Created
August 1, 2019 12:55
-
-
Save OlafenwaMoses/6c948aa66ff19aeaa9fa6d4f5f462981 to your computer and use it in GitHub Desktop.
Code for training custom object detection model with ImageAI
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from imageai.Detection.Custom import DetectionModelTrainer | |
| trainer = DetectionModelTrainer() | |
| trainer.setModelTypeAsYOLOv3() | |
| trainer.setDataDirectory(data_directory="hololens") | |
| trainer.setTrainConfig(object_names_array=["hololens"], batch_size=4, num_experiments=100, train_from_pretrained_model="pretrained-yolov3.h5") | |
| trainer.trainModel() |
It giving an error
i can give object name correctly but it shows file not found error pls clarify it
FileNotFoundError: [Errno 2] No such file or directory: 'Number plate/train/annotations'
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I am having an issue where it is not seeing, or evaluating my validation samples.
As per the walk through provided at https://medium.com/deepquestai/train-object-detection-ai-with-6-lines-of-code-6d087063f6ff
I have split my images and annotations roughly 75% 25% between train and validation folders.
there are 69 files in each of the respective images and annotations folder within the validation folder.
when I run the 6lines I am returned
Generating anchor boxes for training images and annotation...
Average IOU for 9 anchors: 0.91
Anchor Boxes generated.
Detection configuration saved in Samples_forTraining\json\detection_config.json
Evaluating over 0 samples taken from Samples_forTraining\validation
Training over 203 samples given at Samples_forTraining\train
Training on: ['SnowMeter']
Training with Batch Size: 4
Number of Training Samples: 203
Number of Validation Samples: 0
Number of Experiments: 100
Training with transfer learning from pretrained Model