I'm trying to use batch requests on a retrained Inception model using retrain.py from tensorflow(actually available on hub repository). Unfortunately I didn't find any documentation available.
Basically I'm using inception_client.py to perform requests to tensorflow_model_server on a docker container and it works very well when there is one image(even if with the parameter --enable_batching). I've tried to use a stacked list of 10 images using tf.stack() to test batching but it raises the following exception:
AbortionError(code=StatusCode.INVALID_ARGUMENT, details="Input to reshape is a tensor with 20480 values, but the requested shape has 2048 [[Node: pool_3/_reshape = Reshape[T=DT_FLOAT, Tshape=DT_INT32, _output_shapes=[[1,2048]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](pool_3, pool_3/_reshape/shape)]]")
I'm guessing that my model didn't support batching requests. So, how do I should retrain Inception models to support batching requests?
