tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series classification, regression and forecasting.

timeseriesAI

Last update: Jan 8, 2023

Related tags

Deep Learning timeseries deep-learning time-series cnn pytorch transformer rnn sequential sequence fastai time-series-analysis time-series-classification self-supervised state-of-the-art tsai time-series-regression tutorial-nbs

Overview

tsai

State-of-the-art Deep Learning for Time Series and Sequence Modeling. tsai is currently under active development by timeseriesAI.

tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series classification, regression and forecasting.

Self-supervised learning: If you are interested in applying self-supervised learning to time series, you may check our new tutorial notebook: 08_Self_Supervised_TSBERT.ipynb
New visualization: We've also added a new PredictionDynamics callback that will display the predictions during training. This is the type of output you would get in a classification task for example:

Installation

You can install the latest stable version from pip using:

pip install tsai

Or you can install the cutting edge version of this library from github by doing:

pip install -Uqq git+https://github.com/timeseriesAI/tsai.git

Once the install is complete, you should restart your runtime and then run:

from tsai.all import *

Documentation

Here's the link to the documentation.

How to get started

To get to know the tsai package, we'd suggest you start with this notebook in Google Colab: 01_Intro_to_Time_Series_Classification

It provides an overview of a time series classification problem using fastai v2.

If you want more details, you can get them in nbs 00 and 00a.

To use tsai in your own notebooks, the only thing you need to do after you have installed the package is to add this:

from tsai.all import *

Citing tsai

If you use tsai in your research please use the following BibTeX entry:

@Misc{tsai,
    author =       {Ignacio Oguiza},
    title =        {tsai - A state-of-the-art deep learning library for time series and sequential data},
    howpublished = {Github},
    year =         {2020},
    url =          {https://github.com/timeseriesAI/tsai}
}

Comments

Training on Larger than Memory Datasets

Thanks for providing these implementations, they're very high quality. I've successfully reproduced most of the state of the art results in the self-supervised BERT-like time series transformer paper. All of the datasets used in the paper fit nicely into memory, but many real-world production applications will require iterating over datasets that are too large to fit into memory.

Dataset classes that support larger than memory datasets were proposed as a goal for FastAI 2.3 here. Linked in that discussion is NVTabular's Dataset class, an example of a dataloader that supports larger-than-memory datasets.

Supporting larger-than-memory datasets is probably outside the scope of this project, but I wonder if you could provide any pointers on how one might try to implement this in the existing TSDatasets and TSDataloaders/NumpyDataLoaders classes? I find they are pretty complex, and I'm not familiar with the fastai DataLoaders and Datasets classes that they're subclassing.

The NVTabular Dataset class randomly iterates over the partitions of a dask data frame, which supports multiple files on disk and larger-than-memory datasets. My understanding is that this could be challenging with fastai's training because the __getitem__ call requires random seeking of data, and dask does not support random seeks because it could require reading a different file on disk for each element. I wonder if the __getitem__ could just be tricked in some way to randomly iterate over a partition before moving on to the next random partition. NVTabular is using TorchAsyncIter to iterate over the lazily read dask dataframe.

opened by xanderdunn 48
Memory Error with Very Large Dataset
Dear team,

I am in stuck when convert very large numpy array to your TSDatasets. These are what I have tried to fix my issue:

when building time-series, I used tensorflow.keras.preprocessing.timeseries_dataset_from_array. After this step, the memory is still fine

I concatenate all batch data into numpy array, this step produces problem so I use numpy memmap to avoid load data into RAM --> now, this step is ok

I don't know where in your library loads all the data into RAM so it crashes in the line dsets = TSDatasets(X_dl, y_dl, splits=splits, tfms=transformations, inplace=True)

Is there any way to overcome this issue? Or I have to use your model only and design my own pipeline to use Tensorflow Data Generator
enhancement good first issue
opened by HariWu1995 34
Serializing Time Series Forecasts

I've been getting great results on forecasting a multistep horizon for a multivariate time series, but am having a lot of trouble exporting or saving the model to use on other machines or even in the same Jupyter Notebook.

I create the learner with ts_learner, train it, but when I use learner.save or learner.export, the imported model doesn't have the same predictions.

Any help would be appreciated.
under review

opened by DonRomaniello 29
RuntimeError: Can only calculate the mean of floating types. Got Long instead.

RuntimeError: Can only calculate the mean of floating types. Got Long instead. dls = TSDataLoaders.from_dsets(dsets.train, dsets.valid, bs=64, batch_tfms=TSStandardize(by_var=True))

if batch_tfms=TSStandardize(by_var=True) is removed RuntimeError: expected scalar type Long but found Float
under review

opened by maheshs11 20
add support for session windows

TSAI has great support for sliding windows in the data preprocessing methods: https://github.com/timeseriesAI/tsai/blob/62e9348d9e29a6b5f628879bd77056c11db5c0ab/tsai/data/preparation.py#L119

Ideally support for (event-) session based windows can be added as outlined in https://stackoverflow.com/questions/66783615/transform-time-series-to-event-window-sessions

opened by geoHeil 20

AttributeError: 'TSRobustScale' object has no attribute '_setup'

Sorry Ignacio; I had what I thought was a bug in my code, but looks like it's associated with the 0.2.23 tsai package; confirmed not present in 0.2.20. From what I have read, this sort of error is commonly associated with mutual imports (eg, two modules importing from each other).

from tsai.inference import load_learner

model_path = f'/gdrive/MyDrive/***/LVEFCLASS-16735.pkl' 
model      = load_learner(model_path)
print(type(model))
model.export(f'/gdrive/MyDrive/***/LVEFCLASS-16735_.pkl)

<class 'fastai.learner.Learner'>
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-11-a9c25badaf5d> in <module>()
      4 model      = load_learner(model_path)
      5 print(type(model))
----> 6 model.export(f'/gdrive/MyDrive/Semler/QF+/Sensor Data Files/Cached_Data/models/InceptionTime/LVEFCLASS-16735_.pkl')

12 frames
/usr/local/lib/python3.7/dist-packages/tsai/data/preprocessing.py in setups(self, dl)
    327 
    328     def setups(self, dl: DataLoader):
--> 329         if self._setup:
    330             if not self.use_single_batch:
    331                 o = dl.dataset.__getitem__([slice(None)])[0]

AttributeError: 'TSRobustScale' object has no attribute '_setup'

under review

opened by bob-mcrae 18

Tutorial notebook using Optuna

I have added a tutorial notebook that showcases how to use TSAI with Optuna for hyperparameter optimization. Would love to hear your feedbacks as this is only my first pull request on Github. Thank you for the efforts in building TSAI. I would love to contribute more as I learn.

opened by dnth 17
limit on time points ?

Hello, I'm playing the package with the provided datasets. With sample in shape of batchvartime_points, I noticed that when the time_points is too big, the learner will throw error about dimension match.

For example with dataset:

dsid='HouseholdPowerConsumption1' X, y, splits = get_regression_data(dsid, split_data=False) print(dsid, X.shape, y.shape) HouseholdPowerConsumption1 (1431, 5, 1440) (1431,)

It has 1440 time points.

When I run: tfms = [None, [TSRegression()]] batch_tfms = TSStandardize(by_sample=True, by_var=True) dls = get_ts_dls(x, y, splits=splits, tfms=tfms, batch_tfms=batch_tfms, bs=128) dls.one_batch() learn = ts_learner(dls, TSTPlus, metrics=[mae, rmse], cbs=ShowGraph()) learn.lr_find()

I'll get error (from learn.lr_find() ) as below: RuntimeError: The size of tensor a (512) must match the size of tensor b (1440) at non-singleton dimension 3

Please help !

Thanks!

opened by ymlea 17
fast preprocessing

I love your new SlidingWindowPanel, but as you can easily imagine for even a moderate size of devices the current implementation using python loops and numpy (=single threaded implementation) is rather slow.

What do you think about adding a num-jobs parameter similar to scikit-learn which would use the CPU to auto-parallelize (perhaps via dask)?

opened by geoHeil 17
Bug:I just can not get the same result after running get_pred func!!!!

I just can not get the same result after running get_pred func!!!! https://colab.research.google.com/github/timeseriesAI/tsai/blob/master/tutorial_nbs/01_Intro_to_Time_Series_Classification.ipynb I get the 0.9278 after running learn.get_preds first time , but i get 0.7111 then i run learn.get_preds again?????
bug

opened by yifeiSunny 16

Error creating multivariate dataloader

Hi, I was trying to follow the tutorial notebook on how to prepare data:

https://github.com/timeseriesAI/tsai/blob/master/tutorial_nbs/00c_Time_Series_data_preparation.ipynb

I opened this in Google Colab, and have tried with and without the stable flag. My versions in Colab from the top cell are: tsai : 0.2.13 fastai : 2.1.10 fastcore : 1.3.13 torch : 1.7.0+cu101

Under the End-End examples/Single multivariate time series, I can load the first cell fine and see the df. However when I run the second cell to create the data loader, I get the following error:

    ---------------------------------------------------------------------------
    AssertionError                            Traceback (most recent call last)
    <ipython-input-3-dbbb5a3104e2> in <module>()
          7 seq_first = True
          8 
    ----> 9 X, y = SlidingWindow(window_length, stride=stride, start=start, get_x=get_x,  get_y=get_y, horizon=horizon, seq_first=seq_first)(df)
         10 splits = get_splits(y, valid_size=.2, stratify=True, random_state=23, shuffle=False)
         11 tfms  = [None, [Categorize()]]
    
    /usr/local/lib/python3.6/dist-packages/tsai/data/preparation.py in SlidingWindow(window_len, stride, start, get_x, get_y, y_func, horizon, seq_first, sort_by, ascending, check_leakage)
         93     if min_horizon <= 0 and y_func is None and get_y != [] and check_leakage:
         94         assert get_x is not None and  get_y is not None and len([y for y in _get_y if y in _get_x]) == 0,  \
    ---> 95         'you need to change either horizon, get_x, get_y or use a y_func to avoid leakage'
         96     stride = ifnone(stride, window_len)
         97 
    
    AssertionError: you need to change either horizon, get_x, get_y or use a y_func to avoid leakage
    ---------------------------

Any suggestions?

opened by chandrashan 16

TypeError: init() got an unexpected keyword argument 'custom_head'

I am trying to use TSRegressor with the TST architecture. and I get a TypeError: __init__() got an unexpected keyword argument 'custom_head'

I tried the fix from #597, but it doesn't work, the issue too complex for my understanding. @oguiza can you please take a look at it?

Here is the entire stacktrace

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
[<ipython-input-33-8d14c640de6e>](https://localhost:8080/#) in <module>
----> 1 learn = TSRegressor(X, y, batch_tfms=batch_tfms, splits=split, arch=TST,
      2                     metrics=mae, bs=512, train_metrics=True)

3 frames
[/usr/local/lib/python3.8/dist-packages/tsai/tslearner.py](https://localhost:8080/#) in __init__(self, X, y, splits, tfms, inplace, sel_vars, sel_steps, weights, partial_n, train_metrics, bs, batch_size, batch_tfms, shuffle_train, drop_last, num_workers, do_setup, device, arch, arch_config, pretrained, weights_path, exclude_head, cut, init, loss_func, opt_func, lr, metrics, cbs, wd, wd_bn_bias, train_bn, moms, path, model_dir, splitter, verbose)
    133             #     model = build_ts_model(arch, dls=dls, device=device, verbose=verbose, pretrained=pretrained, weights_path=weights_path,
    134             #                        exclude_head=exclude_head, cut=cut, init=init, arch_config=arch_config)
--> 135             model = build_ts_model(arch, dls=dls, device=device, verbose=verbose, pretrained=pretrained, weights_path=weights_path,
    136                                 exclude_head=exclude_head, cut=cut, init=init, arch_config=arch_config)
    137         try:

[/usr/local/lib/python3.8/dist-packages/tsai/models/utils.py](https://localhost:8080/#) in build_ts_model(arch, c_in, c_out, seq_len, d, dls, device, verbose, pretrained, weights_path, exclude_head, cut, init, arch_config, **kwargs)
    158             if v in arch.__name__]):
    159         pv(f'arch: {arch.__name__}(c_in={c_in} c_out={c_out} seq_len={seq_len} device={device}, arch_config={arch_config}, kwargs={kwargs})', verbose)
--> 160         model = arch(c_in, c_out, seq_len=seq_len, **arch_config, **kwargs).to(device=device)
    161     elif 'xresnet' in arch.__name__ and not '1d' in arch.__name__:
    162         pv(f'arch: {arch.__name__}(c_in={c_in} c_out={c_out} device={device}, arch_config={arch_config}, kwargs={kwargs})', verbose)

[/usr/local/lib/python3.8/dist-packages/fastcore/meta.py](https://localhost:8080/#) in __call__(cls, *args, **kwargs)
     38         if type(res)==cls:
     39             if hasattr(res,'__pre_init__'): res.__pre_init__(*args,**kwargs)
---> 40             res.__init__(*args,**kwargs)
     41             if hasattr(res,'__post_init__'): res.__post_init__(*args,**kwargs)
     42         return res

[/usr/local/lib/python3.8/dist-packages/tsai/models/TST.py](https://localhost:8080/#) in __init__(self, c_in, c_out, seq_len, max_seq_len, n_layers, d_model, n_heads, d_k, d_v, d_ff, dropout, act, fc_dropout, y_range, verbose, **kwargs)
    172             self.new_q_len = True
    173             t = torch.rand(1, 1, seq_len)
--> 174             q_len = nn.Conv1d(1, 1, **kwargs)(t).shape[-1]
    175             self.W_P = nn.Conv1d(c_in, d_model, **kwargs) # Eq 2
    176             pv(f'Conv1d with kwargs={kwargs} applied to input to create input encodings\n', verbose)

TypeError: __init__() got an unexpected keyword argument 'custom_head'

opened by deven-gqc 0

'LSTMPlus' object has no attribute '__name__'

Hi,I am trying to train a LSTM model and when I adjust params,it will show that" 'LSTMPlus' object has no attribute 'name' ". Do you know how to fix this? @oguiza Sorry to bother and Thanks.

opened by Sure20220604 1

TSDatasets does not work with torch.Tensor

Hi TSAI contributors,

I would like to create my own datasets with torch.Tensor data:

for iter, (data, label) in enumerate(dataloader):
    if iter == 0:
        X = data
        y = label
    else:
        X = torch.concat([X, data], dim=0)
        y = torch.concat([y, label])

splits = get_splits(y, valid_size=.2, stratify=True, random_state=23, shuffle=True)

and after check_data(X, y, splits), I got

X      - shape: [1472 samples x 3 features x 480 timesteps]  type: Tensor  dtype:torch.float64  isnan: 0
y      - shape: torch.Size([1472])  type: Tensor  dtype:torch.int64  isnan: 0
splits - n_splits: 2 shape: [1178, 294]  overlap: False

Then I run

tfms = [None, [Categorize()]]
dsets = TSDatasets(X, y, tfms=tfms, splits=splits, inplace=True)

and got

KeyError                                  Traceback (most recent call last)
File ~/miniconda3/lib/python3.8/site-packages/fastai/data/transforms.py:261, in Categorize.encodes(self, o)
    260 try:
--> 261     return TensorCategory(self.vocab.o2i[o])
    262 except KeyError as e:

KeyError: tensor(0)

The above exception was the direct cause of the following exception:

KeyError                                  Traceback (most recent call last)
/Users/xiaochen/Project/icudelir_benchmarks/nbs/Baseline_Methods/Baseline_Data_Preparation.ipynb Cell 28 in <cell line: 2>()
      1 tfms = [None, [Categorize()]]
----> 2 dsets = TSDatasets(X, y, tfms=tfms, splits=splits, inplace=True)

File ~/miniconda3/lib/python3.8/site-packages/tsai/data/core.py:422, in TSDatasets.__init__(self, X, y, items, sel_vars, sel_steps, tfms, tls, n_inp, dl_type, inplace, **kwargs)
    420     self.tls = L(lt(item, t, **kwargs) for lt,item,t in zip(lts, items, self.tfms))
    421     if len(self.tls) > 0 and len(self.tls[0]) > 0:
--> 422         self.typs = [type(tl[0]) if isinstance(tl[0], torch.Tensor) else self.typs[i] for i,tl in enumerate(self.tls)]
    423     self.ptls = L([typ(stack(tl[:]))[...,self.sel_vars, self.sel_steps] if (i==0 and self.multi_index) else typ(stack(tl[:])) \
    424                     for i,(tl,typ) in enumerate(zip(self.tls,self.typs))]) if inplace else self.tls
    425 else:

File ~/miniconda3/lib/python3.8/site-packages/tsai/data/core.py:422, in <listcomp>(.0)
...
    261     return TensorCategory(self.vocab.o2i[o])
    262 except KeyError as e:
--> 263     raise KeyError(f"Label '{o}' was not included in the training dataset") from e

KeyError: "Label '0' was not included in the training dataset"

My y looks like tensor([0, 0, 0, ..., 1, 1, 1]).

If I add

X = X.numpy().astype(np.float64)
y = y.numpy().astype(np.int64)

everything works!

Thanks a lot!

opened by xcvil 0

Feature Importance & Step Importance Not working

Hi here is my system versioning,

python          : 3.7.12
tsai            : 0.3.4
fastai          : 2.7.10
fastcore        : 1.5.27
torch           : 1.13.1+cu117
device          : 1 gpu (['Tesla P100-PCIE-16GB'])
cpu cores       : 28
threads per cpu : 1
RAM             : 251.23 GB
GPU memory      : [16.0] GB
os              : Linux-3.10.0-1160.76.1.el7.x86_64-x86_64-with-redhat-7.9-Maipo
python          : 3.7.12
tsai            : 0.3.4
fastai          : 2.7.10
fastcore        : 1.5.27
torch           : 1.13.1+cu117
device          : 1 gpu (['Tesla P100-PCIE-16GB'])
cpu cores       : 28
threads per cpu : 1
RAM             : 251.23 GB
GPU memory      : [16.0] GB

when I am running the following lines of code:

    X = np.random.randn(300, 2, 50)
    Y = np.zeros(300)
    Y[200:] = 1
    X_train , X_test, y_train, y_test = train_test_split(X, Y,  test_size = 0.2, random_state = 333)
    X, y, splits = combine_split_data([X_train, X_test], [y_train, y_test])
    tfms = [None, TSClassification()]
    dls = get_ts_dls(X, y, splits=splits, tfms=tfms) 
    learn = ts_learner(dls, XceptionTimePlus, 
            loss_func = LabelSmoothingCrossEntropyFlat(),
            # metrics = accuracy,)
            metrics= RocAucBinary())
    learn.fit_one_cycle(2, 1e-4)
    learn.step_importance()

it throws an error:

Traceback (most recent call last):                                                                                                      
  File "/gpfs/ysm/project/gerstein/yl2428/conda_envs/ABCD/lib/python3.7/site-packages/tsai/analysis.py", line 367, in step_importance
    try: value = metric(output[0][:, 1], output[1]).item()
  File "/gpfs/ysm/project/gerstein/yl2428/conda_envs/ABCD/lib/python3.7/site-packages/sklearn/metrics/_ranking.py", line 580, in roc_auc_score
    sample_weight=sample_weight,
  File "/gpfs/ysm/project/gerstein/yl2428/conda_envs/ABCD/lib/python3.7/site-packages/sklearn/metrics/_base.py", line 72, in _average_binary_score
    raise ValueError("{0} format is not supported".format(y_type))
ValueError: continuous format is not supported

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/gpfs/ysm/home/yl2428/YL_ABCD/ABCD/ABCD/train/trainer.py", line 177, in <module>
    learn.step_importance()
  File "/gpfs/ysm/project/gerstein/yl2428/conda_envs/ABCD/lib/python3.7/site-packages/tsai/analysis.py", line 368, in step_importance
    except: value = metric(output[0], output[1]).item()
  File "/gpfs/ysm/project/gerstein/yl2428/conda_envs/ABCD/lib/python3.7/site-packages/sklearn/metrics/_ranking.py", line 580, in roc_auc_score
    sample_weight=sample_weight,
  File "/gpfs/ysm/project/gerstein/yl2428/conda_envs/ABCD/lib/python3.7/site-packages/sklearn/metrics/_base.py", line 72, in _average_binary_score
    raise ValueError("{0} format is not supported".format(y_type))
ValueError: continuous-multioutput format is not supported

opened by liyy2 0

Releases(0.3.4)

0.3.4(Nov 18, 2022)
New Features

compatibility with Pytorch 1.13 (#619)

added sel_vars to get_robustscale_params (#610)

added sel_steps to TSRandom2Value (#607)

new walk forward cross-validation in tsai (#582)

Bugs Squashed

fixed issue when printing an empty dataset wo transforms NoTfmLists (#622)

fixed minor issue in get_robustscaler params with sel_vars (#615)

fixed issue when using tsai in dev with VSCode (#614)

issue when using lists as sel_vars and sel_steps in TSRandom2Value (#612)

fixed issue with feature_importance and step_importance when using metrics (#609)

renamed data processing tfms feature_idxs as sel_vars for consistency (#608)

fixed issue when importing 'GatedTabTransformer' (#536)

Source code(tar.gz)
Source code(zip)
0.3.2(Oct 9, 2022)
Breaking Changes

replaced TSOneHot preprocessor by TSOneHotEncode using a different API (#502)

replaced MultiEmbedding n_embeds, embed_dims and padding_idxs by n_cat_embeds, cat_embed_dims and cat_padding_idxs (#497)

New Features

added GaussianNoise transform (#514)

added TSSequencer model based on Sequencer: Deep LSTM for Image Classification paper (#508)

added TSPosition to be able to pass any steps list that will be concatenated to the input (#504)

added TSPosition preprocessor to allow the concatenation of a custom position sequence (#503)

added TSOneHot class to encode a variable on the fly (#501)

added token_size and tokenizer arguments to tsai (#496)

SmeLU activation function not found (#495)

added example on how to perform inference, partial fit and fine tuning (#491)

added get_time_per_batch and get_dl_percent_per_epoch (#489)

added TSDropVars used to removed batch variables no longer needed (#488)

added SmeLU activation function (#458)

Feature request: gMLP and GatedTabTransformer. (#354)

Pay Attention to MLPs - gMLP (paper, implementation)

The GatedTabTransformer (paper, implementation);

Bugs Squashed

after_batch tfms set to empty Pipeline when using dl.new() (#516)

00b_How_to_use_numpy_arrays_in_fastai: AttributeError: attribute 'device' of 'torch._C._TensorBase' objects is not writable (#500)

getting regression data returns _check_X() argument error (#430)

I wonder why only 'Nor' is displayed in dls.show_batch(sharvey=True). (#416)

Source code(tar.gz)
Source code(zip)
0.3.1(Apr 19, 2022)
Release notes

0.3.1

New Features

added StratifiedSampler to handle imbalanced datasets (#479)

added seq_embed_size and seq_embed arguments to TSiT (#476)

added get_idxs_to_keep that can be used to filter indices based on different conditions (#469)

added SmeLU activation function (#458)

added split_in_chunks (#454)

upgraded min Python version to 3.7 (#450)

added sampler argument to NumpyDataLoader and TSDataLoader (#436)

added TSMask2Value transform which supports multiple masks (#431)

added TSGaussianStandardize for improved ood generalization (#428)

added get_dir_size function (#421)

Bugs Squashed

slow import of MiniRocketMultivariate from sktime (#482)

Fixed install from source fails on Windows (UnicodeDecodeError) (#470)

TSDataset error oindex is not an attribute (#462)

split_in_chunks incorrectly calculated (#455)

_check_X() got an unexpected keyword argument 'coerce_to_numpy' (#415)

Source code(tar.gz)
Source code(zip)
0.3.0(Mar 2, 2022)
Release notes

0.3.0

New Features

Added function that pads sequences to same length (#410)

Added TSRandomStandardize preprocessing technique (#396)

New visualization techniques: model's feature importance and step importance (#393)

Allow from tsai.basics import * to speed up loading (#320)

Bugs Squashed

Separate core from non-core dependencies in tsai - pip install tsaiextras. This is an important change that:

reduces the time to pip install tsai

avoid errors during installation

reduces the time to load tsai using from tsai.all import *

Source code(tar.gz)
Source code(zip)
0.2.25(Feb 6, 2022)
0.2.25

Breaking Changes

updated forward_gaps removing nan_to_num (#331)

TSRobustScaler only applied by_var (#329)

remove add_na arg from TSCategorize (#327)

New Features

added IntraClassCutMix1d (#384)

added learn.calibrate_model method (#379)

added analyze_array function (#378)

Added TSAddNan transform (#376)

added dummify function to create dummy data from original data (#366)

added Locality Self Attention to TSiT (#363)

added sel_vars argument to MVP callback (#349)

added sel_vars argument to TSNan2Value (#348)

added multiclass, weighted FocalLoss (#346)

added TSRollingMean batch transform (#343)

added recall_at_specificity metric (#342)

added train_metrics argument to ts_learner (#341)

added hist to PredictionDynamics for binary classification (#339)

add padding_idxs to MultiEmbedding (#330)

Bugs Squashed

sort_by data may be duplicated in SlidingWindowPanel (#389)

create_script splits the nb name if multiple underscores are used (#385)

added torch functional dependency to plot_calibration_curve (#383)

issue when setting horizon to 0 in SlidingWindow (#382)

replace learn by self in calibrate_model patch (#381)

Argument d_head is not used in TSiTPlus (#380)

https://github.com/timeseriesAI/tsai/blob/6baf0ba2455895b57b54bf06744633b81cdcb2b3/tsai/models/TSiTPlus.py#L63

replace default relu activation by gelu in TSiT (#361)

sel_vars and sel_steps in TSDatasets and TSDalaloaders don't work when used simultaneously (#347)

ShowGraph fails when recoder.train_metrics=True (#340)

fixed 'se' always equal to 16 in MLSTM_FCN (#337)

ShowGraph doesn't work well when train_metrics=True (#336)

TSPositionGaps doesn't work on cuda (#333)

XResNet object has no attribute 'backbone' (#332)

import InceptionTimePlus in tsai.learner (#328)

df2Xy: Format correctly without the need to specify sort_by (#324)

bug in MVP code learn.model --> self.learn.model (#323)

Colab install issues: importing the lib takes forever (#315)

Calling learner.feature_importance on larger than memory dataset causes OOM (#310)

Source code(tar.gz)
Source code(zip)
0.2.24(Dec 16, 2021)
Release notes

0.2.24

Breaking Changes

removed InceptionTSiT, InceptionTSiTPlus, ConvTSiT & ConvTSiTPlus (#276)

New Features

add stateful custom sklearn API type tfms: TSShrinkDataFrame, TSOneHotEncoder, TSCategoricalEncoder (#313)

Pytorch 1.10 compatibility (#311)

ability to pad at the start/ end of sequences and filter results in SlidingWindow (#307)

added bias_init to TSiT (#288)

plot permutation feature importance after a model's been trained (#286)

added separable as an option to MultiConv1d (#285)

Modified TSiTPlus to accept a feature extractor and/or categorical variables (#278)

Bugs Squashed

learn modules takes too long to load (#312)

error in roll2d and roll3d when passing index 2 (#304)

TypeError: unhashable type: 'numpy.ndarray' (#302)

ValueError: only one element tensors can be converted to Python scalars (#300)

unhashable type: 'numpy.ndarray' when using multiclass multistep labels (#298)

incorrect data types in NumpyDatasets subset (#297)

create_future_mask creates a mask in the past (#293)

NameError: name 'X' is not defined in learner.feature_importance (#291)

TSiT test fails on cuda (#287)

MultiConv1d breaks when ni == nf (#284)

WeightedPerSampleLoss reported an error when used with LDS_weights (#281)

pos_encoding transfer weight in TSiT fails (#280)

MultiEmbedding cat_pos and cont_pos are not in state_dict() (#277)

fixed issue with MixedDataLoader (#229), thanks to @Wabinab

Source code(tar.gz)
Source code(zip)
0.2.23(Nov 25, 2021)
Release notes

0.2.23

Breaking Changes

removed torch-optimizer dependency (#228)

New Features

added option to train MVP on random sequence lengths (#252)

added ability to pass an arch name (str) to learner instead of class (#217)

created convenience fns create_directory and delete_directory in utils (#213)

added option to create random array of given shapes and dtypes (#212)

my_setup() print your main system and package versions (#202)

added a new tutorial on how to train large datasets using tsai (#199)

added a new function to load any file as a module (#196)

Created CODE_OF_CONDUCT.md in https://github.com/timeseriesAI/tsai/pull/210

Add Optuna tutorial notebook by @dnth in https://github.com/timeseriesAI/tsai/pull/275

Bugs Squashed

Loading code just for inference takes too long (#273)

Fixed out-of-memory issue with large datasets on disk (#126)

AttributeError: module 'torch' has no attribute 'nan_to_num' (#262)

Fixed TypeError: unhashable type: 'numpy.ndarray' (#250)

Wrong link in paper references (#249)

remove default PATH which overwrites custom PATH (#238)

Predictions where not properly decoded when using with_decoded. (#237)

SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame (#221)

InceptionTimePlus wasn't imported by TSLearners (#218)

get_subset_dl fn is not properly creating a subset dataloader (#211)

Bug in WeightedPersSampleLoss (#203)

Bump nokogiri from 1.11.4 to 1.12.5 in /docs by @dependabot in https://github.com/timeseriesAI/tsai/pull/222

New Contributors

@geoHeil made their first contribution in https://github.com/timeseriesAI/tsai/pull/207

@dnth made their first contribution in https://github.com/timeseriesAI/tsai/pull/275

Full Changelog: https://github.com/timeseriesAI/tsai/compare/0.2.20...0.2.23
Source code(tar.gz)
Source code(zip)
0.2.20(Sep 9, 2021)

Source code(tar.gz)
Source code(zip)
0.2.19(Sep 6, 2021)

Source code(tar.gz)
Source code(zip)

tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series classification, regression and forecasting.

Related tags

Overview

tsai

Installation

Documentation

How to get started

Citing tsai

Comments

Releases(0.3.4)

0.3.4(Nov 18, 2022)

New Features

Bugs Squashed

0.3.2(Oct 9, 2022)

Breaking Changes

New Features

Bugs Squashed

0.3.1(Apr 19, 2022)

Release notes

0.3.1

New Features

Bugs Squashed

0.3.0(Mar 2, 2022)

Release notes

0.3.0

New Features

Bugs Squashed

0.2.25(Feb 6, 2022)

0.2.25

Breaking Changes

New Features

Bugs Squashed

0.2.24(Dec 16, 2021)

Release notes

0.2.24

Breaking Changes

New Features

Bugs Squashed

0.2.23(Nov 25, 2021)

Release notes

0.2.23

Breaking Changes

New Features

Bugs Squashed

New Contributors

0.2.20(Sep 9, 2021)

0.2.19(Sep 6, 2021)

Owner

timeseriesAI

Multivariate Time Series Forecasting with efficient Transformers. Code for the paper "Long-Range Transformers for Dynamic Spatiotemporal Forecasting."

Implementation of ETSformer, state of the art time-series Transformer, in Pytorch

A resource for learning about deep learning techniques from regression to LSTM and Reinforcement Learning using financial data and the fitness functions of algorithmic trading

deep-table implements various state-of-the-art deep learning and self-supervised learning algorithms for tabular data using PyTorch.

Fuzzification helps developers protect the released, binary-only software from attackers who are capable of applying state-of-the-art fuzzing techniques

A selection of State Of The Art research papers (and code) on human locomotion (pose + trajectory) prediction (forecasting)

LaneDet is an open source lane detection toolbox based on PyTorch that aims to pull together a wide variety of state-of-the-art lane detection models

The source code and data of the paper "Instance-wise Graph-based Framework for Multivariate Time Series Forecasting".

Library for implementing reservoir computing models (echo state networks) for multivariate time series classification and clustering.

This is the unofficial code of Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes. which achieve state-of-the-art trade-off between accuracy and speed on cityscapes and camvid, without using inference acceleration and extra data

Time Series Forecasting with Temporal Fusion Transformer in Pytorch

Image Classification - A research on image classification and auto insurance claim prediction, a systematic experiments on modeling techniques and approaches

😇A pyTorch implementation of the DeepMoji model: state-of-the-art deep learning model for analyzing sentiment, emotion, sarcasm etc

We evaluate our method on different datasets (including ShapeNet, CUB-200-2011, and Pascal3D+) and achieve state-of-the-art results, outperforming all the other supervised and unsupervised methods and 3D representations, all in terms of performance, accuracy, and training time.

The GitHub repository for the paper: “Time Series is a Special Sequence: Forecasting with Sample Convolution and Interaction“.

Quickly comparing your image classification models with the state-of-the-art models (such as DenseNet, ResNet, ...)

Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting

Implementation of the paper NAST: Non-Autoregressive Spatial-Temporal Transformer for Time Series Forecasting.

Spectral Temporal Graph Neural Network (StemGNN in short) for Multivariate Time-series Forecasting