• operators
• data types
• string methods
• sets
• tuples
• Dictonary

• if
• elif

• conda create
• activate/deactivate
• conda list
• conda install
• conda env export > environment.yaml
• conda env create -f environment.yaml
• conda env remove -n env_name
• pip freeze > requirements.txt

• np.zeros()
• np.ones()
• np.full()
• np.eye()
• np.diag()
• np.arange()
• np.linspace()
• np.reshape()
• np.random.random()
• np.random.randint()
• np.random.normal()

• Accessing array elements.
• Modify array values.
• Delete array values.
• Append array values.
• Insert values to array.
• vstack and hstack.

• nd array slicing using index.
• np.copy()
• np.diag()
• np.unique()

Learned boolean operations, set functions and sorting.

• Boolean indexing (x[x>10])
• Set funnctions
  • np.intersect1d(x, y)
  • np.setdiff1d(x, y)
  • np.union1d(x, y)
• Sort Functions
  • np.sort(x)
  • np.sort(x,axis=0)
  • np.sort(x,axis=1)

Learned Arithmetic operations and numpy broadcasting functions.

• create Series pd.Series(data=[30, 6, 'Yes', 'No'], index=['eggs', 'apple', 'milk', 'bread']).
• get shape groceries.shape.
• get dimension groceries.ndim.
• get size groceries.ndim.
• get index groceries.index.
• get values groceries.values.
• check index in series 'banana' in groceries.

• By index label - groceries['bread'], groceries['bread','apple'].
• By numerical indices - groceries[-1], groceries[[1, 0, -1]].
• By loc() function - groceries.loc[['eggs','apples']].
• By iloc() function - groceries.iloc[[1,-1]].
• change values of the element - groceries['eggs'] = 2.
• remove element from series - groceries.drop('apple',inplace=True).

Learned to create pandas data frames.

• items = {
    'Bob': pd.Series([245, 25, 55],index=['bike', 'pants', 'watch']),
    'Alice': pd.Series([40, 110, 500, 45],index=['book', 'glassess', 'bike', 'pants'])
  }
  shopping_carts = pd.DataFrame(items)
  shopping_carts

• get the index of data frame shopping_carts.index.
• get the column names of data frame shopping_carts.columns.
• get the values of data frame shopping_carts.values.
• get the shape of data frame shopping_carts.shape.
• get the dimension of the data frame shopping_carts.ndim.
• get the size of data frame shopping_carts.size.
• create filtered data frame alice_sel_shopping_cart = pd.DataFrame(items, index=['glassess', 'bike'], columns=['Alice'])alice_sel_shopping_cart.
• create custom index stored_items = pd.DataFrame(items, index=['store 1', 'store 2']).

Learned to accessing Elements in pandas data frames

• access elements using index stored_items[['bikes']].
• creating new columns with arithmetic operations stored_items['suits'] = stored_items['shirts'] + stored_items['pants'].
• insert new column stored_items.insert(5, 'shoes',[8, 5, 0]).
• remove columns using pop stored_items.pop('new_watches').
• rename columns stored_items = stored_items.rename(columns={'bikes':'hats'}).
• set index to data frame stored_items = stored_items.set_index('pants').

Dealing with NaN values

• get the count of NaN values store_items.interpolate(method='linear', axis=0).
• remove NaN values store_items.dropna(axis=1) #inplace = true will effect the actual dataframe.
• fill NaN values with a value store_items.fillna(0).
• fill NaN values with forward and backward values (ffill, bfill) store_items.fillna(method='ffill', axis=0).
• fill NaN values with interpolation store_items.interpolate(method='linear', axis=1).

Loading Data into a Pandas DataFrame

• Load data from csv pd.read_csv('./goog-1.csv').
• list the top rows google_stock.head().
• list the last lines google_stock.tail().
• check for NaN values google_stock.isnull().any().
• get the data description google_stock['Adj Close'].describe().
• get the correlation of data google_stock.corr()
• groupby to group the data data.groupby(['Year','Department'])['Salary'].sum().

Data Vizualization

• for quantitative variable, the common plot type is histogram.
• for qualitative variable, the common plot type is bar chart.
• completed DV for univarient vizualization.

Learned a glimpse of Kernel Density Estimation.

Bivarient vizualization.

• used to look at relationship.
• 3 major bivarient plot.
  1. Scatter plots for quantitative variables vs. qualitative variables.
  2. Violin plots for quantitative variables vs. qualitative variables.
  3. Clustered bar charts for qualitative variables vs. quantitative variables.

Scatter plot

• Overplotting
  • where a plot is created with too many overlapping points.
  • resolve with
    • Sampling - ploting few numbers of data.
    • Transperecy - making point as transperent.
    • Jitter - adds a small amount of random noise to the position of each point.(used for discrete points).

Heat maps or 2D histogram

• for quantitative variables vs. quantitative variables.
• good for discrete variable vs discrete variable.
• good alternative to transperency for a lot of data.
• bin size is important.

• Violin plots for quantitative variable vs qualitative variable.

• for qualitative variable vs quantitative variable.

• multiple copies of the same type of plot visualized on different subsets of the data.

• Ploting the relationship between two quantitative variable, one on each axis.
• Emphasize relative change.
• Emphasize trends across X-values.