Add files via upload

Author: chubi-x

sortit.py

@@ -0,0 +1,51 @@
+# import required libraries
+import tensorflow as tf
+import streamlit as st
+import cv2
+from PIL import Image, ImageOps
+import numpy as np
+# load the model
+model = tf.keras.models.load_model("sortit_model(v4).hdf5")
+# header
+st.write(
+    """
+    # SortIt Image Classifier
+    """
+
+)
+st.write(
+    """
+    This app classifies images of cats, cars, dogs, bicycles, and motorcycles
+    """
+)
+# upload an image for prediction
+file = st.file_uploader("Please upload an image file", type=["jpg","png"])
+
+# function that imports image resizes it, and runs prediction
+def import_and_predict(img_data,model):
+    # specify the image size
+    size = (150,150)
+    # import the image and resize it
+    image = ImageOps.fit(img_data,size,Image.ANTIALIAS)
+    # convert image to numpy array
+    image = np.asarray(image)
+    # convert the image from bgr to rgb
+    # img = cv2.cvtColor(image,cv2.COLOR_BGR2RGB)
+    # resize the numpy array
+    img_resize = (cv2.resize(image,dsize=(75,75),interpolation=cv2.INTER_CUBIC))/255
+    # reshape the image
+    img_reshape = img_resize[np.newaxis,...]
+    # predictions = np.argmax(model.predict(img_reshape), axis=-1)
+    # run prediction on the image
+    prediction = model.predict(img_reshape)
+    return prediction
+
+if file is None:
+    st.text("Please upload an image file")
+else:
+    #open the image and convert it to rgb
+    image = Image.open(file).convert('RGB')
+    st.image(image,use_column_width=True)
+    prediction = import_and_predict(image,model)
+    # print the prediction
+    st.write(prediction)