add GenerationMixin inheritance - needed for transformers >= 4.50

kha-white · Oct 11, 2024 · 0588d90 · 0588d90
1 parent 083ddbd
commit 0588d90
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 3 deletions.
diff --git a/manga_ocr/_version.py b/manga_ocr/_version.py
@@ -1 +1 @@
-__version__ = "0.1.12"
+__version__ = "0.1.13"
diff --git a/manga_ocr/ocr.py b/manga_ocr/ocr.py
@@ -5,15 +5,18 @@
 import torch
 from PIL import Image
 from loguru import logger
-from transformers import ViTImageProcessor, AutoTokenizer, VisionEncoderDecoderModel
+from transformers import ViTImageProcessor, AutoTokenizer, VisionEncoderDecoderModel, GenerationMixin
 
 
+class MangaOcrModel(VisionEncoderDecoderModel, GenerationMixin):
+    pass
+
 class MangaOcr:
     def __init__(self, pretrained_model_name_or_path="kha-white/manga-ocr-base", force_cpu=False):
         logger.info(f"Loading OCR model from {pretrained_model_name_or_path}")
         self.processor = ViTImageProcessor.from_pretrained(pretrained_model_name_or_path)
         self.tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path)
-        self.model = VisionEncoderDecoderModel.from_pretrained(pretrained_model_name_or_path)
+        self.model = MangaOcrModel.from_pretrained(pretrained_model_name_or_path)
 
         if not force_cpu and torch.cuda.is_available():
             logger.info("Using CUDA")