ValueError: Tensor Tensor("mrcnn_detection/Reshape_1:0", shape=(1, 100, 6), dtype=float32) is not an element of this graph

Question

I was trying to detect shapes using rcnn with flask. But got this error: Traceback (most recent call last): File "C:\Users\KushalM\AppData\Local\Continuum\anaconda3\lib\site-packages\tensorflow_core\python\framework\ops.py", line 3505, in as_graph_element return self._as_graph_element_ocked(obj, allow_tensor, allow_operation) raise ValueError("Tensor %s is not an element of this graph." % obj) ValueError: Tensor Tensor("mrcnn_detection/Reshape_1:0", shape=(1, 100, 6), dtype=float32) is not an element of this graph.

This is my code:

app.py

    from flask import request, redirect
    import os
    from flask import Flask
    from flask import render_template

    import os
    #import detect
    from detect import getImage
    app = Flask(__name__)
    import os
    import sys
    import numpy as np

    HOME_DIR = os.getcwd()
    LIB_DIR = os.path.join(HOME_DIR, "libraries")
    #os.chdir(LIB_DIR)
    sys.path.insert(0, LIB_DIR)
    #print(os.getcwd())
    print(os.listdir('./libraries/'))

    from mrcnn.config import Config
    import mrcnn.utils as utils
    import mrcnn.model as modellib
    import mrcnn.visualize as visualize
    from mrcnn.model import log
    import skimage
    import tensorflow as tf
    from tensorflow.python.keras import backend as k


    image_size = 64
    rpn_anchor_template = (1, 2, 4, 8, 16) # anchor sizes in pixels
    rpn_anchor_scales = tuple(i * (image_size // 16) for i in rpn_anchor_template)

    class ShapesConfig(Config):
        """Configuration for training on the shapes dataset.
        """
        NAME = "shapes"

        # Train on 1 GPU and 2 images per GPU. Put multiple images on each
        # GPU if the images are small. Batch size is 2 (GPUs * images/GPU).
        GPU_COUNT = 1
        IMAGES_PER_GPU = 1

        # Number of classes (including background)
        NUM_CLASSES = 1 + 4  # background + 3 shapes (triangles, circles, and squares)

        # Use smaller images for faster training. 
        IMAGE_MAX_DIM = image_size
        IMAGE_MIN_DIM = image_size

        # Use smaller anchors because our image and objects are small
        RPN_ANCHOR_SCALES = rpn_anchor_scales

        # Aim to allow ROI sampling to pick 33% positive ROIs.
        TRAIN_ROIS_PER_IMAGE = 32

        STEPS_PER_EPOCH = 50

        VALIDATION_STEPS = STEPS_PER_EPOCH / 20

    config = ShapesConfig()
    config.display()


    class InferenceConfig(ShapesConfig):
        GPU_COUNT = 1
        IMAGES_PER_GPU = 1

    inference_config = InferenceConfig()
    MODEL_DIR = "./weights/mask_rcnn_coco.h5"
    # Recreate the model in inference mode
    model = modellib.MaskRCNN(mode="inference", 
                              config=inference_config,
                              model_dir=MODEL_DIR)

    # Get path to saved weights
    # Either set a specific path or find last trained weights
    # model_path = os.path.join(ROOT_DIR, ".h5 file name here")
    #model_path = model.find_last()

    ROOT = os.getcwd()

    model_path = ROOT + "/weights/mask_rcnn_coco.h5"
    # Load trained weights
    print("Loading weights from ", model_path)
    model.load_weights(model_path, by_name=True)
    #graph = tf.get_default_graph()


    class_names = ['BG', 'Std Paneer', 'Non Std Paneer', 'Std Chicken', 'Non Std Chicken']
    #graph = tf.get_default_graph()

    app.config["IMAGE_UPLOADS"] = './static'

    @app.route("/upload-image", methods=["GET", "POST"])

    def upload_image():

        if request.method == "POST":

            if request.files:

                image = request.files["image"]
                #print(image.shape)
                image= getImage(image) 

                file = image.filename
                image.save(os.path.join(app.config["IMAGE_UPLOADS"], image.filename))

                print("Image saved")
                return render_template("view.html", image_name = file)
        return render_template("upload.html")

    if __name__ == '__main__':
        app.run(debug = True, use_reloader = False)

detect.py
    def getImage(image):
        image = skimage.io.imread(image)
        print(image.shape)
              #image = image.resize((64,64))
              # Run detection
        results = model.detect([image], verbose=1)

              # Visualize results
        r = results[0]
        print(len(r['class_ids']))
        visualize.display_instances(image, r['rois'], r['masks'], r['class_ids'], class_names, 
                                          r['scores'])

Answer 1

I have changed my web framework for building app from flask to FastAPI and the error got resolved

Answer 2

Change to:

model.load_weights(model_path, by_name=True)
model.keras_model._make_predict_function()