When using a simple NN with Conv1D layer, I get this error:
Layer.js:30 Uncaught (in promise) Error: [Conv1D layer: conv1d_1] Incompatible combination of dilation_rate with strides.
at Conv1D.throwError (webpack:///./src/Layer.js?:30:13)
at new Conv1D (webpack:///./src/layers/convolutional/Conv1D.js?:68:13)
Since my dilation_rate & strides are both 1, I started looking further. The code believes that typeof dilation_rate and strides are both 'object', rather than 'number'. When I re-casted them using 'parseInt' and forced them to be a number, I ended up with this error:
VM18120:8 Uncaught (in promise) Error: cwise: Arrays do not all have the same shape!
at assign_cwise_thunk (eval at createThunk (webpack:///./node_modules/cwise-compiler/lib/thunk.js?), <anonymous>:8:71)
at Object.assign_ndarrayops [as assign] (eval at makeOp (webpack:///./node_modules/ndarray-ops/ndarray-ops.js?), <anonymous>:3:43)
at Conv2D._w2row (webpack:///./src/layers/convolutional/Conv2D.js?:286:61)
at Conv2D.setWeights (webpack:///./src/layers/convolutional/Conv2D.js?:142:12)
at Conv1D.setWeights (webpack:///./src/layers/convolutional/Conv1D.js?:97:20)
at Model._createLayer (webpack:///./src/Model.js?:504:15)
When using a simple NN with Conv1D layer, I get this error:
Since my dilation_rate & strides are both 1, I started looking further. The code believes that typeof dilation_rate and strides are both 'object', rather than 'number'. When I re-casted them using 'parseInt' and forced them to be a number, I ended up with this error: