search

oarriaga / face_classification

Real-time face detection and emotion/gender classification using fer2013/imdb datasets with a keras CNN model and openCV.
MIT License
5.6k stars 1.59k forks source link

ValueError: color_mode must be "grayscale", "rbg", or "rgba" #93

Open omkarhp opened 6 years ago

omkarhp commented 6 years ago

While running image_emotion_gender_demo.py i'm getting error statement as ValueError: color_mode must be "grayscale", "rbg", or "rgba"

Like this:

C:\Users\o.hemant.pawaskar\Downloads\EMOTION_RECOGNITION_ALL_DATA\face_classification-master\src>python C:\Users\o.hemant.pawaskar\Downloads\EMOTION_RECOGNITION_ALL_DATA\face_classification-master\src\image_emotion_gender_demo.py C:\Users\o.hemant.pawaskar\Desktop\happy.jpg Using TensorFlow backend. 2018-10-22 14:48:39.056701: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 Traceback (most recent call last): File "C:\Users\o.hemant.pawaskar\Downloads\EMOTION_RECOGNITION_ALL_DATA\face_classification-master\src\image_emotion_gender_demo.py", line 42, in rgb_image = load_image(image_path, grayscale=False) File "C:\Users\o.hemant.pawaskar\Downloads\EMOTION_RECOGNITION_ALL_DATA\face_classification-master\src\utils\inference.py", line 7, in load_image pil_image = image.load_img(image_path, grayscale, target_size) File "C:\Users\o.hemant.pawaskar\AppData\Local\Continuum\anaconda3\envs\py36\lib\site-packages\keras_preprocessing\image.py", line 509, in load_img raise ValueError('color_mode must be "grayscale", "rbg", or "rgba"') ValueError: color_mode must be "grayscale", "rbg", or "rgba"

Can somebody please help?? Thank you

vajpaye commented 6 years ago

hi @omkarhp

please add color_mode="rgb" in load_image() in inference.py file located in utils folder

def load_image(image_path, grayscale=False, color_mode='rgb', target_size=None): pil_image = image.load_img(image_path, grayscale, color_mode, target_size, ) return image.img_to_array(pil_image)

mazhaojia commented 6 years ago

change to these lines in inference.py to work with the latest keras:

def load_image(image_path, grayscale=False, target_size=None):
    color_mode = 'grayscale'
    if grayscale == False:
        color_mode = 'rgb'
    else:
        grayscale = False
    pil_image = image.load_img(image_path, grayscale, color_mode, target_size)
    return image.img_to_array(pil_image)