fix dataset filter

deepspeech/exps/deepspeech2/model.py

@@ -145,11 +145,11 @@ class DeepSpeech2Trainer(Trainer):
         config.data.manifest = config.data.dev_manifest
         config.data.augmentation_config = ""
         config.data.min_input_len = 0.0  # second
-        config.data.max_input_len: 100.0  # second
-        config.data.min_output_len: 0.0  # tokens
-        config.data.max_output_len: 400.0  # tokens
-        config.data.min_output_input_ratio: 0.00
-        config.data.max_output_input_ratio: 100.0
+        config.data.max_input_len = float('inf')  # second
+        config.data.min_output_len = 0.0  # tokens
+        config.data.max_output_len = float('inf')  # tokens
+        config.data.min_output_input_ratio = 0.00
+        config.data.max_output_input_ratio = float('inf')
         dev_dataset = ManifestDataset.from_config(config)
 
         if self.parallel:
@@ -330,11 +330,11 @@ class DeepSpeech2Tester(DeepSpeech2Trainer):
         config.data.keep_transcription_text = True
         config.data.augmentation_config = ""
         config.data.min_input_len = 0.0  # second
-        config.data.max_input_len: 100.0  # second
-        config.data.min_output_len: 0.0  # tokens
-        config.data.max_output_len: 400.0  # tokens
-        config.data.min_output_input_ratio: 0.00
-        config.data.max_output_input_ratio: 100.0
+        config.data.max_input_len = float('inf')  # second
+        config.data.min_output_len = 0.0  # tokens
+        config.data.max_output_len = float('inf')  # tokens
+        config.data.min_output_input_ratio = 0.00
+        config.data.max_output_input_ratio = float('inf')
         test_dataset = ManifestDataset.from_config(config)
 
         # return text ord id

deepspeech/exps/u2/model.py

@@ -215,14 +215,14 @@ class U2Trainer(Trainer):
         config.data.manifest = config.data.train_manifest
         train_dataset = ManifestDataset.from_config(config)
 
+        config.data.manifest = config.data.dev_manifest
         config.data.augmentation_config = ""
         config.data.min_input_len = 0.0  # second
-        config.data.max_input_len: 100.0  # second
-        config.data.min_output_len: 0.0  # tokens
-        config.data.max_output_len: 400.0  # tokens
-        config.data.min_output_input_ratio: 0.00
-        config.data.max_output_input_ratio: 100.0
-        config.data.manifest = config.data.dev_manifest
+        config.data.max_input_len = float('inf')  # second
+        config.data.min_output_len = 0.0  # tokens
+        config.data.max_output_len = float('inf')  # tokens
+        config.data.min_output_input_ratio = 0.00
+        config.data.max_output_input_ratio = float('inf')
         dev_dataset = ManifestDataset.from_config(config)
 
         collate_fn = SpeechCollator(keep_transcription_text=False)
@@ -257,15 +257,15 @@ class U2Trainer(Trainer):
             collate_fn=collate_fn)
 
         # test dataset, return raw text
+        config.data.manifest = config.data.test_manifest
         config.data.keep_transcription_text = True
         config.data.augmentation_config = ""
         config.data.min_input_len = 0.0  # second
-        config.data.max_input_len: 100.0  # second
-        config.data.min_output_len: 0.0  # tokens
-        config.data.max_output_len: 400.0  # tokens
-        config.data.min_output_input_ratio: 0.00
-        config.data.max_output_input_ratio: 100.0
-        config.data.manifest = config.data.test_manifest
+        config.data.max_input_len = float('inf')  # second
+        config.data.min_output_len = 0.0  # tokens
+        config.data.max_output_len = float('inf')  # tokens
+        config.data.min_output_input_ratio = 0.00
+        config.data.max_output_input_ratio = float('inf')
         test_dataset = ManifestDataset.from_config(config)
         # return text ord id
         self.test_loader = DataLoader(