data frame visualization library using plotly (100$)
Budget: $30 – $250 AUD
I need to build a library that visualizes the data frame. you will have the input as a pandas dataframe, target column, and a dictionary describing each column and its type. The type can be numerical, categorical, binary or time series.
The function should return (1) an image of the plot (2) a path for the saved plot
For each dataframe you will provide 3 sets of plots as described below:
(1) 3 heatmaps describing the relation between the columns and each other
a. heatmap1 (will describe the relations of numerical cols vs numerical cols using Spearman correlation)
b. heatmap2 (will describe the relations numerical cols vs categorical cols using ANOVA correlation)
c. Heatmap3 (will describe the relations between categorical cols vs categorical cols using Chi2 test)
(2) N Histogram charts, where N is the number of columns we have, where each column will be described as a histogram (bar chart + fitting line) , this histogram will describe how many times each value in the column has occurred
(3) N-1 Target charts, where N is the number of columns in the dataframe. This chart describes the target values for each value on the x-axis (if the target is numerical you can use Bar chart with whiskers and boxplot) and if the target is categorical, you should use a scatter plot, horizontal violin plot and horizontal boxplot). The target values may need aggregation).
The deliverable will be (1) a function for every plot (2) a function that calls the plotting function for all of them (3) a restful web service that calls each function and the main function as well (4) the output will be the plot image and the path to the saved image
The function should return (1) an image of the plot (2) a path for the saved plot
For each dataframe you will provide 3 sets of plots as described below:
(1) 3 heatmaps describing the relation between the columns and each other
a. heatmap1 (will describe the relations of numerical cols vs numerical cols using Spearman correlation)
b. heatmap2 (will describe the relations numerical cols vs categorical cols using ANOVA correlation)
c. Heatmap3 (will describe the relations between categorical cols vs categorical cols using Chi2 test)
(2) N Histogram charts, where N is the number of columns we have, where each column will be described as a histogram (bar chart + fitting line) , this histogram will describe how many times each value in the column has occurred
(3) N-1 Target charts, where N is the number of columns in the dataframe. This chart describes the target values for each value on the x-axis (if the target is numerical you can use Bar chart with whiskers and boxplot) and if the target is categorical, you should use a scatter plot, horizontal violin plot and horizontal boxplot). The target values may need aggregation).
The deliverable will be (1) a function for every plot (2) a function that calls the plotting function for all of them (3) a restful web service that calls each function and the main function as well (4) the output will be the plot image and the path to the saved image