Prefer literals over list and dict functions (#1454)

* d = dict() --> d = {}

* l = list() --> []

Author: lgeiger

tensor2tensor/data_generators/audio.py

@@ -60,7 +60,7 @@ def _collect_data(directory, input_ext, target_ext):
   #   if the datafile was "/path/to/datafile.wav" then the key would be
   #   "/path/to/datafile"
   # value: a pair of strings (input_filepath, target_filepath)
-  data_files = dict()
+  data_files = {}
   for root, _, filenames in os.walk(directory):
     input_files = [filename for filename in filenames if input_ext in filename]
     for input_filename in input_files:

tensor2tensor/data_generators/gym_env.py

@@ -640,7 +640,7 @@ def __init__(self, base_env_name=None, batch_size=1, grayscale=False,
 
     if self.should_derive_observation_space:
       with self._tf_graph.obj.as_default():
-        self._resize = dict()
+        self._resize = {}
         orig_height, orig_width = orig_observ_space.shape[:2]
         self._img_batch_t = _Noncopyable(tf.placeholder(
             dtype=tf.uint8, shape=(None, orig_height, orig_width, 3)))

tensor2tensor/data_generators/gym_env_test.py

@@ -84,8 +84,8 @@ def setUp(self):
   def init_batch_and_play(self, env_name, steps_per_epoch=1, epochs=(0,),
                           generate_data=False, batch_size=2, **kwargs):
     env = gym_env.T2TGymEnv(env_name, batch_size=batch_size, **kwargs)
-    obs = list()
-    rewards = list()
+    obs = []
+    rewards = []
     num_dones = 0
     for epoch in epochs:
       env.start_new_epoch(epoch, self.out_dir)
@@ -100,8 +100,8 @@ def init_batch_and_play(self, env_name, steps_per_epoch=1, epochs=(0,),
     return env, obs, rewards, num_dones
 
   def play(self, env, n_steps):
-    obs = list()
-    rewards = list()
+    obs = []
+    rewards = []
     obs.append(env.reset())
     num_dones = 0
     for _ in range(n_steps):

tensor2tensor/data_generators/librispeech.py

@@ -67,7 +67,7 @@ def _collect_data(directory, input_ext, transcription_ext):
   #   if the datafile was "/path/to/datafile.wav" then the key would be
   #   "/path/to/datafile"
   # value: a pair of strings (media_filepath, label)
-  data_files = dict()
+  data_files = {}
   for root, _, filenames in os.walk(directory):
     transcripts = [filename for filename in filenames
                    if transcription_ext in filename]

tensor2tensor/data_generators/mscoco.py

@@ -107,7 +107,7 @@ def get_vocab():
   caption_file = io.open(caption_filepath)
   caption_json = json.load(caption_file)
   # Dictionary from image_id to ((filename, height, width), captions).
-  image_dict = dict()
+  image_dict = {}
   for image in caption_json["images"]:
     image_dict[image["id"]] = [(image["file_name"], image["height"],
                                 image["width"]), []]

tensor2tensor/models/research/attention_lm_moe.py

@@ -308,7 +308,7 @@ def print_shape(x, suffix, debug=False):
                 x,
                 hparams.filter_size)
           else:
-            additional_conv_params = dict()
+            additional_conv_params = {}
             if hparams.use_sepconv:
               additional_conv_params = dict(
                   padding="LEFT",

tensor2tensor/rl/player_utils.py

@@ -389,7 +389,7 @@ def infer_paths(output_dir, **subdirs):
   Returns:
     a dictionary with the directories.
   """
-  directories = dict()
+  directories = {}
   for name, path in six.iteritems(subdirs):
     directories[name] = path if path else os.path.join(output_dir, name)
   directories["output_dir"] = output_dir

tensor2tensor/utils/hparam.py

@@ -495,7 +495,7 @@ def parse(self, values):
       ValueError: If `values` cannot be parsed or a hyperparameter in `values`
       doesn't exist.
     """
-    type_map = dict()
+    type_map = {}
     for name, t in self._hparam_types.items():
       param_type, _ = t
       type_map[name] = param_type

tensor2tensor/utils/metrics.py

@@ -600,7 +600,7 @@ def image_wrapped_metric_fn(predictions,
   def weights_fn_for_mp(problem_task_id):
     return lambda x: common_layers.weights_multi_problem(x, problem_task_id)
 
-  eval_metrics = dict()
+  eval_metrics = {}
   for problem_instance in problems:
     problem_name = problem_instance.name
     if problem_instance.was_reversed:
@@ -681,7 +681,7 @@ def create_eager_metrics_internal(metric_fns,
     (accum_fn(predictions, targets) => None,
      result_fn() => dict<str metric_name, float avg_val>
   """
-  tfe_metrics = dict()
+  tfe_metrics = {}
 
   for name in metric_fns:
     tfe_metrics[name] = tfe.metrics.Mean(name=name)

tensor2tensor/utils/rouge.py

@@ -62,7 +62,7 @@ def _lcs(x, y):
     Table of dictionary of coord and len lcs
   """
   n, m = len(x), len(y)
-  table = dict()
+  table = {}
   for i in range(n + 1):
     for j in range(m + 1):
       if i == 0 or j == 0: