My TFLite running on Desktop

I have TFlite Model and i want to run checking image on Desktop.

Preparation

If you don’t have Python on your PC then you can install it from: Getting Start with Python.

You can use my sample TFLite Model. Inside my TFLite model there are 4 labels. indosiar_iklan, indosiar, sctv and sctv_iklan. If you confused with TFLite or Label you can read this.

Execution

Use following script to run the TFLite

import tensorflow as tf
import cv2.cv2 as cv
import pathlib
import sys
# Load TFLite model and allocate tensors.
interpreter = tf.lite.Interpreter(model_path="model.tflite")
# Get input and output tensors.
input_details = interpreter.get_input_details()
output_details = interpreter.get_output_details()
# input details
print(input_details)
# output details
print(output_details)
interpreter.allocate_tensors()for file in pathlib.Path(sys.argv[1]).iterdir():
# read and resize the image
img = cv.imread(r"{}".format(file.resolve()))
new_img = cv.resize(img, (224, 224))

interpreter.set_tensor(input_details[0]['index'], [new_img])
interpreter.invoke()
output_data = interpreter.get_tensor(output_details[0]['index'])
if (output_data[0][0] > 200):
print("file {} -> indosiar_ads".format(file.stem))
elif (output_data[0][1] > 200):
print("file {} -> indosiar".format(file.stem))
elif (output_data[0][2] > 200):
print("file {} -> sctv".format(file.stem))
elif (output_data[0][3] > 200):
print("file {} -> sctv_ads".format(file.stem))
else:
print("file {} -> unknown".format(file.stem))

My TFLite model file name is model.tflite. So i give that on model_path when creating interpreter object.

On looping section there is sys.argv[1] that the value get from arg given when run the script. The argument is folder test path.

The images for test you can download from this link.

You can use Conda or Pip for this.

pip3 install tensorflow
pip3 install pathlib
pip3 install opencv-python

Run the script with following argument as testing images folder.

python3 main.py test

Still believe, can change the world with code..