WEBVTT 00:00.080 --> 00:00.880 Welcome back. 00:01.160 --> 00:02.240 Go to Colab. 00:03.280 --> 00:03.720 Open. 00:03.720 --> 00:04.440 Colab. 00:04.480 --> 00:10.320 Go to upload, browse and select file number five ML with Python. 00:10.320 --> 00:11.160 Click okay. 00:11.360 --> 00:18.840 And by the way, as I told you in the previous videos, all those files are available to download in 00:18.880 --> 00:21.160 the resources folder under each section. 00:21.160 --> 00:26.360 And you can download them from the first section, all in one package. 00:26.400 --> 00:26.840 Okay. 00:27.040 --> 00:31.480 Now this is our ML with Python notebook. 00:31.560 --> 00:34.240 Introduction to Machine Learning with Python. 00:34.240 --> 00:35.000 Welcome. 00:35.000 --> 00:40.400 In this hands on notebook you'll learn what is machine learning. 00:40.560 --> 00:42.680 Types of ML supervised. 00:42.720 --> 00:51.400 Unsupervised, how to use Python for ML, and a full example predicting house prices with linear regressions. 00:51.400 --> 00:57.560 Tools used pandas, matplotlib, NumPy and scikit learn. 00:57.600 --> 01:04.450 So let's introduce our new libraries that we're going to use in Python. 01:04.450 --> 01:07.930 First of all, we need to use numpy. 01:08.170 --> 01:14.210 NumPy in Python is a powerful library used for numerical and scientific computing. 01:14.330 --> 01:23.610 It provides multi-dimensional arrays ndarray that are much faster and more efficient than Python lists, 01:23.850 --> 01:32.410 supports mathematical operations like linear algebra, statistics, trigonometric and etc. on arrays. 01:32.570 --> 01:38.770 Often used in data science, machine learning, and scientific research. 01:38.930 --> 01:46.810 Because of its speedy speed and ability to handle large data sets, works well with other libraries 01:46.810 --> 01:51.250 like pandas, SciPy, TensorFlow, and others. 01:51.250 --> 01:57.010 In short, numpy equals to fast array plus math operations. 01:57.010 --> 02:00.810 The foundation of numerical computing in Python. 02:00.810 --> 02:03.470 So this is a very important library. 02:03.510 --> 02:04.350 Pandas. 02:04.350 --> 02:06.750 We talked about pandas in the previous videos. 02:07.030 --> 02:10.710 We talked about matplotlib in the previous videos. 02:10.750 --> 02:14.030 Now we're going to talk about sklearn. 02:14.190 --> 02:24.990 Scikit learn or sklearn is a popular machine learning library in Python built on top of NumPy, SciPy 02:24.990 --> 02:30.270 and Matplot library provides easy to use tools for data processing. 02:30.710 --> 02:37.270 Data processing includes scaling, encoding, splitting, data sets, machine learning algorithms, 02:37.390 --> 02:40.230 classification, regression, clustering. 02:40.270 --> 02:48.230 As we talked about in the previous videos, and model evaluation, cross validation metrics and selections 02:48.230 --> 02:53.110 well suited for small to medium sized data sets. 02:53.310 --> 02:58.270 Widely used in AI, data science, and predictive modeling. 02:58.470 --> 03:05.760 In short, scikit learn a user friendly toolkit for building, training and evaluating machine learning 03:05.760 --> 03:07.560 models in Python. 03:07.680 --> 03:15.640 Later we're going to use TensorFlow, but scikit learn if you are working on smaller data sets and traditional 03:15.680 --> 03:22.600 ML, for example classification with decision trees, go and use it using TensorFlow or PyTorch. 03:22.640 --> 03:28.960 If you are building deep learning models, for example image recognition, NLP object detections. 03:29.240 --> 03:31.520 Later we're going to use TensorFlow. 03:31.720 --> 03:36.880 But in this notebook I'm going to show you how to use scikit learn. 03:36.920 --> 03:42.240 If you are using a simple data sets for classification. 03:42.440 --> 03:45.960 And later we're going to work with TensorFlow. 03:45.960 --> 03:52.880 So don't worry we're going to learn and and work with different types of libraries. 03:52.920 --> 03:56.000 Click run to import those libraries. 03:56.200 --> 04:00.520 And when you get libraries imported successfully you are good. 04:00.560 --> 04:01.480 And here we go. 04:01.720 --> 04:04.520 All libraries are imported successfully.