Convert a TensorFlow Model to OpenVINO™

This tutorial is also available as a Jupyter notebook that can be cloned directly from GitHub. See the installation guide for instructions to run this tutorial locally on Windows, Linux or macOS. To run without installing anything, click the launch binder button.

Binder Github

This short tutorial shows how to convert a TensorFlow MobileNetV3 image classification model to OpenVINO Intermediate Representation (OpenVINO IR) format, using Model Optimizer. After creating the OpenVINO IR, load the model in OpenVINO Runtime and do inference with a sample image.


import time
from pathlib import Path

import cv2
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
from IPython.display import Markdown
from openvino.runtime import Core
2023-05-17 22:24:09.842009: I tensorflow/core/util/] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.
2023-05-17 22:24:09.876419: I tensorflow/core/platform/] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-05-17 22:24:10.397418: W tensorflow/compiler/tf2tensorrt/utils/] TF-TRT Warning: Could not find TensorRT


# The paths of the source and converted models.
model_dir = Path("model")

model_path = Path("model/v3-small_224_1.0_float")

ir_path = Path("model/v3-small_224_1.0_float.xml")

Download model

Load model using tf.keras.applications api and save it to the disk.

model = tf.keras.applications.MobileNetV3Small()
WARNING:tensorflow:input_shape is undefined or non-square, or rows is not 224. Weights for input shape (224, 224) will be loaded as the default.
2023-05-17 22:24:11.288808: W tensorflow/core/common_runtime/gpu/] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. model.compile_metrics will be empty until you train or evaluate the model.
2023-05-17 22:24:15.543664: I tensorflow/core/common_runtime/] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'inputs' with dtype float and shape [?,1,1,1024]
     [[{{node inputs}}]]
2023-05-17 22:24:18.727793: I tensorflow/core/common_runtime/] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'inputs' with dtype float and shape [?,1,1,1024]
     [[{{node inputs}}]]
WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 54). These functions will not be directly callable after loading.
INFO:tensorflow:Assets written to: model/v3-small_224_1.0_float/assets
INFO:tensorflow:Assets written to: model/v3-small_224_1.0_float/assets

Convert a Model to OpenVINO IR Format

Convert a TensorFlow Model to OpenVINO IR Format

Use Model Optimizer to convert a TensorFlow model to OpenVINO IR with FP16 precision. The models are saved to the current directory. Add mean values to the model and scale the output with the standard deviation with --scale_values. With these options, it is not necessary to normalize input data before propagating it through the network. The original model expects input images in RGB format. The converted model also expects images in RGB format. If you want the converted model to work with BGR images, use the --reverse-input-channels option. For more information about Model Optimizer, including a description of the command-line options, see the Model Optimizer Developer Guide. For information about the model, including input shape, expected color order and mean values, refer to the model documentation.

First construct the command for Model Optimizer, and then execute this command in the notebook by prepending the command with an !. There may be some errors or warnings in the output. When model optimization is successful, the last lines of the output will include [ SUCCESS ] Generated IR version 11 model.

# Construct the command for Model Optimizer.
mo_command = f"""mo
                 --saved_model_dir "{model_path}"
                 --input_shape "[1,224,224,3]"
                 --model_name "{}"
                 --output_dir "{model_path.parent}"
mo_command = " ".join(mo_command.split())
print("Model Optimizer command to convert TensorFlow to OpenVINO:")
Model Optimizer command to convert TensorFlow to OpenVINO:

mo --saved_model_dir "model/v3-small_224_1.0_float" --input_shape "[1,224,224,3]" --model_name "v3-small_224_1.0_float" --compress_to_fp16 --output_dir "model"

# Run Model Optimizer if the IR model file does not exist
if not ir_path.exists():
    print("Exporting TensorFlow model to IR... This may take a few minutes.")
    ! $mo_command
    print(f"IR model {ir_path} already exists.")
Exporting TensorFlow model to IR... This may take a few minutes.
Check for a new version of Intel(R) Distribution of OpenVINO(TM) toolkit here or on
[ INFO ] The model was converted to IR v11, the latest model format that corresponds to the source DL framework input/output format. While IR v11 is backwards compatible with OpenVINO Inference Engine API v1.0, please use API v2.0 (as of 2022.1) to take advantage of the latest improvements in IR v11.
Find more information about API v2.0 and IR v11 at
[ SUCCESS ] Generated IR version 11 model.
[ SUCCESS ] XML file: /opt/home/k8sworker/cibuilds/ov-notebook/OVNotebookOps-408/.workspace/scm/ov-notebook/notebooks/101-tensorflow-to-openvino/model/v3-small_224_1.0_float.xml
[ SUCCESS ] BIN file: /opt/home/k8sworker/cibuilds/ov-notebook/OVNotebookOps-408/.workspace/scm/ov-notebook/notebooks/101-tensorflow-to-openvino/model/v3-small_224_1.0_float.bin

Test Inference on the Converted Model

Load the Model

ie = Core()
model = ie.read_model(ir_path)
compiled_model = ie.compile_model(model=model, device_name="CPU")

Get Model Information

input_key = compiled_model.input(0)
output_key = compiled_model.output(0)
network_input_shape = input_key.shape

Load an Image

Load an image, resize it, and convert it to the input shape of the network.

# The MobileNet network expects images in RGB format.
image = cv2.cvtColor(cv2.imread(filename="../data/image/coco.jpg"), code=cv2.COLOR_BGR2RGB)

# Resize the image to the network input shape.
resized_image = cv2.resize(src=image, dsize=(224, 224))

# Transpose the image to the network input shape.
input_image = np.expand_dims(resized_image, 0)


Do Inference

result = compiled_model(input_image)[output_key]

result_index = np.argmax(result)
# Convert the inference result to a class name.
imagenet_classes = open("../data/datasets/imagenet/imagenet_2012.txt").read().splitlines()

'n02099267 flat-coated retriever'


Measure the time it takes to do inference on thousand images. This gives an indication of performance. For more accurate benchmarking, use the Benchmark Tool in OpenVINO. Note that many optimizations are possible to improve the performance.

num_images = 1000

start = time.perf_counter()

for _ in range(num_images):

end = time.perf_counter()
time_ir = end - start

    f"IR model in OpenVINO Runtime/CPU: {time_ir/num_images:.4f} "
    f"seconds per image, FPS: {num_images/time_ir:.2f}"
IR model in OpenVINO Runtime/CPU: 0.0010 seconds per image, FPS: 1042.50