georgian-io / pyoats Goto Github PK
View Code? Open in Web Editor NEWQuick and Easy Time Series Outlier Detection
License: Apache License 2.0
Quick and Easy Time Series Outlier Detection
License: Apache License 2.0
Thanks for creating this awesome library. I want to ask if this package is still maintained? Thanks
Hi, I want to ask how I could save the trained models. Thanks!
AttributeError Traceback (most recent call last)
Cell In[4], line 1
----> 1 from oats.models import*
2 TRAIN_SIZE = 5000
3 mp_model = MatrixProfileModel(window=48, use_gpu=True) # Try changing the model or window size
File ~\AppData\Roaming\Python\Python38\site-packages\oats\models_init_.py:8
1 """
2 Models
3 -----------------
4 """
6 import tensorflow as tf
----> 8 gpu_devices = tf.config.experimental.list_physical_devices("GPU")
9 for device in gpu_devices:
10 tf.config.experimental.set_memory_growth(device, True)
AttributeError: module 'tensorflow' has no attribute 'config'
Hi,
I am looking to implement the Reconstruction-Based Models with the Autoencoders but do not understand how to, especially as the documentation seems to be offline. Could someone guide me through doing so?
Hi I was wondering if you have 3.11 compatibility on the horizon, thanks for the nice package.
Best,
Derek
Hi,
I'm using pyoats library, but I cannot find a way in your doc to save the trained model after fitting so that I can decide the best hyperparameters, is there a way to save it?
I am interested in anomaly detection. Your work is great, but there is a lack of detailed examples. I have installed pyoats
, but I don't know what to do next.
What is the error:
Set use_gpu=True
raise unexpected keyword argument in pytorch lightning
Error logs:
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/oats/models/_darts_model.py:173, in DartsModel.fit(self, train_data, epochs, **kwargs)
170 train_data = self._scale_series(train_data)
171 tr, val = self._get_train_val_split(train_data, self.val_split)
--> 173 self.model.fit(
174 TimeSeries.from_values(tr),
175 val_series=TimeSeries.from_values(val),
176 epochs=epochs,
177 num_loader_workers=1,
178 **kwargs,
179 )
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/darts/utils/torch.py:112, in random_method.<locals>.decorator(self, *args, **kwargs)
110 with fork_rng():
111 manual_seed(self._random_instance.randint(0, high=MAX_TORCH_SEED_VALUE))
--> 112 return decorated(self, *args, **kwargs)
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/darts/models/forecasting/torch_forecasting_model.py:755, in TorchForecastingModel.fit(self, series, past_covariates, future_covariates, val_series, val_past_covariates, val_future_covariates, trainer, verbose, epochs, max_samples_per_ts, num_loader_workers)
747 logger.info(f"Train dataset contains {len(train_dataset)} samples.")
749 super().fit(
750 series=seq2series(series),
751 past_covariates=seq2series(past_covariates),
752 future_covariates=seq2series(future_covariates),
753 )
--> 755 return self.fit_from_dataset(
756 train_dataset, val_dataset, trainer, verbose, epochs, num_loader_workers
757 )
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/darts/utils/torch.py:112, in random_method.<locals>.decorator(self, *args, **kwargs)
110 with fork_rng():
111 manual_seed(self._random_instance.randint(0, high=MAX_TORCH_SEED_VALUE))
--> 112 return decorated(self, *args, **kwargs)
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/darts/models/forecasting/torch_forecasting_model.py:896, in TorchForecastingModel.fit_from_dataset(self, train_dataset, val_dataset, trainer, verbose, epochs, num_loader_workers)
893 train_num_epochs = epochs if epochs > 0 else self.n_epochs
895 # setup trainer
--> 896 self.trainer = self._setup_trainer(trainer, verbose, train_num_epochs)
898 # TODO: multiple training without loading from checkpoint is not trivial (I believe PyTorch-Lightning is still
899 # working on that, see https://github.com/PyTorchLightning/pytorch-lightning/issues/9636)
900 if self.epochs_trained > 0 and not self.load_ckpt_path:
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/darts/models/forecasting/torch_forecasting_model.py:471, in TorchForecastingModel._setup_trainer(self, trainer, verbose, epochs)
466 trainer_params["enable_model_summary"] = (
467 verbose if self.model.epochs_trained == 0 else False
468 )
469 trainer_params["enable_progress_bar"] = verbose
--> 471 return self._init_trainer(trainer_params=trainer_params, max_epochs=epochs)
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/darts/models/forecasting/torch_forecasting_model.py:482, in TorchForecastingModel._init_trainer(trainer_params, max_epochs)
479 if max_epochs is not None:
480 trainer_params_copy["max_epochs"] = max_epochs
--> 482 return pl.Trainer(**trainer_params_copy)
File /layers/dap-buildpacks_pip-install/site-packages/virtual-env/lib/python3.9/site-packages/pytorch_lightning/utilities/argparse.py:70, in _defaults_from_env_vars.<locals>.insert_env_defaults(self, *args, **kwargs)
67 kwargs = dict(list(env_variables.items()) + list(kwargs.items()))
69 # all args were already moved to kwargs
---> 70 return fn(self, **kwargs)
**TypeError: __init__() got an unexpected keyword argument 'gpus'**
How to replicate:
from oats.models import *
model = NBEATSModel(48, use_gpu=True)
I'm so interested in this project,but i find your community construction is incomplete, so could you provide some simple example for us?
such as a short code with short datasets. Expect your reply!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.