TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi
TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi copied to clipboard
EdjeElectronics
Not able to change model to EfficientDet-D1 or D2
Hi EJ
I am trying to change the model to D1 or D2. But it is not working. I received this error ValueError: Cannot assign value to variable ' WeightSharedConvolutionalBoxPredictor/BoxPredictionTower/conv2d_0/BatchNorm/feature_0/beta:0': Shape mismatch.The variable shape (88,), and the assigned value shape (112,) are incompatible.
Possible to help? Here is my code.
chosen_model = 'efficientdet-d2'
MODELS_CONFIG = { 'efficientdet-d2': { 'model_name': 'efficientdet_d2_coco17_tpu-32', 'base_pipeline_file': 'ssd_efficientdet_d2_768x768_coco17_tpu-8.config', 'pretrained_checkpoint': 'efficientdet_d2_coco17_tpu-32.tar.gz', } }
model_name = MODELS_CONFIG[chosen_model]['model_name'] pretrained_checkpoint = MODELS_CONFIG[chosen_model]['pretrained_checkpoint'] base_pipeline_file = MODELS_CONFIG[chosen_model]['base_pipeline_file']
Hello did you solve the problem for this? if so, how did you add the efficientdet-d2?
