inputs=tf.Tensor(shape=(1, 1030, 1030, 3), dtype=float32)

[LorenzoInspire]

I am trying to finetune a db_resnet50 model for ocr detection, and I'm using the exact same code of the Train your own model paragraph (https://github.com/mindee/doctr/blob/main/references/detection/train_pytorch.py).
However, in the line to instantiate the model:

model = detection.__dict__[args.arch](
        pretrained=args.pretrained,
        assume_straight_pages=not args.rotation,
        class_names=val_set.class_names,
    )

I get the following error:

tensorflow.python.framework.errors_impl.InvalidArgumentError: Exception encountered when calling layer 'conv1_conv' (type Conv2D).
{{function_node _wrapped__Conv2D_device/job:localhost/replica:0/task:0/device:GPU:0}} No DNN in stream executor. [Op:Conv2D] name:
Call arguments received by layer 'conv1_conv' (type Conv2D):
• inputs=tf.Tensor(shape=(1, 1030, 1030, 3), dtype=float32)

Why does it happen? I'm starting to believe that it is something related to the gpu because when I try to run the code on the cpu the code continues from that point.
Did it ever happen to you? How can I solve?
Thank you in advance, and if you need any additional detail feel free to ask