opt/conda/lib/python3.6/site-packages/keras/callbacks.py in on_batch_end(self, batch, logs)
234 else:
235 if k in self.totals:
--> 236 self.totals[k] += v * batch_size
237 else:
238 self.totals[k] = v * batch_size
ValueError: operands could not be broadcast together with shapes (32,) (20,) (32,)
This error pops up midway through training - it seems to be at batch end. Any body have pointers to what shapes are being referred to here?
Per Kyle's request :
class batch_creator(Sequence):
def __init__(self, data, batch_size,window_size, V):
self.data = data
self.batch_size = batch_size
def __len__(self):
return int(len(self.data) / float(self.batch_size))
def __getitem__(self, idx):
batch_x = self.data[idx*self.batch_size :(idx +1)* self.batch_size]
#if idx % self.batch_size==0:
#print(idx)
#words=[]
# contexts=[]
x,z,y =next(generate_data(batch_x, window_size, V))
x = np.array(x,dtype=np.int32)
z = np.array(z,dtype=np.int32)
y= np.array(y,dtype=np.int32)
#print(x)
return [x,y]
"""Instantiate Batchmanager"""
model.fit_generator(batch_manager,epochs=5)
