Prophet
darts
Our great sponsors
| Prophet | darts | |
|---|---|---|
| 88 | 23 | |
| 14,431 | 4,048 | |
| 1.4% | 5.3% | |
| 6.5 | 9.4 | |
| 17 days ago | 4 days ago | |
| Python | Python | |
| MIT License | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Prophet
-
Reverse Engineering Video Game Stock Prices
Prophet is a forecasting model made by Facebook.
-
ARIMA models are solid though
I like FB Prophet and LinkedIn's GreyKite
-
LSTM/CNN architectures for time series forecasting[Discussion]
Prophet
-
Basic Time Series Prediction
I like Meta's Prophet. It's a flexible model that I feel is easy to understand and explain to non-technical stakeholders. Implementations for R and python. It can be as simple as a univariate series, or accept exogenous explanatory variables (categorical & continuous). Seasonality is part of the model, out of the box. It also has decent default parameters if you're not looking to do a lot of tuning. Lastly, the documentation is quite thorough and approachable. https://facebook.github.io/prophet/
- prophet: NEW Data - star count:14301.0
darts
-
Basic Time Series Prediction
Try Dart https://unit8co.github.io/darts/ Has many inbuilt statistical techniques for time series prediction.
-
Forecasting three months ahead.
Darts can be used to train ML-based forecasting models on tens of thousands of time series in a few lines of code only. Such a model can then be used for fast inference (e.g., it takes 1-2 seconds to forecast 1,300 time series in some of the experiments we conducted). Here is a completely self-contained notebook with an example of how to do this on large datasets with different kinds of models: https://github.com/unit8co/darts/blob/master/examples/14-transfer-learning.ipynb
-
tsqsim's benchmark (new tool designed for Monero Research Lab)
darts
-
Interactive Timeseries Forecasting! Predicting the future with Darts + Streamlit
I wanted to explore the claim of "Time Series Made Easy in Python" by the Darts library. Turns out it takes ~12 lines of code including imports to get started with Darts.
-
[P] Beware of false (FB-)Prophets: Introducing the fastest implementation of auto ARIMA [ever].
To name a few: https://github.com/jdb78/pytorch-forecasting, https://github.com/unit8co/darts, https://github.com/Nixtla/neuralforecast
-
Discussion: How would you approach a time series problem with spatial components (Spatio-Temporal Forecasting)
Moreover I found the DARTS Library by Unit 8 which offers a nice tool case for time series problems, however they unfortunately don't seem to include and possibilities for covering space aspects, so using this I would need to train an independent model for each cluster and totally loose and space relations between them.
-
Why are there so many more statistics courses in R and so few in Python?
You should check out darts for time series in python, its like if sklearn was expanded to cover an impressive quantity of time series algos: https://unit8co.github.io/darts/
-
Time series forecasting with non-temporal information
This notebook by darts library should help to answer your question: https://github.com/unit8co/darts/blob/master/examples/01-multi-time-series-and-covariates.ipynb
-
Automated Time Series Processing and Forecasting
Great, another open source tool purporting to solve time series analysis in an "automated way" (lol @ attempting automated statistics) that my manager will link me tomorrow and ask me to review.
Why should I use this over Darts[1] or just Statsmodels[2], if I need more lower level access and diagnostics? Both of these are far more established.
I dislike that Facebook Prophet was chosen as a benchmark; it's not a difficult benchmark to beat for the majority of time series use cases. It signifies to me that this project might targeting cargo cult data science. Prophet is not particularly good at non-daily timeseries and non-seasonal timeseries. The paper itself admits this[3]. Moreover, it's just a generalized additive model that incorporates holidays.
--
1 https://github.com/unit8co/darts
-
Comparison of Time Series Forecasting Discussion
How is this different from Unit8's Darts (https://github.com/unit8co/darts)?
What are some alternatives?
tensorflow - An Open Source Machine Learning Framework for Everyone
xgboost - Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library, for Python, R, Java, Scala, C++ and more. Runs on single machine, Hadoop, Spark, Dask, Flink and DataFlow
scikit-learn - scikit-learn: machine learning in Python
pytorch-forecasting - Time series forecasting with PyTorch
greykite - A flexible, intuitive and fast forecasting library
sktime - A unified framework for machine learning with time series
MLflow - Open source platform for the machine learning lifecycle
Keras - Deep Learning for humans
Kats - Kats, a kit to analyze time series data, a lightweight, easy-to-use, generalizable, and extendable framework to perform time series analysis, from understanding the key statistics and characteristics, detecting change points and anomalies, to forecasting future trends.
tsai - Time series Timeseries Deep Learning Machine Learning Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai