I have a list of tensors, each of 300x1
dimensions and I need to find the mean and variance for each dimension based on the list, how exactly can I achieve this. I could make this work in numpy with numpy arrays, with the following code snippet:
for item in embeddingLists: # embeddingLists is a list of lists
# item contains list of numpy arrays
tempVal = np.mean(np.array(item),axis=0)
meanVects.append(tempVal)
temVar = np.var(np.array(item),axis=0,ddof=1)
varVects.append(temVar)