Gist f2b16fc0ccd955514d00fbf44683516f

100%

Gist f2b16fc0ccd955514d00fbf44683516f

✓ Published0🌍 Public

Last edited May 26, 2021

Created on May 26, 2021

This example demonstrates a temporal split of simulated financial transaction data into training and test sets for fraud detection. The code loads transaction records spanning from July 25 to August 14, 2018, then partitions them into three consecutive seven-day windows: training, delay, and test periods. It uses the custom `read_from_files` function to ingest the data and `get_train_test_set` to perform the split, while counting fraudulent transactions with the `TX_FRAUD` column. The visualization relies on Pandas DataFrames and Python’s `datetime` module to manage the time-based segmentation.

AI-generated description

Similar vizzes

Loading thumbnail…

Gist 392220bb33ff81da26fe5b1787cd505b

This example visualizes the results of a fraud-detection decision tree model trained on transaction data, showing how the model classifies transactions as fraudulent or non-fraudulent. It displays the test set predictions alongside the true labels, highlighting the top 100 predicted fraud cases for performance assessment. The code uses scikit-learn’s `DecisionTreeClassifier` with a maximum depth of two for interpretability, and evaluates predictions using a custom `performance_assessment` function. The dataset includes engineered features such as transaction amounts, weekend/night indicators, and customer/terminal historical statistics over one, seven, and thirty-day windows.

YYannael

74% match

Loading thumbnail…

Gist b5bb72e72049a84c7a5a0b3f0e1a3498

The example shows how to load and prepare simulated credit-card transaction data for fraud-detection analysis. It initializes a Python environment by downloading shared helper functions and cloning a GitHub repository containing pre-transformed datasets. The code uses the `%run` magic command to execute `shared_functions.py` and checks for local data before cloning, ensuring reproducible access to the simulated data source. This setup is a prerequisite for subsequent visualization steps, though no chart or rendering API is invoked in this initialization phase.

YYannael

69% match

Loading thumbnail…

Fork of Transaction Analysis with Time (Hopefully)

This visualization analyzes a personal transaction dataset, showing spending patterns across categories and time. It uses a scatter plot to display individual transactions by category and amount on a logarithmic scale, alongside a time-based chart for chronological trends. Users can brush the time chart to filter the scatter plot, and click on categories to filter the time view. The code uses the Vega-Lite API with D3.js to load data from a CSV file and render linked selections between the two views.

Ggyan9427

54% match

Loading thumbnail…

Temporal Analysis

This visualization presents a temporal and categorical analysis of financial transactions, filtering by AccountID, Channel, and TransactionType. It shows median amounts by type and channel, transaction counts by hour and day, and a volume heatmap. The D3.js v7 code reads CSV data from a gist, renders SVG charts, and updates them based on user inputs, with an option to highlight anomalies at or above the 95th percentile.

SSreedevi Itta

53% match

Loading thumbnail…

Transaction Analysis with Time

A financial dashboard combines a temporal scatter plot and a time-series bar chart to analyze personal transactions, with linked selections across both views. The upper chart displays credit/debit amounts over time, while the lower plot correlates category, amount, and date. Built using D3.js v7 and Vega-Lite API within React, it renders via the `vl.vconcat` layout, reads CSV data from a GitHub gist, and includes interactive interval and point selection parameters for coordinated filtering.

CChris Fritz

53% match

Loading thumbnail…

Tut3

This visualization demonstrates a machine-learning workflow for splitting a Brexit-related dataset into training and testing subsets. It shows how 70 percent of the data is randomly sampled for model training, with the remaining 30 percent held out for validation. The code uses base R’s `sample()` function and `nrow()` to compute indices, then applies bracket indexing to partition the data frame. The example highlights a common data-preparation step, with the split proportion clearly visible in the sampling logic. No plotting or rendering libraries are used; the focus is purely on the data-splitting mechanism.

BBenHeubl

53% match

Loading thumbnail…

Gist 25342708e9edeb5099b496447772d743

The visualization shows a technique for analyzing log files by measuring time gaps between consecutive timestamped entries. It uses Python’s `datetime.strptime` to parse timestamps from `"YYYY-MM-DD HH:MM:SS.mmm"` formats and calculates differences in milliseconds, flagging intervals exceeding 10 ms. The code filters log lines by keyword using standard file I/O operations and prints matching entries, or reads all lines from a file to identify timing anomalies. This example demonstrates text-based log inspection rather than graphical rendering, relying on Python console output for insights.

JJastor11

52% match

Loading thumbnail…

Transaction Analysis with Time (Hopefully)

This visualization pairs a scatter plot of transaction amounts by category with a time-series view of the same data, enabling linked filtering through interval and point selections. The top chart plots logged amounts against categories, while the bottom chart displays amounts over time from a CSV file of personal financial transactions. Built with Vega-Lite and D3, the charts use cross-filtering to highlight correlated selections, with tooltips showing amount, description, date, credit/debit status, and category.

CChris Fritz

52% match