* Modern string formatting in csp.py.

* Modern string formatting in games.py.

* Modern string formatting in learning.py.

* Modern string formatting in nlp.py.

* Modern string formatting in probability.py.

* Modern string formatting in search.py.

* Replaced {0\!r} by {} if %s.

* Corrected a typo.

Syntax error. Used the function len instead of count.

* Update search.py

Commenting issues fixed (spacing and punctuation was off sometimes).

* Update agents.py

* Update canvas.py

Grammar

* Update grid.py

* Update learning.py

Added period

* Update logic.py

Fix quoting

* Update mdp.py

Fixed quoting

* Update nlp.py

Capitalization and punctuation fixes

* Update planning.py

* Update probability.py

* Update rl.py

'th' to 'the'

* Update search.py

* Update text.py

* Update utils.py

* Update utils.py

* Update utils.py

* Update learning.py

Typo

* Update utils.py

Seems like a typo, results in error:
TypeError: 'int' object is not iterable.

Fixing it:
for i in len(population) -> for i in range(len(population))

* Corrected the bad bold formatting.

* Changed Python version locally.

* Reverted back the local environment changes.

* Added implementation of the cake problem

* Added test for three_block_tower and fixed a bug in three_block_tower code

In the precondition positive list, I am not sure whether b!=x is the best way to represent the condition. IsNot(b, x) can be used instead.

Added test for spare_tire problem of planning module.

* Implementing HITS algorithm

* Moving HITS work to nlp.py and test_nlp.py

* Minor docstring changes

* Added Spare Tire Problem

* Fixed a bug in substitute method of class Action

* Fixed minor typo in comment

The link to "Pseudocode algorithm (pdf)" was pointing to https://github.com/aimacode/pseudocode/blob/master/algorithms.pdf, which was causing the 404 error. The name of the file algorithms.pdf was changed to aima3e-algorithms.pdf by @ctjoreilly in commit 80286be. Edited the link to reflect that change and point to a valid URL.

this classifier of learning module is not optimized to run on this huge MNIST data

having this flag drastically reduces time to load & sanity check large image datasets for practical ML tasks