Train TensorFlow Lite Model for Custom Object (License Plate) Detection with Custom Dataset

In this video, we delve into the exciting realm of custom object detection using TensorFlow Lite! We'll guide you through the process of training a personalized model specifically designed to recognize license plates. From gathering and preprocessing the data to training the model and exporting it to TensorFlow Lite for optimal efficiency and deployment on edge devices, this tutorial has got you covered.
💡 Whether you're a developer, hobbyist, or tech enthusiast, this video will empower you to create your very own custom object detection model for license plates using TensorFlow Lite. Get ready to enhance your understanding of machine learning and computer vision while unlocking the potential for innovative applications!
👍 Don't forget to like, share, and subscribe for more exciting tutorials and updates on AI, machine learning, and cutting-edge technologies! Let's embark on this exciting journey together. 🚀🤖
Githiub Repo: github.com/Arijit1080/Licence...
Train YOLO V8 on Custom Dataset for Object Detection : • Train YOLO V8 on Custo...
Custom Object (Licence Plate) Detection in Raspberry Pi with YOLO V8 and Python: • Custom Object (Licence...
Licence Plate Recognition with YOLO V8 and Easy OCR using Custom Dataset: • Licence Plate Recognit...
For future updates follow us on facebook: / sparklers2018
You can also follow me on instagram for regular updates : / ari_the_maker
#TensorFlowLite #ObjectDetection #CustomModel #MachineLearning #DeepLearning #ComputerVision #AI #LicensePlateDetection #CustomObjectDetection #TFLite #TensorFlow #NeuralNetworks #OpenCV #Python #ImageProcessing #ArtificialIntelligence #TrainingModels #CustomVision #MachineVision #EdgeComputing #ConvolutionalNeuralNetworks #Tutorial #Coding #Programming #DataScience #ModelTraining #ObjectLocalization #ComputerScience #VisualRecognition #CustomDataset #DataLabeling #VideoTutorial #SoftwareDevelopment #EmbeddedAI #AutomatedRecognition #CustomAI #Customization #OpenSource #CustomObjectRecognition #CustomTagging #MobileAppDevelopment #SmartTechnology #LicensePlateRecognition #DIYAI #AdvancedMachineLearning #TechExplained #LearnAI #BuildYourOwnAIModel

[@svarsha2579]

Hi, the training part doesn't complete even after 6 hours. what could be the issue?

[@minhazulmahmudriyad1559]

Hi there how can i use other models of tfod zoo. Just cant find the other two(checkpoint,config) except model name. It would be great if you show us how to do it.

[@Wikipedieu]

hi, could you help me? i'm already finish training my own dataset (big thanks) but i've got error during running your LicenseDetector.py the messages is "from tensorflow.lite.python.interpreter import Interpreter
ModuleNotFoundError: No module named 'tensorflow.lite.python.interpreter'"
may you have any solution to solve this problem, thank you. keep it up 🙌

[@user-yo5oz5td6h]

sir is this compatible in flutter?

[@abdullahsuheylozmen3255]

I can't connect to the gpu while training the model. Even though TPU4 is selected, training takes place via the CPU.

[@animationp973]

same problem here

[@TlTAN]

hi, is a really nice video, its working, but i have a question, my dataset is large, when the time of google colab is done, how can start the train again but from the las checkpoint?

[@uriel2348]

I think it deletes everything when the time is done. Try checking the colab pro, maybe there you can get more time

[@Ashwanth777]

hi bro is the code is running ?

[@user-ry8wu7ko8s]

From where i get my key in roboflow

[@oliverexcellent]

Roboflow automatically generates a key for you when you get your dataset. Choose "Download dataset" and then "show download code"

[@jaybryanodero8110]

send collab note book link bro

[@chandramohan1281]

How to create our own data set and use that ?

[@pocketgamer8446]

You can use the labelimg python package to label your own images

[@mohanrajudhandapani3241]

Hello Brother
I am getting error in this section and unable to proceed further , could you please help on this?
pipeline_fname = '/content/models/mymodel/' + base_pipeline_file
fine_tune_checkpoint = '/content/models/mymodel/' + model_name + '/checkpoint/ckpt-0'
def get_num_classes(pbtxt_fname):
from object_detection.utils import label_map_util
label_map = label_map_util.load_labelmap(pbtxt_fname)
categories = label_map_util.convert_label_map_to_categories(
label_map, max_num_classes=90, use_display_name=True)
category_index = label_map_util.create_category_index(categories)
return len(category_index.keys())
num_classes = get_num_classes(label_map_pbtxt_fname)
print('Total classes:', num_classes)

[@mohanrajudhandapani3241]

Error :
NotFoundError Traceback (most recent call last)
in ()
9 category_index = label_map_util.create_category_index(categories)
10 return len(category_index.keys())
---> 11 num_classes = get_num_classes(label_map_pbtxt_fname)
12 print('Total classes:', num_classes)
3 frames
/usr/local/lib/python3.10/dist-packages/tensorflow/python/lib/io/file_io.py in _preread_check(self)
74 raise errors.PermissionDeniedError(None, None,
75 "File isn't open for reading")
---> 76 self._read_buf = _pywrap_file_io.BufferedInputStream(
77 compat.path_to_str(self.__name), 1024 * 512)
78
NotFoundError: /content/labelmap.pbtxt; No such file or directory

[@user-nl2rw5vg4y]

check the path and make sure that .pbtxt file created before

[@chaimaez6651]

my model takes so much time to train (hours..). is that normal or can there be a problem ?

[@animationp973]

same problem here

[@ceynacecia_]

2024-05-12 07:59:55.623431: W tensorflow/core/framework/dataset.cc:768] Input of GeneratorDatasetOp::Dataset will not be optimized because the dataset does not implement the AsGraphDefInternal() method needed to apply optimizations.
how to fix?