Dead kernel

[benzom]

I follow the example from README except for the part where model is being quantized

from fastT5 import (OnnxT5, get_onnx_runtime_sessions,
                    generate_onnx_representation, quantize)
from transformers import AutoTokenizer

model_or_model_path = 't5-small'

# Step 1. convert huggingfaces t5 model to onnx
onnx_model_paths = generate_onnx_representation(model_or_model_path)
model_sessions = get_onnx_runtime_sessions(onnx_model_paths)

however get_onnx_runtime_sessions is killing the kernel, what's wrong? same for

from fastT5 import export_and_get_onnx_model
from transformers import AutoTokenizer

model_name = 't5-small'
model = export_and_get_onnx_model(model_name, quantized=False)

I'm using
fastt5==0.1.4
torch==1.13.1
onnx==1.8.0