tsqsim
darts
tsqsim | darts | |
---|---|---|
4 | 47 | |
7 | 7,366 | |
- | 2.8% | |
0.7 | 9.1 | |
about 1 year ago | 6 days ago | |
C++ | Python | |
GNU Affero General Public License v3.0 | 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.
tsqsim
-
mj's dev report Apr.-May 2022 (2/3)
Hey Lynnaignet, it's very nice that you asked. My proposal is fully funded and I receive regular payments, as agreed, therefore so far I can't complain about this part. I do have a "tip" address, that can be found below, if you like: http://monerodevs.org/monero-research-lab.html But at the current stage, a nice word is enough. Not only for the emotional support, that I sometimes do need after being trashed by the restless troll(s). The thing is, that many Anons will simply never say a word, but will at most upvote comments as direct as yours. This alone gives me confidence, that I'm going in the right direction, despite the mentioned reviewing/merging resistance. It helps me plan the future proposals better. Then, after this non-verbal communication is done and the proposal is laid out, the Anons will just drop their Moneros in quite large quantities individually, onto the proposal's address. I find it quite interesting. What more can be done? I'd suggest that whenever you see a CCS Proposal of a Developer that sounds concrete and modest at the same time, like these: https://ccs.getmonero.org/proposals/seraphis-wallet-poc.html https://ccs.getmonero.org/proposals/j-berman-3months-full-time-2.html https://ccs.getmonero.org/proposals/tobtoht-feather-dev-2021-3.html https://ccs.getmonero.org/proposals/Rucknium-OSPEAD-Fortifying-Monero-Against-Statistical-Attack.html , then go ahead and fund them. These Devs with their social skills will typically be of a large value in whatever area that they focus on. Some time later they will stumble upon my PRs, after their high-priority tasks are done. So I hope at least. This would be a more passive way to help. An active way to help, that I'd normally not promote, as it sounds statist and bureaucratic, is to go to my PRs directly and show some support via emoticons, maybe writing some text to wake others up. If this doesn't help, then pinging potential reviewers directly (namely those, who are currently being funded by YOUR money after all) would be "The Last Card" here, but this should be avoided if possible, I think. Lastly other developers are indeed affected, indeed. Perhaps it sounds weird and I don't understand it fully, but they will usually accept the status quo, because they haven't yet seen a C++ project that compiles fast and delivers more functionality, than a typical C project would. I'm trying to prove that it's possible with my tsqsim project.
-
tsqsim's benchmark (new tool designed for Monero Research Lab)
pip3 install statsmodels # New optional dependencies pip3 install darts git clone --recursive https://github.com/mj-xmr/tsqsim.git # Clone this repo (assuming it's not a fork) cd tsqsim # Enter the cloned repo's dir rm build/* -fr || true # Clear up previous configuration files git checkout benchmark # checkout the relevant branch ./util/prep-env.sh # Prepare the environment - downloads example data and creates useful symlinks ./util/deps-pull.sh # Download the maintaned dependencies ./util/deps-build.sh # Build and install the unmanaged dependencies (uses sudo for installation) ./util/build.py --benchmark --plot
-
mj's dev report: Dec. 2021
I was very happy to work overtime in the 1st half of December, since I knew that family matters would slow me down later on. Therefore this month’s work was clearly divided into two portions. In the 1st one I was doing a lot of the necessary Continuous Integration (CI) work and made sure that the software works on all platforms, that are available on GitHub CI, which are:
-
mj's dev report: Nov. 2021
For the whole month I’ve worked on being able to deliver the first public version of my Time Series Quick Simulator (dubbed "tsqsim"), that aims to support the Monero Research Lab in detecting and predicting transaction patterns, however the Researchers want to use them at a given time. The goal has been achieved and you can use the quite simple instructions written here, to see the first public version in action (so far Debian / Ubuntu 20.04 only):
darts
- Darts: Python lib for forecasting and anomaly detection on time series
-
[D] Doubts on the implementation of LSTMs for timeseries prediction (like including weather forecasts)
Don't use an LSTM. Get up to date with SoTA methods and read the papers in the field. LSTMs are not the way forward. Read the papers I suggested. It would be very useful to come to grips with both the Time Series Repository (https://github.com/thuml/Time-Series-Library) and Darts (https://github.com/unit8co/darts) as these are widely used for research and in industry.
-
Facebook Prophet: library for generating forecasts from any time series data
As others have pointed out, Prophet is not a particularly good model for forecasting, and has been superseded by a multitude of other models. If you want to do time series forecasting, I'd recommend using Darts: https://github.com/unit8co/darts. Darts implements a wide range of models and is fairly easy to use.
The problem with time series forecasting in general is that they make a lot of assumptions on the shape of your data, and you'll find you're spending a lot of time figuring out mutating your data. For example, they expect that your data comes at a very regular interval. This is fine if it's, say, the data from a weather station. This doesn't work well in clinical settings (imagine a patient admitted into the ER -- there is a burst of data, followed by no data).
That said, there's some interesting stuff out there that I've been experimenting with that seems to be more tolerant of irregular time series and can be quite useful. If you're interested in exchanging ideas, drop me a line (email in my profile).
-
Elevate Your Python Skills: Machine Learning Packages That Transformed My Journey as ML Engineer
3. darts
-
Aeon: A unified framework for machine learning with time series
Looking forward to checking this out! How does this compare with darts[1]?
[1] https://unit8co.github.io/darts/
- [D] Hybrid forecasting framework ARIMA-LSTM
-
[D] Do any of you have experience using Darts for forecasting?
Darts is an open-source Python library by Unit8 for easy handling, pre-processing, and forecasting of time series. It contains an array of models, from standard statistical models such as ARIMA to deep neural networks. https://unit8co.github.io/darts/
-
gluonts VS darts - a user suggested alternative
2 projects | 13 Apr 2023
active support
-
A Simple Guide to Feature Engineering in the Forecast Menu
The new Forecast menu, featuring the open-source Darts Time Series library, offers script-friendly functionality. It's also easy to use. Don't have any data to load yet? Enter through the Stocks or Crypto menus.
-
Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
I would recommend Darts in Python [1]. It's easy to use (think fit()/predict()) and includes
* Statistical models (ETS, (V)ARIMA(X), etc)
* ML models (sklearn models, LGBM, etc)
* Many recent deep learning models (N-BEATS, TFT, etc)
* Seamlessly works on multi-dimensional series
* Models can be trained on multiple series
* Many models offer rich support for probabilistic forecasts
* Model evaluation is easy: Darts has many metrics, offers backtest etc
* Deep learning scales to large datasets, using GPUs, TPUs, etc
* There's even now an explainability module for some of the models - showing you what matters for computing the forecasts
* (coming soon): an anomaly detection module :)
* (also, it even include FB Prophet if you really want to use it)
Warning: I'm probably biased because I'm Darts creator.
[1] https://github.com/unit8co/darts
What are some alternatives?
meta - A Meta Repository for General Monero Project Matters
sktime - A unified framework for machine learning with time series
pytorch-forecasting - Time series forecasting with PyTorch
Prophet - Tool for producing high quality forecasts for time series data that has multiple seasonality with linear or non-linear growth.
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 Python Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai
statsforecast - Lightning ⚡️ fast forecasting with statistical and econometric models.
nixtla - TimeGPT-1: production ready pre-trained Time Series Foundation Model for forecasting and anomaly detection. Generative pretrained transformer for time series trained on over 100B data points. It's capable of accurately predicting various domains such as retail, electricity, finance, and IoT with just a few lines of code 🚀.
neural_prophet - NeuralProphet: A simple forecasting package
neuralforecast - Scalable and user friendly neural :brain: forecasting algorithms.
greykite - A flexible, intuitive and fast forecasting library
flow-forecast - Deep learning PyTorch library for time series forecasting, classification, and anomaly detection (originally for flood forecasting).