Fix type annotations and minor issues (#515)

* fix some type annotations

* add missing range

* translator: check for compound inputs

* trigger event first

* remove extra event trigger

* another try to fix event trigger

xnmt/batchers.py

@@ -71,7 +71,7 @@ def sent_len(self):
     return sum(b.sent_len() for b in self.batches)
 
   def __iter__(self):
-    for i in self.batch_size():
+    for i in range(self.batch_size()):
       yield sent.CompoundSentence(sents=[b[i] for b in self.batches])
 
   def __getitem__(self, key):

xnmt/eval/tasks.py

@@ -39,11 +39,16 @@ class LossEvalTask(EvalTask, Serializable):
   yaml_tag = '!LossEvalTask'
 
   @serializable_init
-  def __init__(self, src_file: str, ref_file: Optional[str] = None, model: 'model_base.GeneratorModel' = Ref("model"),
+  def __init__(self,
+               src_file: Union[str, Sequence[str]],
+               ref_file: Optional[str] = None,
+               model: 'model_base.GeneratorModel' = Ref("model"),
                batcher: Batcher = Ref("train.batcher", default=bare(xnmt.batchers.SrcBatcher, batch_size=32)),
-               loss_calculator: LossCalculator = bare(MLELoss), max_src_len: Optional[int] = None,
+               loss_calculator: LossCalculator = bare(MLELoss),
+               max_src_len: Optional[int] = None,
                max_trg_len: Optional[int] = None,
-               loss_comb_method: str = Ref("exp_global.loss_comb_method", default="sum"), desc: Any = None):
+               loss_comb_method: str = Ref("exp_global.loss_comb_method", default="sum"),
+               desc: Any = None):
     self.model = model
     self.loss_calculator = loss_calculator
     self.src_file = src_file

xnmt/experiments.py

@@ -1,4 +1,4 @@
-from typing import List, Optional, Dict
+from typing import Any, Dict, List, Optional
 
 from xnmt.param_initializers import ParamInitializer, GlorotInitializer, ZeroInitializer
 from xnmt.settings import settings
@@ -48,7 +48,7 @@ def __init__(self,
                loss_comb_method: str = "sum",
                compute_report: bool = False,
                commandline_args: dict = {},
-               placeholders: Dict[str, str] = {}) -> None:
+               placeholders: Dict[str, Any] = {}) -> None:
     self.model_file = model_file
     self.log_file = log_file
     self.dropout = dropout

xnmt/models/translators.py

@@ -119,7 +119,6 @@ def shared_params(self):
 
 
   def _encode_src(self, src: Union[batchers.Batch, sent.Sentence]):
-    event_trigger.start_sent(src)
     embeddings = self.src_embedder.embed_sent(src)
     encoding = self.encoder.transduce(embeddings)
     final_state = self.encoder.get_final_states()
@@ -129,6 +128,8 @@ def _encode_src(self, src: Union[batchers.Batch, sent.Sentence]):
     return initial_state
 
   def calc_nll(self, src: Union[batchers.Batch, sent.Sentence], trg: Union[batchers.Batch, sent.Sentence]) -> dy.Expression:
+    event_trigger.start_sent(src)
+    if isinstance(src, batchers.CompoundBatch): src = src.batches[0]
     # Encode the sentence
     initial_state = self._encode_src(src)
 
@@ -201,8 +202,9 @@ def generate_search_output(self,
     if src.batch_size()!=1:
       raise NotImplementedError("batched decoding not implemented for DefaultTranslator. "
                                 "Specify inference batcher with batch size 1.")
-    # Generating outputs
     event_trigger.start_sent(src)
+    if isinstance(src, batchers.CompoundBatch): src = src.batches[0]
+    # Generating outputs
     cur_forced_trg = None
     src_sent = src[0]
     sent_mask = None
@@ -233,6 +235,7 @@ def generate(self,
     """
     assert src.batch_size() == 1
     search_outputs = self.generate_search_output(src, search_strategy, forced_trg_ids)
+    if isinstance(src, batchers.CompoundBatch): src = src.batches[0]
     sorted_outputs = sorted(search_outputs, key=lambda x: x.score[0], reverse=True)
     assert len(sorted_outputs) >= 1
     outputs = []