This notebook runs network analysis of the table specified by the input_table parameter.
For input_table , this notebook assumes a transition matrix that consists of source, target, and weight/count columns. A sample table:
| source | target | weight/count |
|---|---|---|
| google.com | /learn | 10 |
| google.com | / | 7.354033 |
| NULL | / | 5.698033 |
| /about | /about | 4.249822 |
| NULL | /press_release_jp | 4.034131 |
This notebook outputs visualization of transitions. Supported network visualization methods:
- Directed Graph Visualization using networkx. Graph nodes are weighted and connected based on PageRank.
- Sankey Diagram Visualization using Plotly.
Some sample network analysis visualizations are as follows:
Find a sample workflow here in Treasure Boxes.
+network_analysis:
ipynb>:
notebook: network_analysis
input_table: ml_datasets.transition_matrix| Parameter Name | Parameter on Console | Description | Default Value |
|---|---|---|---|
| docker.task_name | Docker Task Mem | Task memory size. Available values are 64g, 128g (default), 256g, 384g, or 512g depending on your contracted tiers. | 128g |
| input_table | Input Table | Specify a TD table used for analysis as dbname.table_name. | - |
| limit | Limit | Maximum number of edges to search. | 1000 |