@mdespriee firstly the problem mentioned in the log means the shape of the data changed and there maybe data missing. As explained in [here](http://mxnet.apache.org/api/python/ndarray/ndarray.html#mxnet.ndarray.NDArray.reshape), users should not try to reshape the data with loss of data. However, I can use python to do the same thing without any issues: ```Python import mxnet as mx a = mx.nd.array([[1,2],[3,4]]) b = a.reshape(1, -2) >>> b [[1. 2.]] <NDArray 1x2 @cpu(0)> ``` Not sure what is the issue here, will look into it and see.... @mxnet-label-bot please label [scala, bug]
Reproducible code: ```Scala val nd = NDArray.array(Array(1, 2, 3, 4), Shape(2, 2)) println(nd.shape) nd.toArray.foreach(println) val nd2 = NDArray.api.reshape(nd, shape = Some(Shape(1, -2))) nd2.toArray.foreach(println) ``` [ Full content available at: https://github.com/apache/incubator-mxnet/issues/12425 ] This message was relayed via gitbox.apache.org for [email protected]
