SNN_RBM shape mismatch

[CorpaciLC]

Hello,

When trying to run SNN_RBM on the 1458 campaign, we get the following error.

''' ../../make-ipinyou-data/1458/train.yzx.txt drop_mlp4da.py|ad:1458|drop:1|b_size:1000 | X:133465 | Hidden 0:300 | Hidden 1:300 | Hidden 2:100 | L_r:0.0006 | activation1:tanh | lambda:0.0001 training RBM line: 133465 line: 300 Traceback (most recent call last): File "SNN_RBM.py", line 81, in ww0,bb0,ww1,bb1,ww2,bb2=gbrbm.get_rbm_weights(train_file,arr,ncases=train_size,batch_size=100000,fm_model_file=fm_model_file) File "C:\Users\corpa\Master\Sem3\Experiment Design For Data Science\Assignment2\deep-ctr\python\sampling_based_gaussian_binary_rbm_sparse.py", line 533, in get_rbm_weights trainer.train(file,epochs,ncases,cdsteps=1,k=k) File "C:\Users\corpa\Master\Sem3\Experiment Design For Data Science\Assignment2\deep-ctr\python\sampling_based_gaussian_binary_rbm_sparse.py", line 448, in train thisweightstep -= np.dot(vis.T, hid) ValueError: operands could not be broadcast together with shapes (4,300) (32,300) (4,300) '''

Any hard-sought solutions at hand?

Thanks