expected str, bytes or os.PathLike object, not Sequential

I am trying to deploy my Image Classification model and facing this issue while reading the selected image.

Code snippet:

def prediction(model):
    image_file = st.file_uploader("Choose a bird image..", type="jpg")
    if(image_file is not None): 
        image = Image.open(image_file)
        st.image(image, width=40, caption='Uploaded bird image', use_column_width=True)
        np_image = np.array(image).astype('float32')/255
        np_image = transform.resize(np_image, (224, 224, 3))
        np_image = np.expand_dims(np_image, axis=0)
        model = load_model(model)
        prediction = model.predict(img.reshape(1,224,224,3))
                
        if(np.argmax(prediction)==0.0):
            st.write('According to Model, it can be: COVID-19 with an Accuracy of {0:.2f}%'.format(np.max(prediction)*100))
        elif(np.argmax(prediction)==1.0):
            st.write('According to Model, it can be: Normal with an Accuracy of {0:.2f}%'.format(np.max(prediction)*100))
        elif(np.argmax(prediction)==2.0):
            st.write('According to Model, it can be: Pneumonia with an Accuracy of {0:.2f}%'.format(np.max(prediction)*100))

I executed the same code a month ago, and it worked fine—Donno why I am facing this error now.

Debug info

• Streamlit version: 0.72.0
• Python version: Python 3.7.5
• Tensorflow Version: tensorflow==2.4.1
• Keras Version: Keras==2.3.1
• Using IDLE
• OS version: Windows
• Browser version: Chrome

[snehitvaddi]

Yes @kmcgrady , I did use file.read()
code: image = Image.open(io.BytesIO(file.read())) .

I came across some working code as given below:

def prediction(model):
    file = st.file_uploader("Choose a X-Ray image..", type=["jpg","png"])
    if(file is not None):
        img = Image.open(file)
        img = ImageOps.fit(img,(224,224))
        img = np.asarray(img)
        img_reshape = img[np.newaxis,...]
        
        st.image(img, width=40, caption='Uploaded X-ray image', use_column_width=True)
        st.write("")
        st.write("Classifying...")
        prediction = model.predict(img_reshape)

        print(np.argmax(prediction))
        if(np.argmax(prediction)==0.0):
            st.write('According to Model, it can be: COVID-19 with an Accuracy of {0:.2f}%'.format(np.max(prediction)*100))
        elif(np.argmax(prediction)==1.0):
            st.write('According to Model, it can be: Normal with an Accuracy of {0:.2f}%'.format(np.max(prediction)*100))
        elif(np.argmax(prediction)==2.0):
            st.write('According to Model, it can be: Pneumonia with an Accuracy of {0:.2f}%'.format(np.max(prediction)*100))

Right now, I’m facing two issues.

Issue-1: Here, the Model is wrongly predicting images on Streamlit. I tested the model on Jupyter notebook, and it predicted correctly there.

Issue-2: This worked fine without any errors for images in the same directory but raising errors for images on other directories as below images.

[kmcgrady]

I am going to close this issue after not hearing anything back. Feel free to reopen it if you are still seeing issues.