PANDASDATAFRAMES_SCIKIT_METRICS PPT.pptx

PANDASDATAFRAMES_SCIKIT_METRICS PPT.pptx

• 1.
  PANDA LIBRARY Introduction • “Pandas”– “Panel Data” and “Python Data Analysis” • Powerful and versatile library, simplifies the task of data manipulation in Python. • Built upon Numpy library, so lot of data structures of Numpy are used or replicated in Pandas. • Pandas – well suited for working with tabular data – spreadsheets and SQL tables. • Versatality – makes it an essential tool for data analytics, scientists and engineers working with structured data in Python. Where and why Pandas in AI? • Case-study: Marketing campaigns conducted by a company – dataset contains info about campaign type, budget, duration, reach, engagement metrics, sales performance. • Pandas – used to locate datasets, display summary statistics, perform group-wise analysis to understand the performance of the different marketing campaigns. • Data visualization – of the sales performance and average engagement metrics for each campaign – Plotted using Matplotlib – a popular plotting library in Python.
• 2.
  Pandas and AI •Pandas provide powerful data manipulation and aggregation functionalities • Makes it easy to perform complex analysis and generate insightful data visualizations. • This capability is invaluable in AI and data-driven decision making processes, allowing businesses to gain actionable insights from their data.
• 3.
  Data structures inPanda • Provides – two data structures for manipulating data. • a) Series and b) Data Frame • Series – 1d array containing a series of values of any data type (int, float, list, string etc) – default numeric data labels from 0 called ‘index’. • Panda Series – similar to a column in a spreadsheet. • Data Frame – 2d data structures – used when we need to work with multiple columns at a time – to process tabular data • Eg- result of a class, items in a restaurant menu, reservation chart of a train etc.
• 4.
  Creating a DataFrame • 2 methods 1. using NumPy ndarrays 2. using list of Dictionaries 1. Syntax array_object = np.array([list of items]) dataframe_object = pd.DataFrame([list of array_object], rows =[list of row labels], columns=[list of column labels]) Rows accessed with index or row labels. Columns accessed with column labels or key in the case of Dictionary. 2. Syntax listDict = [{key:value} pairs] dataframe_object = pd.DataFrame(listDict, index = [list of row names as strings])
• 5.
  Dealing with Rowand Columns • Over to Python Script and Shell modes.