_id' : item['path_id'],
'found': item['found']
}) })
if 'instr_encoding' in item: if 'instr_encoding' in item:
obs[-1]['instr_encoding'] = item['instr_encoding'] obs[-1]['instr_encoding'] = item['instr_encoding']