make style

Author: tolgacangoz

examples/research_projects/anytext/frozen_clip_embedder_t3.py

@@ -110,8 +110,9 @@ def text_encoder_forward(
             )
             # CLIP's text model uses causal mask, prepare it here.
             # https://github.com/openai/CLIP/blob/cfcffb90e69f37bf2ff1e988237a0fbe41f33c04/clip/model.py#L324
-            causal_attention_mask = _create_4d_causal_attention_mask(input_shape, hidden_states.dtype,
-                                                                     device=hidden_states.device)
+            causal_attention_mask = _create_4d_causal_attention_mask(
+                input_shape, hidden_states.dtype, device=hidden_states.device
+            )
             # expand attention_mask
             if attention_mask is not None:
                 # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]

examples/research_projects/anytext/recognizer.py

@@ -134,13 +134,13 @@ def get_image_file_list(img_file):
 
 class TextRecognizer(object):
     def __init__(self, args, predictor):
-        self.rec_image_shape = [int(v) for v in args['rec_image_shape'].split(",")]
-        self.rec_batch_num = args['rec_batch_num']
+        self.rec_image_shape = [int(v) for v in args["rec_image_shape"].split(",")]
+        self.rec_batch_num = args["rec_batch_num"]
         self.predictor = predictor
-        self.chars = self.get_char_dict(args['rec_char_dict_path'])
+        self.chars = self.get_char_dict(args["rec_char_dict_path"])
         self.char2id = {x: i for i, x in enumerate(self.chars)}
         self.is_onnx = not isinstance(self.predictor, torch.nn.Module)
-        self.use_fp16 = args['use_fp16']
+        self.use_fp16 = args["use_fp16"]
 
     # img: CHW
     def resize_norm_img(self, img, max_wh_ratio):