"Python for data analysis" interactive course
Mastering Python from scratch
Mastering Python from scratch
3 months / 50 mini sessions / certificate
Starts June 15 / from 24 900 rubles
Starts June 15 / from 24 900 rubles
Book minimum price
Apply and our manager will secure the price for you
By clicking the button, I consent to the processing of personal data and agree to the privacy policy и and the offer agreement.
Call me back
Apply and our manager will call you back
By clicking the button, I consent to the processing of personal data and agree to the privacy policy и and the offer agreement.
Why Python?
- Python is one of the easiest languages to learnPython is an ultra-high level language with simple syntax.
- Python is used by the world's largest companies: Google, Facebook, Instagram, Dropbox.For example, parts of the Youtube and Youtube API are written on Python
- Python is a universal environment for scientific calculations.Python with NumPy, SciPy and MatPlotLib is actively used as a universal environment for scientific calculations as a replacement for common commercial Matlab, IDL and other packages.
For example, the Astropy library is a popular tool for astronomical calculations. - The world's leading neural network libraries have an API on PythonTensorflow, PyTorch, Keras are neural network libraries with Python shells. The libraries are aimed at operative work with networks of deep learning.
- Starting salary of Python data analysis specialist is 120.000 rublesAccording to statistics from hh.ru, the starting salary of a Python data analyst (Junior Data Scientist) is 120,000 rubles.
- Python is one of the 5 most popular programming languages.TIOBE programming community index is an index that measures the popularity of programming languages. According to this index, Python is in the top 5 most popular programming languages in 2019.
Python is a high-level programming language focused on improving developer productivity and code readability.
The syntax of the Python core is minimalistic. At the same time, the standard library includes a large number of useful functions
The syntax of the Python core is minimalistic. At the same time, the standard library includes a large number of useful functions
Course program
Contents:
Practical assignments:
- Why Python?
- Using the Terminal
- Installing Anaconda on Windows, Linux, MacOS
- Connecting to remote servers, Putty/ssh
- Launching JupyterNotebook
- Code design rules: PEP8, PEP20, PEP257
- Basic operations: Python as calculator
- Variables and types, type conversion
- Basic constructs: conditional operator, if...else construct
- Cycles: for, while
Practical assignments:
- Using the instruction from the lecture, install Anaconda, run JupyterNotebook via terminal
- Display "Hello World!"
- Task to work with conditional operator
- Using cycles
- Example: display all letters in the word "Concatenation" one by one, display only consonants, only unique characters; calculate the sum of numbers from 1 to 100.
Contents:
Practical assignments:
- Built-in structures:
- list
- tuple
- dict
- set, frozenset
- stack and queue
- Functions:
- Signature
- Visibility area
- Default arguments
- Argument transfer
- Lambda-functions
- Exceptions
- Generators
Practical assignments:
- One task to use each of the structures
- Rewrite task solution from homework assignment 1 with the for cycle using the lambda-function
- Write a function that calculates the weighted average
- Add "foolproofing" to the function from the previous task, use exceptions
- Write a generator that divides the list into non-intersecting subsets of constant length
Contents:
Practical assignments:
- Packages
- Classes:
- Design
- Attributes
- Methods
- Privacy
- Magical methods
- Inheritance
- Abstract class
- Multiple inheritance
Practical assignments:
- Write a class with specified attributes and methods, save in .py files
- Load the module/modules into a new notebook, demonstrate correct operation of codes
Contents:
Practical assignments:
- np.ndarray structure:
- Data types
- Indexing
- Arithmetic operations
- Type conversion
- Operations with arrays
- Multivariate arrays
- Multivariate arithmetic
- Basic statistical functions:
- np.sum, np.mean…
- Sampling from distributions
- Matrix operations
- Linear algebra np.linalg
Practical assignments:
- Read previously prepared one-dimensional data using the corresponding function in NumPy
- Write several simple functions that perform the specified arithmetic operations with data (for example, calculate the sum of all array elements, the sum of elements not multiple of 10)
- Compare the performance over time of your functions with the functions built into the library
- Load a pre-prepared multivariate array
- Make a subsample of an array that meets certain conditions, to perform specified matrix operations with it
- Write a function to search for a reverse matrix to a given one, compare the result and performance with the built-in function in the library
Contents:
Practical assignments:
- Abandoning cycles:
- Vector operations
- Tensor operations
- Special structures:
- Sparse matrices
- Toeplitz matrices scipy scipy.linalg.toeplitz
- QR decomposition
- LU decomposition
- SVD decomposition
Practical assignments:
- Write a function that calculates the sum of squares of elements of a given array
- Write a function that does the same without using loops
- Compare results and performance over time
- The problem on the alternative to the cycle for c using the Toeplitz matrices
- Implement SVD decomposition without using cycles, compare the result with the built-in function
Contents:
- pd.Series and pd.DataFrame structures
- Basic functionality
- Reading data from different file types
- Sub samples
- Indexing, types of indexes
- Multi-indexing
- Concatenation, Join several tables
- Summary tables
- Read pre-prepared data using the corresponding function in the Pandas library
- Display a sub-sample of the dataset that meets some of the specified conditions by column value
- Rename the specified column
- Add a new column for the sum of values in the specified columns
- Read data from the second set table, make left join from the first one
- Make a summary table with the specified parameters
- Make a subsample of the summary table according to the given logical conditions (using the multi-index)
Contents:
- Datetime index
- GroupBy
- Window functions
- Resampling
- Built-in statistical functions
- Principles of working with text data
- Read data, convert index to datetime
- Calculate a simple moving average in the window of a given length
- Convert minute data into 10-minute data
- Convert 10-minute data into minutes using the backfill method
- Task on groupby: calculate quantiles, average, median for a given grouping level
- Text data task
Contents:
Practical assignments:
- Displaying one-dimensional and two-dimensional data
- Graph properties:
- Axes names
- Scale
- Axes scaling
- Drawing options
- Color scheme
- Axes: Several graphs as a single object
- Common axes
- Scaling, polar coordinates
- Drawing categorical data
- Charts, histograms
- Contour graphs
- Heatmap
- Drawing of images, scales
- Affine image transformation
Practical assignments:
- Build sine graph, sign the axes
- Build a sinusoid graph in polar coordinates
- Load a known standard dataset, such as Iris.
- Build a scatter plot of several variables on one chart, add graph name, axes names, labels, color class labels
- Build 4 variable graphs from the dataset as one object in the logarithmic scale, select font size, scale, sign each graph
- Build a histogram with a specified number of bins for a specific variable from the array above
- Calculate the correlation matrix and build a heatmap
- Using the instruction, download the .jpg picture and display it in black and white, in color
- Rotate the picture at a given angle and display the image.
Contents:
- Drawing of statistical relationships
- Drawing categorical data
- Visualization of distributions
- Control of main graph parameters
- Load a known standard dataset, e.g. Iris.
- Set the parameters of the graphs: dark background and casting (darkgrid), set the color palette to suit the taste of students
- Build graphs of paired variable dependencies using the built-in function
- Build boxplot, violinplot.
- Load a standard dataset for a regression task, e.g. bostonhouses, draw a graph with drawing of the regression line (sns.jointplot).
- Build histograms with distribution drawing using the built-in function
- Based on the graphs drawn, make conclusions about the dependencies in the dataset.
Contents:
- Loading, reading data
- Missed value monitoring, filling options
- Coding different types of categorical data
- Descriptive statistics
- Visualization of paired relationships
- Variable distribution visualization
- Heatmap similarity matrix
- Select a known standard dataset or use the data you would like to learn for your work tasks
- Load the data, check for Nan/Null
- Choose how to fill in the passes or throw out the corresponding rows, justify the choice.
- If there are categorical variables in the dataset, encode them using one of the methods described in the lesson, justify the choice.
- Calculate the mean, median, standard deviation and other statistical indicators for all numeric variables
- Build all the graphs you think are relevant from the previous homework, justify the choice.
- Calculate the correlation matrix, draw heatmap using the seaborn library.
- Make conclusions about the found interrelationships, substantiate in detail
Book minimum price
Apply and book the price of 24.900 rubles
instead of 34.900 rubles
instead of 34.900 rubles
What will you learn from the course
- Basic syntax and basics of Python programming
- Installing the Python distribution kit - Anaconda and other auxiliary tools, connecting to remote computing servers
- Basic concepts of scientific programming on Python: from working with different data types and built-in structures to writing your own modules.
- Conducting descriptive data analysis (EDA): from reading and pre-processing to searching and visualizing relationships
- High level vector computing using NumPy, SciPy libraries
- Manipulation and analysis of data using Pandas library structures
- Visualization with libraries Matplotlib, Seaborn
"Python for data analysis"
course teacher
course teacher
Oleg Konorev
Moscow
Works in application of classical machine learning and neural networks in various tasks related to the processing of natural language (NLP). Experience in developing models for image recognition. Technologies and tools: Python 3, Pandas, Matplotlib, Numpy, scikit-learn, Keras/Tensorflow.
Course curator
Marat Khaidarov
Kazan
Technical education: Engineer of automated information processing and control systems.
Additional Education: Neural Networks and Deep Learning, Coursera
Work experience: Senior SCM Analyst; Logistics Supervisor at Ford Sollers Holding - more than 5 years (tasks: predictive analytics, process automation and optimization, assessment and improvement of key metrics of the department).
At the University of Artificial Intelligence worked in the following areas: classification of texts, generation of texts (chat bots), object recognition, speech recognition, training with reinforcements, networks with attention.
Additional Education: Neural Networks and Deep Learning, Coursera
Work experience: Senior SCM Analyst; Logistics Supervisor at Ford Sollers Holding - more than 5 years (tasks: predictive analytics, process automation and optimization, assessment and improvement of key metrics of the department).
At the University of Artificial Intelligence worked in the following areas: classification of texts, generation of texts (chat bots), object recognition, speech recognition, training with reinforcements, networks with attention.
Training format
50 mini-sessions
Each session is a small 15-20 minute video, discussing one small topic.
All videos are available to you on a special training platform. You can review important points and refresh your memory.
You can study these sessions in any order you like
All videos are available to you on a special training platform. You can review important points and refresh your memory.
You can study these sessions in any order you like
Practical assignments
Practical tasks will be available for you to work on your own.
Tasks are optional, but we strongly recommend practicing them. At any time, you can proceed to the next session.
Tasks are optional, but we strongly recommend practicing them. At any time, you can proceed to the next session.
Curator support
All communication on the course is organized on our training platform.
If you have any questions about the course material, write to the curator, who will help you understand the topic of the session or the practical task.
If you have any questions about the course material, write to the curator, who will help you understand the topic of the session or the practical task.
Certificate
Upon completion of the course you will receive a certificate on the University letterhead.
Our manager will answer any questions
- Artem Voronov-GashevCustomer service managerEmail: info@neural-university.ru
Phone: +7 (499) 648-67-44
Whatsapp/viber/telegram:
+7(918) 916-41-84
"Python for data analysis" interactive course
Mastering Python from scratch
3 months / 50 mini-seesion / certificate
Starts June 15 / price from 24 900 rubles
Starts June 15 / price from 24 900 rubles