Hello,
when I try to import a Conv2dTranspose function i get the following error:
Traceback (most recent call last): File “micro_gemmini_conv2d_transpose.py”, line 165, in mod, params = relay.frontend.from_tflite( File “/home/rh8588/Dokumente/git/tvm/python/tvm/relay/frontend/tflite.py”, line 4063, in from_tflite op_converter.convert_op_to_relay() File “/home/rh8588/Dokumente/git/tvm/python/tvm/relay/frontend/tflite.py”, line 273, in convert_op_to_relay ret = self.convert_mapop_code_str File “/home/rh8588/Dokumente/git/tvm/python/tvm/relay/frontend/tflite.py”, line 3146, in convert_transpose_conv output_shape_value = self.get_tensor_value(output_shape_tensor) File “/home/rh8588/Dokumente/git/tvm/python/tvm/relay/frontend/tflite.py”, line 466, in get_tensor_value return np.frombuffer(data, dtype=dtype).reshape(shape) TypeError: a bytes-like object is required, not ‘int’
Is there a way to fix this? I created the function in Tensorflow.keras.layers and then quantized it with the TFLiteConverter.