Tool for producing high quality forecasts for time series data that has multiple seasonality with li forecast-model, python, forecasting, python.

如何安装 Prophet？

运行命令：claude skill add --url github.com/openclaw/skills/tree/main/skills/bytesagain/forecast-model

Forecast Model

Name: 时序预测
Rating: 5 (4155 reviews)
Author: bytesagain

Forecast Model v2.0.0 — an AI toolkit for managing forecasting workflows from the command line. Log configurations, benchmarks, prompts, evaluations, fine-tuning runs, cost tracking, and optimization notes. Each entry is timestamped and persisted locally. Works entirely offline — your data never leaves your machine.

Why Forecast Model?

Works entirely offline — your data never leaves your machine
Simple command-line interface with no GUI dependency
Export to JSON, CSV, or plain text at any time for sharing or archival
Automatic activity history logging across all commands
Each domain command doubles as both a logger and a viewer

Commands

Domain Commands

Each domain command works in two modes: log mode (with arguments) saves a timestamped entry, view mode (no arguments) shows the 20 most recent entries.

Command	Description
`forecast-model configure <input>`	Log a configuration note such as model parameters, seasonal settings, or data pipeline configurations. Use this to track which settings were active during each forecasting experiment.
`forecast-model benchmark <input>`	Log a benchmark result or performance observation. Record MAE, RMSE, MAPE, or other accuracy metrics to compare forecast quality across model versions and datasets.
`forecast-model compare <input>`	Log a comparison note between models, methods, or configurations. Useful for side-by-side evaluations like Prophet vs ARIMA on specific time series datasets.
`forecast-model prompt <input>`	Log a prompt template or prompt engineering note. Track iterations on how you frame forecasting tasks for LLM-based approaches or hybrid models.
`forecast-model evaluate <input>`	Log an evaluation result or quality metric. Record accuracy scores, confidence intervals, prediction intervals, or backtesting results from forecast runs.
`forecast-model fine-tune <input>`	Log a fine-tuning run or hyperparameter note. Track changepoint settings, seasonality orders, holiday effects, and the resulting forecast accuracy.
`forecast-model analyze <input>`	Log an analysis observation or insight. Record trend patterns, seasonality decomposition results, anomaly detections, or data quality issues found.
`forecast-model cost <input>`	Log cost tracking data including compute expenses, API costs for cloud-based forecasting, and resource consumption. Essential for budget monitoring across forecast pipelines.
`forecast-model usage <input>`	Log usage metrics or consumption data. Track how many forecasts were generated, data volumes processed, and resource utilization patterns.
`forecast-model optimize <input>`	Log optimization attempts or improvements. Record hyperparameter tuning results, feature engineering experiments, and their impact on forecast accuracy.
`forecast-model test <input>`	Log test results or test case notes. Record backtesting outcomes, hold-out set performance, and cross-validation results from forecast models.
`forecast-model report <input>`	Log a report entry or summary finding. Capture weekly forecast accuracy summaries, model comparison reports, or stakeholder-ready findings.

Utility Commands

Command	Description
`forecast-model stats`	Show summary statistics across all log files, including entry counts per category and total data size on disk.
`forecast-model export <fmt>`	Export all data to a file in the specified format. Supported formats: `json`, `csv`, `txt`. Output is saved to the data directory.
`forecast-model search <term>`	Search all log entries for a term using case-insensitive matching. Results are grouped by log category for easy scanning.
`forecast-model recent`	Show the 20 most recent entries from the unified activity log, giving a quick overview of recent work across all commands.
`forecast-model status`	Health check showing version, data directory path, total entry count, disk usage, and last activity timestamp.
`forecast-model help`	Show the built-in help message listing all available commands and usage information.
`forecast-model version`	Print the current version (v2.0.0).

Data Storage

All data is stored locally at ~/.local/share/forecast-model/. Each domain command writes to its own log file (e.g., configure.log, benchmark.log). A unified history.log tracks all actions across commands. Use export to back up your data at any time.

Requirements

Bash (4.0+)
No external dependencies — pure shell script
No network access required

When to Use

Tracking AI/ML model forecasting experiments and their configurations across different time series datasets
Logging benchmark results and accuracy metrics to compare forecast quality across model versions
Managing cost and usage data for forecasting API calls and compute resources
Building a local knowledge base of prompt engineering attempts for LLM-based forecasting
Exporting experiment logs for sharing with stakeholders or archiving completed forecast projects

Examples

bash

# Log a new configuration
forecast-model configure "ARIMA(2,1,2) with seasonal=True, period=7, dataset=retail_sales"

# Record a benchmark result
forecast-model benchmark "MAE=0.032, RMSE=0.048 on test set, 10k samples, horizon=14 days"

# Compare two models
forecast-model compare "Prophet vs ARIMA: Prophet wins on weekly data by 12% MAPE reduction"

# Track costs
forecast-model cost "Cloud forecasting API: $4.20 for 3 batch runs, 50k data points each"

# Search across all logs
forecast-model search "Prophet"

# Export everything as JSON
forecast-model export json

# Check overall status
forecast-model status

# View recent activity
forecast-model recent

时序预测

安装

文档

Forecast Model

Why Forecast Model?

Commands

Domain Commands

Utility Commands

Data Storage

Requirements

When to Use

Examples

相关 Skills

Claude API

并行代理

思否热榜

评论