def _collate_fn_train(batch):
# sort the batch by its feature length in a descending order
batch = sorted(batch, key=lambda sample: sample[1], reverse=True)
max_seqlength = batch[0][1]
feat_dim = batch[0][0].size(1)
minibatch_size = len(batch)
feats = torch.zeros(minibatch_size, max_seqlength, feat_dim)
feat_lengths = torch.zeros(minibatch_size, dtype=torch.int)
graph_list = []
utt_ids = []
max_num_transitions = 0
max_num_states = 0
for i in range(minibatch_size):
feat, length, utt_id, graph = batch[i]
feats[i, :length, :].copy_(feat)
utt_ids.append(utt_id)
feat_lengths[i] = length
graph_list.append(graph)
if graph.num_transitions > max_num_transitions:
max_num_transitions = graph.num_transitions
if graph.num_states > max_num_states:
max_num_states = graph.num_states
num_graphs = ChainGraphBatch(graph_list,
max_num_transitions=max_num_transitions,
max_num_states=max_num_states)
return feats, feat_lengths, utt_ids, num_graphs
def merge(key):
if key == "source":
return speech_utils.collate_frames([s[key] for s in samples], 0.0)
elif key == "target":
max_num_transitions = max(s["target"].num_transitions for s in samples)
max_num_states = max(s["target"].num_states for s in samples)
return ChainGraphBatch(
[s["target"] for s in samples],
max_num_transitions=max_num_transitions,
max_num_states=max_num_states,
)
else:
raise ValueError("Invalid key.")
def _collate_fn(batch):
# sort the batch by its feature length in a descending order
batch = sorted(batch, key=lambda sample: sample[0].size(0), reverse=True)
max_seqlength = batch[0][0].size(0)
feat_dim = batch[0][0].size(1)
minibatch_size = len(batch)
feats = torch.zeros(minibatch_size, max_seqlength, feat_dim)
graph_list = []
max_num_transitions = 0
max_num_states = 0
for x in range(minibatch_size):
sample = batch[x]
feat, graph = sample
feat_length = feat.size(0)
feats[x].narrow(0, 0, feat_length).copy_(feat)
graph_list.append(graph)
if graph.num_transitions > max_num_transitions:
max_num_transitions = graph.num_transitions
if graph.num_states > max_num_states:
max_num_states = graph.num_states
num_graphs = ChainGraphBatch(graph_list,
max_num_transitions=max_num_transitions,
max_num_states=max_num_states)
return feats, num_graphs
def __len__(self):
return len(self.samples)
if __name__ == '__main__':
from torch.utils.data import DataLoader
from pychain.loss import ChainLoss
feat_dir = '/export/b08/yshao/kaldi/egs/wsj/s5/data/train_si284_spe2e_hires'
tree_dir = '/export/b08/yshao/kaldi/egs/wsj/s5/exp/chain/e2e_tree'
trainset = ChainDataset(feat_dir, tree_dir)
trainloader = AudioDataLoader(trainset, batch_size=2, shuffle=True)
feat, graphs = next(iter(trainloader))
print(feat.size())
den_fst_path = '/export/b08/yshao/kaldi/egs/wsj/s5/exp/chain/e2e_tree/den.fst'
den_fst = simplefst.StdVectorFst.read(den_fst_path)
den_graph = ChainGraph(den_fst, initial='recursive')
print(den_graph.num_states)
den_graph_batch = ChainGraphBatch(den_graph, batch_size=2)
criterion = ChainLoss(den_graph_batch)
torch.manual_seed(1)
nnet_output = torch.randn(2, 10, 100) # (B, T, D)
nnet_output.requires_grad = True
obj = criterion(nnet_output, graphs)
obj.backward()
print(obj)
print(nnet_output.grad)
