https://github.com/adityalab/samay
State-of-art Foundational Time-series models
https://github.com/adityalab/samay
deep-learning foundation-models machine-learning neural-networks pytorch time-series time-series-analysis
Last synced: 28 days ago
JSON representation
State-of-art Foundational Time-series models
- Host: GitHub
- URL: https://github.com/adityalab/samay
- Owner: AdityaLab
- License: apache-2.0
- Created: 2024-10-17T19:38:43.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-03-31T15:26:54.000Z (about 1 month ago)
- Last Synced: 2025-03-31T16:56:39.691Z (about 1 month ago)
- Topics: deep-learning, foundation-models, machine-learning, neural-networks, pytorch, time-series, time-series-analysis
- Language: TypeScript
- Homepage:
- Size: 20 MB
- Stars: 16
- Watchers: 1
- Forks: 5
- Open Issues: 7
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Samay: Time-series Foundational Models Library
Package for training and evaluating time-series foundational models.
Current repository contains the following models:
1. [LPTM](https://arxiv.org/abs/2311.11413)
2. [MOMENT](https://arxiv.org/abs/2402.03885)
3. [TimesFM](https://arxiv.org/html/2310.10688v2)
4. [Chronos](https://arxiv.org/abs/2403.07815)
5. [MOIRAI](https://arxiv.org/abs/2402.02592)
6. [TinytTimeMixers](https://arxiv.org/abs/2401.03955)
More models will be added soon...
## Installation
You can add the package to your project by running the following command:
```bash
pip install git+https://github.com/AdityaLab/Samay.git
```
**Note:** If the installation fails because rust is missing run:
For MacOS:
```bash
brew install rustup
rustup-init
source ~/.cargo/env
```
For Linux:
```bash
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
source ~/.cargo/env
```
### Development workflow
To develop on the project, you can clone the repository and install the package in editable mode:
```bash
## Clone repo
git clone https://github.com/AdityaLab/Samay.git
## Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
## Install dependencies
uv sync --reinstall
```
## Usage Examples
Check out example notebooks at `examples/` for more detailed examples. We also have google colab notebooks at `examples/colab/`.
### LPTM
#### Loading Model
```python
from samay.model import LPTMModel
config = {
"task_name": "forecasting",
"forecast_horizon": 192,
"freeze_encoder": True, # Freeze the patch embedding layer
"freeze_embedder": True, # Freeze the transformer encoder
"freeze_head": False, # The linear forecasting head must be trained
}
model = LPTMModel(config)
```
#### Loading Dataset
```python
from samay.dataset import LPTMDataset
train_dataset = LPTMDataset(
name="ett",
datetime_col="date",
path="./data/data/ETTh1.csv",
mode="train",
horizon=192,
)
finetuned_model = model.finetune(train_dataset)
```
#### Zero-Forecasting
```python
avg_loss, trues, preds, histories = lptm.evaluate(val_dataset)
```
### TimesFM
#### Loading Model
```python
from samay.model import TimesfmModel
from samay.dataset import TimesfmDataset
repo = "google/timesfm-1.0-200m-pytorch"
config = {
"context_len": 512,
"horizon_len": 192,
"backend": "gpu",
"per_core_batch_size": 32,
"input_patch_len": 32,
"output_patch_len": 128,
"num_layers": 20,
"model_dims": 1280,
"quantiles": [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
}
tfm = TimesfmModel(config=config, repo=repo)
```
#### Loading Dataset
```python
train_dataset = TimesfmDataset(name="ett", datetime_col='date', path='data/ETTh1.csv',
mode='train', context_len=config["context_len"], horizon_len=128)
val_dataset = TimesfmDataset(name="ett", datetime_col='date', path='data/ETTh1.csv',
mode='test', context_len=config["context_len"], horizon_len=config["horizon_len"])
```
#### Zero-Forecasting
```python
avg_loss, trues, preds, histories = tfm.evaluate(val_dataset)
```
### MOIRAI
#### Loading Model
```python
from samay.dataset import MoiraiDataset
from samay.model import MoiraiTSModel
repo = "Salesforce/moirai-moe-1.0-R-small"
config = {
"context_len": 128,
"horizon_len": 64,
"num_layers": 100,
"model_type": "moirai-moe",
"model_size": "small"
}
moirai_model = MoiraiTSModel(repo=repo, config=config)
```
#### Loading Dataset
```python
train_dataset = MoiraiDataset(name="ett", mode="train", path="data/ETTh1.csv", datetime_col="date", freq="h",
context_len=config['context_len'], horizon_len=config['horizon_len'])
test_dataset = MoiraiDataset(name="ett", mode="test", path="data/ETTh1.csv", datetime_col="date", freq="h",
context_len=config['context_len'], horizon_len=config['horizon_len'])
```
#### Zero-Forecasting
```python
eval_results, trues, preds, histories = moirai_model.evaluate(test_dataset, metrics=["MSE", "MASE"])
```
### Support
Tested on Python 3.11-3.13 on Linux (CPU + GPU) and MacOS (CPU). Supports NVIDIA GPUs.
Support for Windows and Apple Silicon GPUs is planned.
## Citation
If you use this code in your research, please cite the following paper:
```bibtex
@inproceedings{
kamarthi2024large,
title={Large Pre-trained time series models for cross-domain Time series analysis tasks},
author={Harshavardhan Kamarthi and B. Aditya Prakash},
booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems},
year={2024},
url={https://openreview.net/forum?id=vMMzjCr5Zj}
}
```
## Contact
If you have any feedback or questions, you can contact us via email: , .