BertModel output dictionary access, specify cache path for model, req…

…uirements update

config_reader.py

@@ -4,7 +4,7 @@
 
 def process_configs(target, arg_parser):
     args, _ = arg_parser.parse_known_args()
-    ctx = mp.get_context('fork')
+    ctx = mp.get_context('spawn')
 
     for run_args, _run_config, _run_repeat in _yield_configs(arg_parser, args):
         p = ctx.Process(target=target, args=(run_args,))

requirements.txt

@@ -1,8 +1,7 @@
 Jinja2==2.10.3
 numpy==1.17.4
 tensorboardX==1.6
-torch==1.3.1
-torchvision==0.4.2
+torch==1.4.0
 tqdm==4.19.5
-transformers==2.2.0
-scikit-learn==0.21.3
+transformers==4.1.1
+scikit-learn==0.24.0

spert/evaluator.py

@@ -399,9 +399,9 @@ def _score(self, gt: List[List[Tuple]], pred: List[List[Tuple]], print_results:
 
     def _compute_metrics(self, gt_all, pred_all, types, print_results: bool = False):
         labels = [t.index for t in types]
-        per_type = prfs(gt_all, pred_all, labels=labels, average=None)
-        micro = prfs(gt_all, pred_all, labels=labels, average='micro')[:-1]
-        macro = prfs(gt_all, pred_all, labels=labels, average='macro')[:-1]
+        per_type = prfs(gt_all, pred_all, labels=labels, average=None, zero_division=0)
+        micro = prfs(gt_all, pred_all, labels=labels, average='micro', zero_division=0)[:-1]
+        macro = prfs(gt_all, pred_all, labels=labels, average='macro', zero_division=0)[:-1]
         total_support = sum(per_type[-1])
 
         if print_results:

spert/models.py

@@ -58,7 +58,7 @@ def _forward_train(self, encodings: torch.tensor, context_masks: torch.tensor, e
                        entity_sizes: torch.tensor, relations: torch.tensor, rel_masks: torch.tensor):
         # get contextualized token embeddings from last transformer layer
         context_masks = context_masks.float()
-        h = self.bert(input_ids=encodings, attention_mask=context_masks)[0]
+        h = self.bert(input_ids=encodings, attention_mask=context_masks)['last_hidden_state']
 
         batch_size = encodings.shape[0]
 
@@ -85,7 +85,7 @@ def _forward_eval(self, encodings: torch.tensor, context_masks: torch.tensor, en
                       entity_sizes: torch.tensor, entity_spans: torch.tensor, entity_sample_masks: torch.tensor):
         # get contextualized token embeddings from last transformer layer
         context_masks = context_masks.float()
-        h = self.bert(input_ids=encodings, attention_mask=context_masks)[0]
+        h = self.bert(input_ids=encodings, attention_mask=context_masks)['last_hidden_state']
 
         batch_size = encodings.shape[0]
         ctx_size = context_masks.shape[-1]

spert/spert_trainer.py

@@ -84,7 +84,8 @@ def train(self, train_path: str, valid_path: str, types_path: str, input_reader_
                                             max_pairs=self.args.max_pairs,
                                             prop_drop=self.args.prop_drop,
                                             size_embedding=self.args.size_embedding,
-                                            freeze_transformer=self.args.freeze_transformer)
+                                            freeze_transformer=self.args.freeze_transformer,
+                                            cache_dir=self.args.cache_path)
 
         # SpERT is currently optimized on a single GPU and not thoroughly tested in a multi GPU setup
         # If you still want to train SpERT on multiple GPUs, uncomment the following lines
@@ -161,7 +162,8 @@ def eval(self, dataset_path: str, types_path: str, input_reader_cls: BaseInputRe
                                             max_pairs=self.args.max_pairs,
                                             prop_drop=self.args.prop_drop,
                                             size_embedding=self.args.size_embedding,
-                                            freeze_transformer=self.args.freeze_transformer)
+                                            freeze_transformer=self.args.freeze_transformer,
+                                            cache_dir=self.args.cache_path)
 
         model.to(self._device)