Naive-Bayes Probabiities: too many indices for array: array is 1-dimensional, but 2 were indexed

32 Views Asked by At

I am trying to fix the Python indexError triggered in Goggle Collab, and have tried to read other solutions previously shared, but cannot find the right solution.

IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed

import math
math.log(2)
pred = []
# TODO -- implement this for loop to make predictions
for d in data_ts_prep:
  max_p = -float('inf')
  max_c = -1
  for c_i in range(c):
    p = math.log(prior_prob[c_i])
    for t in d:
      if t in tok2idx:
        p += math.log(cond_prob[tok2idx[t], c_i])
        if p > max_p:
          max_p = p
          max_c = c_i
      pred.append(max_c)
      
      import numpy as np
      V=3
      cond_prob = np.zeros(10, float)
      cond_prob = cond_prob + 1
      cond_prob /= (np.sum(cond_prob, axis=0) + V)
      cond_prob
  pass

I tried to use the reshape function on the array to make it 1-dimensionl and my attempts did not work.

0

There are 0 best solutions below