用于将包含自定义时间列（如'bob'）的Excel金融数据读入Backtrader。该技能包含处理时间格式转换、映射OHLCV列，以及通过Pandas倒序解决K线图时间轴反向问题的完整流程。

使用Python根据基准文件的第一列顺序，对多个Excel文件进行行重排序，保持文件内容不变且输出结构一致。

针对股票日度面板数据，按日期分组后，根据各指标列的30%和70%分位数筛选出底部和头部的股票代码，并横向合并为DataFrame。

针对具有Datetime索引的时间序列DataFrame，计算指定日期在历史上同月同日数据中的相对分位值（基于最大最小值计算）。

用于从Excel文件读取问题，调用LLM接口获取答案，将答案格式化为JSON字符串后写回Excel指定列的自动化脚本任务。

使用 Pandas 和 SQLAlchemy 将 DataFrame 同步到 MySQL，处理 Merge 后缀，并应用复杂的字段比较逻辑（JSON解析、数值归一化、条件排除、字符串排序）以实现精确的 Upsert。

用于读取、预处理股票5分钟K线CSV数据，并使用tslearn库的TimeSeriesKMeans进行聚类分析的技能。包含数据清洗（长度过滤、NaN过滤）、百分比变化计算、模型训练、保存及代表性样本提取。

用于将Pandas DataFrame同步到MySQL数据库的技能，包含针对JSON字段、数值类型（int/float）、字符串顺序（如effect_desc）的归一化比较逻辑，以及基于type字段的条件排除规则。

用于在Pandas DataFrame执行合并操作后，清理由merge产生的后缀列（_x, _y）和指示列，使其符合数据库表结构以便进行to_sql插入操作。

使用Python和pandas读取Excel文件，对比“依存关系”和“识别的依存关系”列，在剔除特定符号（如?）后计算正确率和错误率，并将不匹配的记录保存到新的Excel文件中。

使用Python Tkinter Treeview组件读取Excel文件，实现点击列头排序、添加动态行序号、双击单元格弹窗显示内容以及自定义列宽的功能。

Advanced Python visualization expert supporting Pandas DataFrames and raw data. Handles dynamic single/dual-axis plotting, Chinese font configuration, Tensor conversion, and safe file saving without display.

使用Hugging Face Transformers和sacrebleu库，对预训练的机器翻译模型进行评估。支持从制表符分隔的CSV文件读取源文本和参考翻译，计算BLEU分数，并提供Seq2SeqTrainer的compute_metrics函数实现。

读取股票5分钟K线CSV文件，按日期分组，计算基于昨日收盘价的百分比变化，过滤无效数据（长度不一致、NaN、Inf），进行Z-score归一化，并使用TimeSeriesKMeans（DTW/SoftDTW度量）进行聚类。

Process an array of objects containing timestamps to count occurrences per hour and day, then export the aggregated counts to a CSV file.

Implement a high score tracking system for Codesters games using CSV files. This involves reading existing scores, appending new scores, sorting them, writing back to the file, and displaying the results on the stage.

Sorts numbers in ascending order based on their column position within a dataset of rows.

Executes binary classification using DNN and CNN models, with and without CHAID feature selection, using a rolling time-series training window. Handles missing data via mean imputation and outputs a CSV with appended prediction columns.

Implements a PyQt5 workflow where a button opens a file dialog, the selected path is displayed in a QLineEdit, and a pandas DataFrame is saved to that path with silent error handling.

Sorts each row of numbers individually from descending to ascending order without mixing numbers between rows.

Process invoice data vectors for analytics and charting. Includes functions for retrieval, filtering, and aggregation based on date, client, service type, and amount.

Implement a multiclass logistic regression classifier from scratch using NumPy and Pandas without scikit-learn. Use the One-vs-Rest strategy to handle multiple classes (e.g., 0, 1, 2) and save the trained model coefficients to a pickle file.

Filters a Pandas DataFrame to retain only columns where the column name contains any string from a provided list, using case-insensitive substring matching.

Implement a multiclass logistic regression classifier from scratch using NumPy and Pandas, avoiding libraries like scikit-learn. The implementation uses a One-vs-Rest strategy to handle multiple classes (e.g., 0, 1, 2) and saves the trained model coefficients to a pickle file.

Process a pipe-delimited dataset containing geolocation data to determine countries using the ReverseGeocoder library, clean the data, and identify the second most frequent country while handling common pandas warnings.

Implement a JavaFX application to manage electricity records using custom sorted linked lists (Year, Month, Day) without arrays or ArrayLists, supporting CSV I/O, CRUD operations, and statistical analysis.

A skill to join user activity and user info CSV datasets using PySpark 1.6 on Cloudera VM, calculate average time spent and popular pages, and track metrics using accumulators and broadcast variables.

Implements a Python-based query executor that converts informal keywords to SQL commands, parses queries preserving string literals, and executes operations on pandas DataFrames.

Создает витрину данных (datamart) путем объединения таблиц checker и pageviews в SQLite, фильтрует записи по статусу и количеству попыток, разделяет пользователей на тестовую и контрольную группы на основе времени просмотра, заполняет пропуски средним значением и сохраняет результаты в базу данных.

指导用户如何继承bt.feeds.PandasData类，通过定义lines和params将包含自定义列名（如symbol, frequency, open, close等）的Pandas DataFrame正确映射到Backtrader回测引擎中。

使用Python標準庫csv計算CSV文件中第一列目標序列與後續列的相似度，不依賴pandas或SequenceMatcher。

使用Python Pandas对Excel文件中指定列进行分组去重，当目标列值为空（NaN）时不执行去重操作。

对数据按指定列（如年龄）进行分组OLS回归分析，提取所有自变量的系数、t值和p值，并将结果整理保存为CSV文件。

使用Python处理包含SNP基因型和表型数据的CSV文件，筛选纯合基因型，计算表型均值并找出每个位点表现最好的基因型。

根据用户指定的数据主题、时间范围和列结构，将数据整理并导出为CSV格式。

针对包含多只股票和多指标的日度平衡面板数据，按日期分组后，根据各指标的分位数阈值（如前30%或后30%）筛选股票代码，并将结果横向合并为DataFrame。

Build event streaming and real-time data pipelines with Kafka, Pulsar, Redpanda, Flink, and Spark. Covers producer/consumer patterns, stream processing, event sourcing, and CDC across TypeScript, Python, Go, and Java. When building real-time systems, microservices communication, or data integration pipelines.

>

Draw ROB2 risk-of-bias plots, including a Traffic Light Plot and a Summary Bar Plot. Input is a CSV file with ROB2 assessments for each study; output are two PNG plot files.

Statistical visualization library integrated with pandas; use it when you need fast EDA of distributions, relationships, and categorical comparisons (e.g., box/violin/pair plots and heatmaps) with strong default aesthetics on top of matplotlib.

Import local literature into a managed library; trigger when you need offline deduplication, tagging, and a searchable index.

Panel data analysis with Python using linearmodels and pandas.

Clean and transform messy data in Stata with reproducible workflows

Run an end-to-end data analysis in R or Python: load, explore, analyze, and produce publication-ready output.

>-

Scans notebooks for data file references and verifies each file exists on disk. Use when checking for broken data paths.

Load, explore, clean, and analyze CSV data with statistical summaries

Systematic data cleaning workflows for research datasets

Upload messy CSVs with minimal prompting for deep automated analysis

Clean, transform, and validate messy research data using Stata