I am new in computer vision. I was trying to train model for image detection and it was working fine but from all of sudden now my model.predict() is showing error of Unidentified image error in google colab. this is full error:
UnidentifiedImageError Traceback (most recent call last) in <cell line: 1>() ----> 1 best_model.predict("img2.jpg")
8 frames /usr/local/lib/python3.10/dist-packages/super_gradients/training/models/detection_models/customizable_detector.py in predict(self, images, iou, conf, batch_size, fuse_model, skip_image_resizing, nms_top_k, max_predictions, multi_label_per_box, class_agnostic_nms, fp16) 306 fp16=fp16, 307 ) --> 308 return pipeline(images, batch_size=batch_size) # type: ignore 309 310 def predict_webcam(
/usr/local/lib/python3.10/dist-packages/super_gradients/training/pipelines/pipelines.py in call(self, inputs, batch_size) 118 return self.predict_video(inputs, batch_size) 119 elif check_image_typing(inputs): --> 120 return self.predict_images(inputs, batch_size) 121 else: 122 raise ValueError(f"Input {inputs} not supported for prediction.")
/usr/local/lib/python3.10/dist-packages/super_gradients/training/pipelines/pipelines.py in predict_images(self, images, batch_size) 131 from super_gradients.training.utils.media.image import load_images 132 --> 133 images = load_images(images) 134 135 result_generator = self._generate_prediction_result(images=images, batch_size=batch_size)
/usr/local/lib/python3.10/dist-packages/super_gradients/training/utils/media/image.py in load_images(images) 32 :return: List of images as numpy arrays. If loaded from string, the image will be returned as RGB. 33 """ ---> 34 return [image for image in generate_image_loader(images=images)] 35 36
/usr/local/lib/python3.10/dist-packages/super_gradients/training/utils/media/image.py in (.0) 32 :return: List of images as numpy arrays. If loaded from string, the image will be returned as RGB. 33 """ ---> 34 return [image for image in generate_image_loader(images=images)] 35 36
/usr/local/lib/python3.10/dist-packages/super_gradients/training/utils/media/image.py in generate_image_loader(images) 66 yield load_image(image=image) 67 else: ---> 68 yield load_image(image=images) 69 70
/usr/local/lib/python3.10/dist-packages/super_gradients/training/utils/media/image.py in load_image(image, input_image_channels) 120 return load_np_image_from_pil(image) 121 elif isinstance(image, str): --> 122 image = load_pil_image_from_str(image_str=image) 123 return load_np_image_from_pil(image) 124 else:
/usr/local/lib/python3.10/dist-packages/super_gradients/training/utils/media/image.py in load_pil_image_from_str(image_str) 139 return PIL.Image.open(io.BytesIO(response.content)) 140 else: --> 141 return PIL.Image.open(image_str) 142 143
/usr/local/lib/python3.10/dist-packages/PIL/Image.py in open(fp, mode, formats) 3281 continue 3282 except BaseException: -> 3283 if exclusive_fp: 3284 fp.close() 3285 raise
UnidentifiedImageError: cannot identify image file 'img2.jpg'
I tried to traceback and my image path was correct and image is not corrupted because when I tried with train_data.dataset._load_image("img2.jpg") it is opening.
For more, I am attaching my code in google colab: https://colab.research.google.com/drive/17DGZzUHsaClbjiJPlfDadmx4x9sX-aYC?usp=sharing
I tried to open image using PIL but still error was same and tried using other image but error is same and unsolved.