From 1da4a16ec81279a4b92b3a38a7e4ae52bc91ede8 Mon Sep 17 00:00:00 2001
From: onyx-and-iris <code@onyxandiris.online>
Date: Fri, 19 Jan 2024 17:21:42 +0000
Subject: [PATCH] add some notes

---
 chapter10/README.md | 3 +++
 chapter11/README.md | 6 ++++++
 chapter12/README.md | 9 +++++++++
 chapter4/binary.py  | 1 -
 chapter9/README.md  | 5 +++++
 5 files changed, 23 insertions(+), 1 deletion(-)
 create mode 100644 chapter10/README.md
 create mode 100644 chapter11/README.md
 create mode 100644 chapter12/README.md
 create mode 100644 chapter9/README.md

diff --git a/chapter10/README.md b/chapter10/README.md
new file mode 100644
index 0000000..ec8632b
--- /dev/null
+++ b/chapter10/README.md
@@ -0,0 +1,3 @@
+# Approximation algorithm
+
+-   Easy to write, fast to run, useful for obtaining approximate solutions for NP-hard problems.
diff --git a/chapter11/README.md b/chapter11/README.md
new file mode 100644
index 0000000..d348b66
--- /dev/null
+++ b/chapter11/README.md
@@ -0,0 +1,6 @@
+# Dynamic Programming
+
+A programming technique for decomposing a problem into smaller discrete subproblems.
+
+-   Useful when trying to optimize something given a constraint.
+    -   Example, items in a knapsack of size W that gives the greates value.
diff --git a/chapter12/README.md b/chapter12/README.md
new file mode 100644
index 0000000..e94123f
--- /dev/null
+++ b/chapter12/README.md
@@ -0,0 +1,9 @@
+# K-Nearest Neighbours
+
+Useful for classification, regression and feature extraction. By examining a data point against its K nearest neighbours we can:
+
+-   categorize into a group
+-   predict responses
+-   convert the item into a list of features
+
+A good starting point for machine learning.
diff --git a/chapter4/binary.py b/chapter4/binary.py
index 9382bcc..a16a86a 100644
--- a/chapter4/binary.py
+++ b/chapter4/binary.py
@@ -1,6 +1,5 @@
 import logging
 import random
-import time
 
 logging.basicConfig(level=logging.DEBUG)
 logger = logging.getLogger(__name__)
diff --git a/chapter9/README.md b/chapter9/README.md
new file mode 100644
index 0000000..cca7401
--- /dev/null
+++ b/chapter9/README.md
@@ -0,0 +1,5 @@
+# Shortest path for weighted graph (cost associated edges)
+
+-   Dijkstra's algorithm works when all weights are non-negative
+    -   If there are negative weights use Bellman-Ford.
+-   Priority queue + min heap is optimal when compared to a function that operates on a list.