search

StefanBloemheuvel / GCNTimeseriesRegression

Github page for: Graph Neural Networks for Multivariate Time Series Regression with Application to Seismic Data
33 stars 5 forks source link

Questions about gcn model #2

Closed CaoZPLQ closed 1 year ago

CaoZPLQ commented 1 year ago

When I execute "python main_gcn.py network1 main 1" when I go to main_gcn.py line 121: "conv1 = layers.Conv1D(filters=32, kernel_size=125, strides=2, activation=activation_func, kernel_regularizer=regularizers.l2(reg_const), name='conv1')(wav_input)", ValueError: Input 0 of layer conv1 is incompatible with the layer: expected ndim=3, found ndim=4. Full shape received: [None, 39, 1000, 3] is displayed. Is this a problem with my Keras version? My environment tensorflow=2.2.0,keras=2.3.1, or other problems? Thank you

StefanBloemheuvel commented 1 year ago

Hello!

For me it seems to work, so it has to be dependencies indeed! I will update the readme but also answer here :)

geopy==2.2.0 keras==2.8.0 networkx==2.7.1 numba==0.56.2 numpy==1.22.3 scikit-learn==1.0.2 scipy==1.8.0 sklearn==0.0 spektral==1.1.0 tensorflow==2.8.0

CaoZPLQ commented 1 year ago

Thank you very much! It works.