pmingkr
February 27, 2024, 7:59am
1
I’m using a 1d convolution layer as the output layer with multi channels.
so it multiplies the mini batch size with the channel size.
the layer get the batch size from the graph but it indicates the batch size of the output layer.
FeedForwardToRnnPreProcessor.preProcess()
divides the shape with the over-sized mini batch size. and it becomes zero.
@pmingkr you can use the reshape preprocessor here:
/*
* ******************************************************************************
* *
* *
* * This program and the accompanying materials are made available under the
* * terms of the Apache License, Version 2.0 which is available at
* * https://www.apache.org/licenses/LICENSE-2.0.
* *
* * See the NOTICE file distributed with this work for additional
* * information regarding copyright ownership.
* * Unless required by applicable law or agreed to in writing, software
* * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
* * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
* * License for the specific language governing permissions and limitations
* * under the License.
* *
* * SPDX-License-Identifier: Apache-2.0
* *****************************************************************************
*/
show original
for the multilayenr network api.
For anything more complicated, consider using the computation graph api which has a reshape vertex.
