Random generation of kterms in a kterm-DNF.
Before, the kterm has a fixed number of '0's and '1's
Now, the kterm has a random number of '0's and '1's

-- Table.py/merge function: Remove error when trying to merge two terms that are equal.
-- test_k_d_s_parallel.py: Addition of padding when printing the graphic of 'num_terms'.
-- test_k_d_s_parallel.py: Alignment of testing_sample_par and training_sample_par for running parallel tasks.

- doc/example with script for parallel learning

- The algorithm reaches k-terms in the result DNF
- The algorithm reaches an error (false negatives) below a value 'e'
- The algorithm reaches a point where the hypothesis function makes a jump that is higher than a threshold

- merge function of Table class (i.e., trying to remove a key that is not in the dictionary)
- arithmetic operations in _overlap functions for generating samples (i.e., force to have int numbers in python 3)

- Minor changes for performance improvement

- Implementation of first version of plotting functions for showing the evolution of estimation errors
- Minor changes