Model loading error tensorflow 2.7.0

TensorFlow
, ,
1

i am using below version to train my model and i succesfully trained it.
Python version: 3.8.19 | packaged by conda-forge | (default, Mar 20 2024, 12:47:35)
[GCC 12.3.0]
Tensorflow version: 2.7.0
Keras version: 2.7.0

After succesfully training , the inferencing is also happening using the same model which i saved as tensorflow saved model.

Now i want to deploy it on Jetson nano which have below versions
python version 3.6.15
tensorflow version 2.7.0
keras version 2.7.0

The problem is, i am not able to load my model in jetson nano.
The error i am encountering is in load file.
that (path).meta_graphs[0]
The list index(0) is out of the range
Below is line of codde causing error
tf.keras.models.load_model(‘model2’)

3

Hi @Shikhar_Sharma ,

The error you’re encountering when deploying the model on Jetson Nano is likely due to a compatibility issue between the Python versions on your training environment (Python 3.8.19) and the Jetson Nano environment (Python 3.6.15).

The error you’re encountering suggests that the SavedModel file might not be compatible or might be corrupted.

Try upgrading the Python version on your Jetson Nano to 3.8 or higher to match your training environment

Thank You .