batch_size = 10 nlabel = 105 debug = 1 model = Model() train_data = Music(name='train', path=data_path, nlabel=nlabel) valid_data = Music(name='valid', path=data_path, nlabel=nlabel) # Choose the random initialization method init_W = InitCell('randn') init_U = InitCell('ortho') init_b = InitCell('zeros') x, y, mask = train_data.theano_vars() # You must use THEANO_FLAGS="compute_test_value=raise" python -m ipdb if debug: x.tag.test_value = np.zeros((10, batch_size, nlabel), dtype=np.float32) y.tag.test_value = np.zeros((10, batch_size, nlabel), dtype=np.float32) mask.tag.test_value = np.ones((10, batch_size), dtype=np.float32) h1 = LSTM(name='h1', parent=['x'], parent_dim=[105], nout=50, unit='tanh', init_W=init_W, init_U=init_U, init_b=init_b)
model = Model() train_data = Music(name='train', path=data_path, nlabel=nlabel) valid_data = Music(name='valid', path=data_path, nlabel=nlabel) # Choose the random initialization method init_W = InitCell('randn') init_U = InitCell('ortho') init_b = InitCell('zeros') x, y, mask = train_data.theano_vars() # You must use THEANO_FLAGS="compute_test_value=raise" python -m ipdb if debug: x.tag.test_value = np.zeros((10, batch_size, nlabel), dtype=np.float32) y.tag.test_value = np.zeros((10, batch_size, nlabel), dtype=np.float32) mask.tag.test_value = np.ones((10, batch_size), dtype=np.float32) h1 = LSTM(name='h1', parent=['x'], parent_dim=[105], nout=50, unit='tanh', init_W=init_W, init_U=init_U, init_b=init_b)
debug = 0 model = Model() trdata = Music(name='train', path=data_path, nlabel=nlabel) valdata = Music(name='valid', path=data_path, nlabel=nlabel) # Choose the random initialization method init_W = InitCell('randn') init_U = InitCell('ortho') init_b = InitCell('zeros') model.inputs = trdata.theano_vars() x, y, mask = model.inputs # You must use THEANO_FLAGS="compute_test_value=raise" python -m ipdb if debug: x.tag.test_value = np.zeros((10, batch_size, nlabel), dtype=np.float32) y.tag.test_value = np.zeros((10, batch_size, nlabel), dtype=np.float32) mask.tag.test_value = np.ones((10, batch_size), dtype=np.float32) inputs = [x, y, mask] inputs_dim = {'x':nlabel} h1 = LSTM(name='h1', parent=['x'], batch_size=batch_size, nout=50, unit='tanh',