Improvement in train_test_split function

Shuffling has been removed

* Update in NeuralNetLearner function

* made the changes as suggested

* no need of if else

* no need of if - else .   

As if hidden_layer_sizes is zero then it will not affect layer_sizes

* removed comment

* Update learning.py

* added relu activation

* added default parameters

* Correction in the formula for mean square error

* Added cross-entropy loss

* Test case for cross-entropy loss

* Decimal point mistake

* Added spaces around = and ==

* styling changes and bug fixes in learning.py

* Fix #833 and other pep corrections in mdp.py

* minor change mdp.py

* renamed train_and_test() to train_test_split() #55 #830

* typo fix

* Update learning.py

* Update test_learning.py

* added example for Decision Tree Learner

* fixed docstring in learning.py and description in learning.ipynb

* add failure_test method to utils

* comment fix

* Update test_learning.py

* Update test_csp.py

* add a simple naive bayes classifier

* Update test_learning.py

* spacing

* minor fix

* lists to strings

* Update README.md

* add powerset to utils

* add powerset test

* Create knowledge.py

* Create test_knowledge.py

* add header docstring to knowledge.py

* update learning docstring

* minor edits in knowledge.py

* Fixed truncated_svd

* Moved SVD algorithm to learning.py

* Minor fixes

* Typo fix

* Added decisiontreelearner to notebook

* Added RandomForest

* Update test_learning.py

* Update learning.py

* DataFile Update

* Update learning.py

* Update search.py

* Update test_learning.py

* Update test_text.py

* Add files via upload

* Update text.ipynb

* Update search-4e.ipynb

* Create CONTRIBUTING.md

is supposed to return an answer when errT converges, not errV
used to return size of when err_val converges but is supposed to return the size with minimum err_val

* Create function to initialize random weights

* Add sigmoid derivative function

* Fix flake8 warnings

* Remove unnecessary #noqa

* Fix doctest

* Update learning.ipynb

* Delete perceptron.png

* Add new Perceptron image

* Update Perceptron Implementation

* Add Gaussian Function

* Added Tests

Add tests for Continuous Naive Bayes + Means/Standard Deviation

* Update learning.py

* Commenting Fix

* Add test for gaussian

* test for every class

* Update test_learning.py

* Round float results to make sure test passes

* Update learning.py

* Update test_learning.py

* Update test_learning.py

* Add grade_learner

* Update test_learning.py

* Update test_learning.py

* Update learning.py

* set max_score to -1 (for now)

* Update learning.py

* Make find_max more pythonic

* Update learning.py

* Add Euclidean Distance

* Update learning.ipynb

* minor fix in notebook

* minor spacing in learning.py

* Added Euclidean Test

* Exclude test files from flake8 check

* Fix flake8 for main files

* Add flake8 check to build

* Bugfixing

* Test for "exclude"

* Update test_learning.py

* update_values

* Bug fixing

* Spacing

* Update test_agents.py

pep8 changes, showed flake8 errors

* Update test_agents.py

* Update test_agents.py

* Update test_agents.py

* Update test_text.py

added missing whitespace after comma

* Update utils.py

added space after comma

* Update search.py

added space after comma

* Update probability.py

added space after comma

* Update learning.py

added space after comma

* Update planning.py

added double_tennis_problem

* Update rl.py

In the pseudocode figure 21.8, the first 'if' starts with argument 's', which is the previous state, not s1(i.e, the current state).

* Update search.py

the 'uniform_cost_search' in notebook 'search-4e.ipynb' resembles more to the pseudocode in book.

* Update search.py

* Update search.py

* Update search.py

created a parameter size in the function leave_one_out(learner, dataset, size=None):

* Update learning.py

* Added remove_examples function