Deep learning examples

3dacdf9e7b724bf89544c85afe044809

This notebooks contains examples with neural network models.

Table of Contents

[1]:

import torch
import random

import pandas as pd
import numpy as np

from etna.datasets.tsdataset import TSDataset
from etna.pipeline import Pipeline
from etna.transforms import DateFlagsTransform
from etna.transforms import LagTransform
from etna.transforms import LinearTrendTransform
from etna.metrics import SMAPE, MAPE, MAE
from etna.analysis import plot_backtest
from etna.models import SeasonalMovingAverageModel

import warnings


def set_seed(seed: int = 42):
    """Set random seed for reproducibility."""
    random.seed(seed)
    np.random.seed(seed)
    torch.manual_seed(seed)
    torch.cuda.manual_seed_all(seed)


warnings.filterwarnings("ignore")

1. Creating TSDataset

We are going to take some toy dataset. Let’s load and look at it.

[2]:

original_df = pd.read_csv("data/example_dataset.csv")
original_df.head()

[2]:
timestamp segment target
0 2019-01-01 segment_a 170
1 2019-01-02 segment_a 243
2 2019-01-03 segment_a 267
3 2019-01-04 segment_a 287
4 2019-01-05 segment_a 279

Our library works with the special data structure TSDataset. Let’s create it as it was done in “Get started” notebook.

[3]:

df = TSDataset.to_dataset(original_df)
ts = TSDataset(df, freq="D")
ts.head(5)

[3]:
segment segment_a segment_b segment_c segment_d
feature target target target target
timestamp
2019-01-01 170 102 92 238
2019-01-02 243 123 107 358
2019-01-03 267 130 103 366
2019-01-04 287 138 103 385
2019-01-05 279 137 104 384

2. Architecture

Our library uses PyTorch Forecasting to work with time series neural networks. There are two ways to use pytorch-forecasting models: default one and via using PytorchForecastingDatasetBuilder for using extra features.

To include extra features we use PytorchForecastingDatasetBuilder class.

Let’s look at it closer.

[4]:

from etna.models.nn.utils import PytorchForecastingDatasetBuilder

[5]:

?PytorchForecastingDatasetBuilder

We can see a pretty scary signature, but don’t panic, we will look at the most important parameters.

  • time_varying_known_reals — known real values that change across the time (real regressors), now it it necessary to add “time_idx” variable to the list;

  • time_varying_unknown_reals — our real value target, set it to ["target"];

  • max_prediction_length — our horizon for forecasting;

  • max_encoder_length — length of past context to use;

  • static_categoricals — static categorical values, for example, if we use multiple segments it can be some its characteristics including identifier: “segment”;

  • time_varying_known_categoricals — known categorical values that change across the time (categorical regressors);

  • target_normalizer — class for normalization targets across different segments.

Our library currently supports these models: * DeepAR, * TFT.

3. Testing models

In this section we will test our models on example.

3.1 DeepAR

Before training let’s fix seeds for reproducibility.

[6]:

set_seed()

Default way

[7]:

from etna.models.nn import DeepARModel

HORIZON = 7


model_deepar = DeepARModel(
    encoder_length=HORIZON,
    decoder_length=HORIZON,
    trainer_params=dict(max_epochs=150, gpus=0, gradient_clip_val=0.1),
    lr=0.01,
    train_batch_size=64,
)
metrics = [SMAPE(), MAPE(), MAE()]

pipeline_deepar = Pipeline(model=model_deepar, horizon=HORIZON)

[8]:

metrics_deepar, forecast_deepar, fold_info_deepar = pipeline_deepar.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name                   | Type                   | Params
------------------------------------------------------------------
0 | loss                   | NormalDistributionLoss | 0
1 | logging_metrics        | ModuleList             | 0
2 | embeddings             | MultiEmbedding         | 0
3 | rnn                    | LSTM                   | 1.6 K
4 | distribution_projector | Linear                 | 22
------------------------------------------------------------------
1.6 K     Trainable params
0         Non-trainable params
1.6 K     Total params
0.006     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=150` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  2.6min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name                   | Type                   | Params
------------------------------------------------------------------
0 | loss                   | NormalDistributionLoss | 0
1 | logging_metrics        | ModuleList             | 0
2 | embeddings             | MultiEmbedding         | 0
3 | rnn                    | LSTM                   | 1.6 K
4 | distribution_projector | Linear                 | 22
------------------------------------------------------------------
1.6 K     Trainable params
0         Non-trainable params
1.6 K     Total params
0.006     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=150` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:  5.2min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name                   | Type                   | Params
------------------------------------------------------------------
0 | loss                   | NormalDistributionLoss | 0
1 | logging_metrics        | ModuleList             | 0
2 | embeddings             | MultiEmbedding         | 0
3 | rnn                    | LSTM                   | 1.6 K
4 | distribution_projector | Linear                 | 22
------------------------------------------------------------------
1.6 K     Trainable params
0         Non-trainable params
1.6 K     Total params
0.006     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=150` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  7.8min remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  7.8min finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    2.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    4.2s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    6.4s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    6.4s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[9]:

metrics_deepar

[9]:
segment SMAPE MAPE MAE fold_number
1 segment_a 11.458003 10.746780 58.296696 0
1 segment_a 3.176028 3.188012 16.515891 1
1 segment_a 7.292190 7.075453 38.055756 2
2 segment_b 8.014092 7.647167 20.117558 0
2 segment_b 4.404586 4.387303 10.633235 1
2 segment_b 5.720607 6.126164 13.026143 2
0 segment_c 6.136990 6.092246 10.315199 0
0 segment_c 4.311422 4.218119 7.638397 1
0 segment_c 9.405851 9.125047 16.484022 2
3 segment_d 5.807083 5.653968 50.962106 0
3 segment_d 4.531924 4.636149 36.712524 1
3 segment_d 3.950340 3.901889 31.104309 2

To summarize it we will take mean value of SMAPE metric because it is scale tolerant.

[10]:

score = metrics_deepar["SMAPE"].mean()
print(f"Average SMAPE for DeepAR: {score:.3f}")

Average SMAPE for DeepAR: 6.184

Dataset Builder: creating dataset for DeepAR with etxtra features.

[11]:

from pytorch_forecasting.data import GroupNormalizer

set_seed()

HORIZON = 7

transform_date = DateFlagsTransform(day_number_in_week=True, day_number_in_month=False, out_column="dateflag")
num_lags = 10
transform_lag = LagTransform(
    in_column="target",
    lags=[HORIZON + i for i in range(num_lags)],
    out_column="target_lag",
)
lag_columns = [f"target_lag_{HORIZON+i}" for i in range(num_lags)]

dataset_builder_deepar = PytorchForecastingDatasetBuilder(
    max_encoder_length=HORIZON,
    max_prediction_length=HORIZON,
    time_varying_known_reals=["time_idx"] + lag_columns,
    time_varying_unknown_reals=["target"],
    time_varying_known_categoricals=["dateflag_day_number_in_week"],
    target_normalizer=GroupNormalizer(groups=["segment"]),
)

Now we are going to start backtest.

[12]:

from etna.models.nn import DeepARModel


model_deepar = DeepARModel(
    dataset_builder=dataset_builder_deepar,
    trainer_params=dict(max_epochs=150, gpus=0, gradient_clip_val=0.1),
    lr=0.01,
    train_batch_size=64,
)
metrics = [SMAPE(), MAPE(), MAE()]

pipeline_deepar = Pipeline(
    model=model_deepar,
    horizon=HORIZON,
    transforms=[transform_lag, transform_date],
)

[13]:

metrics_deepar, forecast_deepar, fold_info_deepar = pipeline_deepar.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name                   | Type                   | Params
------------------------------------------------------------------
0 | loss                   | NormalDistributionLoss | 0
1 | logging_metrics        | ModuleList             | 0
2 | embeddings             | MultiEmbedding         | 35
3 | rnn                    | LSTM                   | 2.2 K
4 | distribution_projector | Linear                 | 22
------------------------------------------------------------------
2.3 K     Trainable params
0         Non-trainable params
2.3 K     Total params
0.009     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=150` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  2.3min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name                   | Type                   | Params
------------------------------------------------------------------
0 | loss                   | NormalDistributionLoss | 0
1 | logging_metrics        | ModuleList             | 0
2 | embeddings             | MultiEmbedding         | 35
3 | rnn                    | LSTM                   | 2.2 K
4 | distribution_projector | Linear                 | 22
------------------------------------------------------------------
2.3 K     Trainable params
0         Non-trainable params
2.3 K     Total params
0.009     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=150` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:  4.8min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name                   | Type                   | Params
------------------------------------------------------------------
0 | loss                   | NormalDistributionLoss | 0
1 | logging_metrics        | ModuleList             | 0
2 | embeddings             | MultiEmbedding         | 35
3 | rnn                    | LSTM                   | 2.2 K
4 | distribution_projector | Linear                 | 22
------------------------------------------------------------------
2.3 K     Trainable params
0         Non-trainable params
2.3 K     Total params
0.009     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=150` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  7.1min remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  7.1min finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    1.9s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    3.9s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    5.9s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    5.9s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

Let’s compare results across different segments.

[14]:

metrics_deepar

[14]:
segment SMAPE MAPE MAE fold_number
1 segment_a 6.352869 6.129627 32.770220 0
1 segment_a 3.673939 3.631979 18.586190 1
1 segment_a 4.346741 4.239818 23.106293 2
2 segment_b 6.138560 5.955818 15.267419 0
2 segment_b 3.833563 3.768379 9.465275 1
2 segment_b 3.281513 3.298270 7.616302 2
0 segment_c 5.416998 5.287060 9.203807 0
0 segment_c 5.808150 5.624208 10.211833 1
0 segment_c 5.375506 5.229224 9.724481 2
3 segment_d 5.030111 4.966043 41.805071 0
3 segment_d 4.040230 4.141370 32.495876 1
3 segment_d 3.253994 3.182568 28.029567 2

To summarize it we will take mean value of SMAPE metric because it is scale tolerant.

[15]:

score = metrics_deepar["SMAPE"].mean()
print(f"Average SMAPE for DeepAR: {score:.3f}")

Average SMAPE for DeepAR: 4.713

Visualize results.

[16]:

plot_backtest(forecast_deepar, ts, history_len=20)
../_images/tutorials_NN_examples_36_0.png

3.2 TFT

Let’s move to the next model.

[17]:

set_seed()

Default way

[18]:

from etna.models.nn import TFTModel

model_tft = TFTModel(
    encoder_length=HORIZON,
    decoder_length=HORIZON,
    trainer_params=dict(max_epochs=200, gpus=0, gradient_clip_val=0.1),
    lr=0.01,
    train_batch_size=64,
)

pipeline_tft = Pipeline(
    model=model_tft,
    horizon=HORIZON,
)

[19]:

metrics_tft, forecast_tft, fold_info_tft = pipeline_tft.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

   | Name                               | Type                            | Params
----------------------------------------------------------------------------------------
0  | loss                               | QuantileLoss                    | 0
1  | logging_metrics                    | ModuleList                      | 0
2  | input_embeddings                   | MultiEmbedding                  | 0
3  | prescalers                         | ModuleDict                      | 96
4  | static_variable_selection          | VariableSelectionNetwork        | 1.7 K
5  | encoder_variable_selection         | VariableSelectionNetwork        | 1.8 K
6  | decoder_variable_selection         | VariableSelectionNetwork        | 1.2 K
7  | static_context_variable_selection  | GatedResidualNetwork            | 1.1 K
8  | static_context_initial_hidden_lstm | GatedResidualNetwork            | 1.1 K
9  | static_context_initial_cell_lstm   | GatedResidualNetwork            | 1.1 K
10 | static_context_enrichment          | GatedResidualNetwork            | 1.1 K
11 | lstm_encoder                       | LSTM                            | 2.2 K
12 | lstm_decoder                       | LSTM                            | 2.2 K
13 | post_lstm_gate_encoder             | GatedLinearUnit                 | 544
14 | post_lstm_add_norm_encoder         | AddNorm                         | 32
15 | static_enrichment                  | GatedResidualNetwork            | 1.4 K
16 | multihead_attn                     | InterpretableMultiHeadAttention | 676
17 | post_attn_gate_norm                | GateAddNorm                     | 576
18 | pos_wise_ff                        | GatedResidualNetwork            | 1.1 K
19 | pre_output_gate_norm               | GateAddNorm                     | 576
20 | output_layer                       | Linear                          | 119
----------------------------------------------------------------------------------------
18.4 K    Trainable params
0         Non-trainable params
18.4 K    Total params
0.074     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=200` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  4.1min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

   | Name                               | Type                            | Params
----------------------------------------------------------------------------------------
0  | loss                               | QuantileLoss                    | 0
1  | logging_metrics                    | ModuleList                      | 0
2  | input_embeddings                   | MultiEmbedding                  | 0
3  | prescalers                         | ModuleDict                      | 96
4  | static_variable_selection          | VariableSelectionNetwork        | 1.7 K
5  | encoder_variable_selection         | VariableSelectionNetwork        | 1.8 K
6  | decoder_variable_selection         | VariableSelectionNetwork        | 1.2 K
7  | static_context_variable_selection  | GatedResidualNetwork            | 1.1 K
8  | static_context_initial_hidden_lstm | GatedResidualNetwork            | 1.1 K
9  | static_context_initial_cell_lstm   | GatedResidualNetwork            | 1.1 K
10 | static_context_enrichment          | GatedResidualNetwork            | 1.1 K
11 | lstm_encoder                       | LSTM                            | 2.2 K
12 | lstm_decoder                       | LSTM                            | 2.2 K
13 | post_lstm_gate_encoder             | GatedLinearUnit                 | 544
14 | post_lstm_add_norm_encoder         | AddNorm                         | 32
15 | static_enrichment                  | GatedResidualNetwork            | 1.4 K
16 | multihead_attn                     | InterpretableMultiHeadAttention | 676
17 | post_attn_gate_norm                | GateAddNorm                     | 576
18 | pos_wise_ff                        | GatedResidualNetwork            | 1.1 K
19 | pre_output_gate_norm               | GateAddNorm                     | 576
20 | output_layer                       | Linear                          | 119
----------------------------------------------------------------------------------------
18.4 K    Trainable params
0         Non-trainable params
18.4 K    Total params
0.074     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=200` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:  8.4min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

   | Name                               | Type                            | Params
----------------------------------------------------------------------------------------
0  | loss                               | QuantileLoss                    | 0
1  | logging_metrics                    | ModuleList                      | 0
2  | input_embeddings                   | MultiEmbedding                  | 0
3  | prescalers                         | ModuleDict                      | 96
4  | static_variable_selection          | VariableSelectionNetwork        | 1.7 K
5  | encoder_variable_selection         | VariableSelectionNetwork        | 1.8 K
6  | decoder_variable_selection         | VariableSelectionNetwork        | 1.2 K
7  | static_context_variable_selection  | GatedResidualNetwork            | 1.1 K
8  | static_context_initial_hidden_lstm | GatedResidualNetwork            | 1.1 K
9  | static_context_initial_cell_lstm   | GatedResidualNetwork            | 1.1 K
10 | static_context_enrichment          | GatedResidualNetwork            | 1.1 K
11 | lstm_encoder                       | LSTM                            | 2.2 K
12 | lstm_decoder                       | LSTM                            | 2.2 K
13 | post_lstm_gate_encoder             | GatedLinearUnit                 | 544
14 | post_lstm_add_norm_encoder         | AddNorm                         | 32
15 | static_enrichment                  | GatedResidualNetwork            | 1.4 K
16 | multihead_attn                     | InterpretableMultiHeadAttention | 676
17 | post_attn_gate_norm                | GateAddNorm                     | 576
18 | pos_wise_ff                        | GatedResidualNetwork            | 1.1 K
19 | pre_output_gate_norm               | GateAddNorm                     | 576
20 | output_layer                       | Linear                          | 119
----------------------------------------------------------------------------------------
18.4 K    Trainable params
0         Non-trainable params
18.4 K    Total params
0.074     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=200` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed: 13.3min remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed: 13.3min finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    2.3s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    5.2s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    7.6s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    7.6s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[20]:

metrics_tft

[20]:
segment SMAPE MAPE MAE fold_number
1 segment_a 41.527952 34.017107 184.644741 0
1 segment_a 8.790662 8.873269 45.135646 1
1 segment_a 7.237774 7.172966 38.038308 2
2 segment_b 31.933354 39.245824 93.498051 0
2 segment_b 8.893341 8.661787 21.545558 1
2 segment_b 6.301268 6.526554 14.349936 2
0 segment_c 66.813178 101.317549 172.212315 0
0 segment_c 13.978753 12.814422 24.043097 1
0 segment_c 8.142750 8.414981 14.742528 2
3 segment_d 83.912047 58.772187 513.073303 0
3 segment_d 15.097971 14.550268 118.252816 1
3 segment_d 26.567021 22.921870 205.484480 2 
[21]:

score = metrics_tft["SMAPE"].mean()
print(f"Average SMAPE for TFT: {score:.3f}")

Average SMAPE for TFT: 26.600

Dataset Builder

[22]:

set_seed()


transform_date = DateFlagsTransform(day_number_in_week=True, day_number_in_month=False, out_column="dateflag")
num_lags = 10
transform_lag = LagTransform(
    in_column="target",
    lags=[HORIZON + i for i in range(num_lags)],
    out_column="target_lag",
)
lag_columns = [f"target_lag_{HORIZON+i}" for i in range(num_lags)]

dataset_builder_tft = PytorchForecastingDatasetBuilder(
    max_encoder_length=HORIZON,
    max_prediction_length=HORIZON,
    time_varying_known_reals=["time_idx"],
    time_varying_unknown_reals=["target"],
    time_varying_known_categoricals=["dateflag_day_number_in_week"],
    static_categoricals=["segment"],
    target_normalizer=GroupNormalizer(groups=["segment"]),
)

[23]:

model_tft = TFTModel(
    dataset_builder=dataset_builder_tft,
    trainer_params=dict(max_epochs=200, gpus=0, gradient_clip_val=0.1),
    lr=0.01,
    train_batch_size=64,
)

pipeline_tft = Pipeline(
    model=model_tft,
    horizon=HORIZON,
    transforms=[transform_lag, transform_date],
)

[24]:

metrics_tft, forecast_tft, fold_info_tft = pipeline_tft.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

   | Name                               | Type                            | Params
----------------------------------------------------------------------------------------
0  | loss                               | QuantileLoss                    | 0
1  | logging_metrics                    | ModuleList                      | 0
2  | input_embeddings                   | MultiEmbedding                  | 47
3  | prescalers                         | ModuleDict                      | 96
4  | static_variable_selection          | VariableSelectionNetwork        | 1.8 K
5  | encoder_variable_selection         | VariableSelectionNetwork        | 1.9 K
6  | decoder_variable_selection         | VariableSelectionNetwork        | 1.3 K
7  | static_context_variable_selection  | GatedResidualNetwork            | 1.1 K
8  | static_context_initial_hidden_lstm | GatedResidualNetwork            | 1.1 K
9  | static_context_initial_cell_lstm   | GatedResidualNetwork            | 1.1 K
10 | static_context_enrichment          | GatedResidualNetwork            | 1.1 K
11 | lstm_encoder                       | LSTM                            | 2.2 K
12 | lstm_decoder                       | LSTM                            | 2.2 K
13 | post_lstm_gate_encoder             | GatedLinearUnit                 | 544
14 | post_lstm_add_norm_encoder         | AddNorm                         | 32
15 | static_enrichment                  | GatedResidualNetwork            | 1.4 K
16 | multihead_attn                     | InterpretableMultiHeadAttention | 676
17 | post_attn_gate_norm                | GateAddNorm                     | 576
18 | pos_wise_ff                        | GatedResidualNetwork            | 1.1 K
19 | pre_output_gate_norm               | GateAddNorm                     | 576
20 | output_layer                       | Linear                          | 119
----------------------------------------------------------------------------------------
18.9 K    Trainable params
0         Non-trainable params
18.9 K    Total params
0.075     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=200` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  5.4min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

   | Name                               | Type                            | Params
----------------------------------------------------------------------------------------
0  | loss                               | QuantileLoss                    | 0
1  | logging_metrics                    | ModuleList                      | 0
2  | input_embeddings                   | MultiEmbedding                  | 47
3  | prescalers                         | ModuleDict                      | 96
4  | static_variable_selection          | VariableSelectionNetwork        | 1.8 K
5  | encoder_variable_selection         | VariableSelectionNetwork        | 1.9 K
6  | decoder_variable_selection         | VariableSelectionNetwork        | 1.3 K
7  | static_context_variable_selection  | GatedResidualNetwork            | 1.1 K
8  | static_context_initial_hidden_lstm | GatedResidualNetwork            | 1.1 K
9  | static_context_initial_cell_lstm   | GatedResidualNetwork            | 1.1 K
10 | static_context_enrichment          | GatedResidualNetwork            | 1.1 K
11 | lstm_encoder                       | LSTM                            | 2.2 K
12 | lstm_decoder                       | LSTM                            | 2.2 K
13 | post_lstm_gate_encoder             | GatedLinearUnit                 | 544
14 | post_lstm_add_norm_encoder         | AddNorm                         | 32
15 | static_enrichment                  | GatedResidualNetwork            | 1.4 K
16 | multihead_attn                     | InterpretableMultiHeadAttention | 676
17 | post_attn_gate_norm                | GateAddNorm                     | 576
18 | pos_wise_ff                        | GatedResidualNetwork            | 1.1 K
19 | pre_output_gate_norm               | GateAddNorm                     | 576
20 | output_layer                       | Linear                          | 119
----------------------------------------------------------------------------------------
18.9 K    Trainable params
0         Non-trainable params
18.9 K    Total params
0.075     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=200` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed: 10.4min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

   | Name                               | Type                            | Params
----------------------------------------------------------------------------------------
0  | loss                               | QuantileLoss                    | 0
1  | logging_metrics                    | ModuleList                      | 0
2  | input_embeddings                   | MultiEmbedding                  | 47
3  | prescalers                         | ModuleDict                      | 96
4  | static_variable_selection          | VariableSelectionNetwork        | 1.8 K
5  | encoder_variable_selection         | VariableSelectionNetwork        | 1.9 K
6  | decoder_variable_selection         | VariableSelectionNetwork        | 1.3 K
7  | static_context_variable_selection  | GatedResidualNetwork            | 1.1 K
8  | static_context_initial_hidden_lstm | GatedResidualNetwork            | 1.1 K
9  | static_context_initial_cell_lstm   | GatedResidualNetwork            | 1.1 K
10 | static_context_enrichment          | GatedResidualNetwork            | 1.1 K
11 | lstm_encoder                       | LSTM                            | 2.2 K
12 | lstm_decoder                       | LSTM                            | 2.2 K
13 | post_lstm_gate_encoder             | GatedLinearUnit                 | 544
14 | post_lstm_add_norm_encoder         | AddNorm                         | 32
15 | static_enrichment                  | GatedResidualNetwork            | 1.4 K
16 | multihead_attn                     | InterpretableMultiHeadAttention | 676
17 | post_attn_gate_norm                | GateAddNorm                     | 576
18 | pos_wise_ff                        | GatedResidualNetwork            | 1.1 K
19 | pre_output_gate_norm               | GateAddNorm                     | 576
20 | output_layer                       | Linear                          | 119
----------------------------------------------------------------------------------------
18.9 K    Trainable params
0         Non-trainable params
18.9 K    Total params
0.075     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=200` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed: 15.4min remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed: 15.4min finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    2.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    4.3s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    6.6s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    6.6s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[25]:

metrics_tft

[25]:
segment SMAPE MAPE MAE fold_number
1 segment_a 5.691613 5.637601 29.017325 0
1 segment_a 8.505932 8.236387 43.995488 1
1 segment_a 3.996154 3.857195 22.390054 2
2 segment_b 5.383510 5.193353 13.481519 0
2 segment_b 6.155114 5.938890 15.099969 1
2 segment_b 3.670533 3.766233 8.381853 2
0 segment_c 4.223216 4.200796 7.158681 0
0 segment_c 2.977674 2.924484 5.396820 1
0 segment_c 6.673994 6.361899 12.073672 2
3 segment_d 9.949646 9.882357 83.306362 0
3 segment_d 2.769709 2.808341 22.188468 1
3 segment_d 4.419314 4.331473 37.982762 2 
[26]:

score = metrics_tft["SMAPE"].mean()
print(f"Average SMAPE for TFT: {score:.3f}")

Average SMAPE for TFT: 5.368

[27]:

plot_backtest(forecast_tft, ts, history_len=20)
../_images/tutorials_NN_examples_51_0.png

3.3 Simple model

For comparison let’s train a much more simpler model.

[28]:

model_sma = SeasonalMovingAverageModel(window=5, seasonality=7)
linear_trend_transform = LinearTrendTransform(in_column="target")

pipeline_sma = Pipeline(model=model_sma, horizon=HORIZON, transforms=[linear_trend_transform])

[29]:

metrics_sma, forecast_sma, fold_info_sma = pipeline_sma.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[30]:

metrics_sma

[30]:
segment SMAPE MAPE MAE fold_number
1 segment_a 6.343943 6.124296 33.196532 0
1 segment_a 5.346946 5.192455 27.938101 1
1 segment_a 7.510347 7.189999 40.028565 2
2 segment_b 7.178822 6.920176 17.818102 0
2 segment_b 5.672504 5.554555 13.719200 1
2 segment_b 3.327846 3.359712 7.680919 2
0 segment_c 6.430429 6.200580 10.877718 0
0 segment_c 5.947090 5.727531 10.701336 1
0 segment_c 6.186545 5.943679 11.359563 2
3 segment_d 4.707899 4.644170 39.918646 0
3 segment_d 5.403426 5.600978 43.047332 1
3 segment_d 2.505279 2.543719 19.347565 2 
[31]:

score = metrics_sma["SMAPE"].mean()
print(f"Average SMAPE for Seasonal MA: {score:.3f}")

Average SMAPE for Seasonal MA: 5.547

[32]:

plot_backtest(forecast_sma, ts, history_len=20)
../_images/tutorials_NN_examples_58_0.png

As we can see, neural networks are a bit better in this particular case.

4. Etna native deep models

We’ve used models from pytorch-forecasting above. Now let’s talk about etna native implementations of deep models for time series.
There is small thing to change: we dont need special PytorchForecastingTransform now.

RNNModel

We’ll use RNN model based on LSTM cell

[33]:

from etna.models.nn import RNNModel
from etna.transforms import StandardScalerTransform

[34]:

model_rnn = RNNModel(
    decoder_length=HORIZON,
    encoder_length=2 * HORIZON,
    input_size=11,
    trainer_params=dict(max_epochs=5),
    lr=1e-3,
)

pipeline_rnn = Pipeline(
    model=model_rnn,
    horizon=HORIZON,
    transforms=[StandardScalerTransform(in_column="target"), transform_lag],
)

[35]:

metrics_rnn, forecast_rnn, fold_info_rnn = pipeline_rnn.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name       | Type    | Params
---------------------------------------
0 | loss       | MSELoss | 0
1 | rnn        | LSTM    | 4.0 K
2 | projection | Linear  | 17
---------------------------------------
4.0 K     Trainable params
0         Non-trainable params
4.0 K     Total params
0.016     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=5` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    4.2s remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name       | Type    | Params
---------------------------------------
0 | loss       | MSELoss | 0
1 | rnn        | LSTM    | 4.0 K
2 | projection | Linear  | 17
---------------------------------------
4.0 K     Trainable params
0         Non-trainable params
4.0 K     Total params
0.016     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=5` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    8.8s remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name       | Type    | Params
---------------------------------------
0 | loss       | MSELoss | 0
1 | rnn        | LSTM    | 4.0 K
2 | projection | Linear  | 17
---------------------------------------
4.0 K     Trainable params
0         Non-trainable params
4.0 K     Total params
0.016     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=5` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   13.5s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   13.5s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.2s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.3s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[36]:

score = metrics_rnn["SMAPE"].mean()
print(f"Average SMAPE for LSTM: {score:.3f}")

Average SMAPE for LSTM: 5.643

[37]:

plot_backtest(forecast_rnn, ts, history_len=20)
../_images/tutorials_NN_examples_65_0.png

Deep State Model

Deep State Model works well with multiple similar time-series. It inffers shared patterns from them.

We have to determine the type of seasonality in data (based on data granularity), SeasonalitySSM class is responsible for this. In this example, we have daily data, so we use day-of-week (7 seasons) and day-of-month (31 seasons) models. We also set the trend component using the LevelTrendSSM class. Also in the model we use time-based features like day-of-week, day-of-month and time independent feature representing the segment of time series.

[38]:

from etna.models.nn import DeepStateModel
from etna.models.nn.deepstate import CompositeSSM, SeasonalitySSM, LevelTrendSSM
from etna.transforms import StandardScalerTransform, DateFlagsTransform, SegmentEncoderTransform

[39]:

HORIZON = 7
metrics = [SMAPE(), MAPE(), MAE()]

[40]:

transforms = [
    SegmentEncoderTransform(),
    StandardScalerTransform(in_column="target"),
    DateFlagsTransform(
        day_number_in_week=True,
        day_number_in_month=True,
        week_number_in_month=False,
        week_number_in_year=False,
        month_number_in_year=False,
        year_number=False,
        is_weekend=False,
        out_column="df",
    ),
]

[41]:

monthly_smm = SeasonalitySSM(num_seasons=31, timestamp_transform=lambda x: x.day - 1)
weekly_smm = SeasonalitySSM(num_seasons=7, timestamp_transform=lambda x: x.weekday())

[42]:

model_dsm = DeepStateModel(
    ssm=CompositeSSM(seasonal_ssms=[weekly_smm, monthly_smm], nonseasonal_ssm=LevelTrendSSM()),
    decoder_length=HORIZON,
    encoder_length=2 * HORIZON,
    input_size=3,
    trainer_params=dict(max_epochs=5),
    lr=1e-3,
)

pipeline_dsm = Pipeline(
    model=model_dsm,
    horizon=HORIZON,
    transforms=transforms,
)

[43]:

metrics_dsm, forecast_dsm, fold_info_dsm = pipeline_dsm.backtest(ts, metrics=metrics, n_folds=3, n_jobs=1)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name       | Type       | Params
------------------------------------------
0 | RNN        | LSTM       | 7.2 K
1 | projectors | ModuleDict | 5.0 K
------------------------------------------
12.2 K    Trainable params
0         Non-trainable params
12.2 K    Total params
0.049     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=5` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   13.0s remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name       | Type       | Params
------------------------------------------
0 | RNN        | LSTM       | 7.2 K
1 | projectors | ModuleDict | 5.0 K
------------------------------------------
12.2 K    Trainable params
0         Non-trainable params
12.2 K    Total params
0.049     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=5` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:   24.2s remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name       | Type       | Params
------------------------------------------
0 | RNN        | LSTM       | 7.2 K
1 | projectors | ModuleDict | 5.0 K
------------------------------------------
12.2 K    Trainable params
0         Non-trainable params
12.2 K    Total params
0.049     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=5` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   40.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   40.0s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.2s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.3s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[44]:

score = metrics_dsm["SMAPE"].mean()
print(f"Average SMAPE for DeepStateModel: {score:.3f}")

Average SMAPE for DeepStateModel: 5.523

[45]:

plot_backtest(forecast_dsm, ts, history_len=20)
../_images/tutorials_NN_examples_74_0.png

N-BEATS Model

This architecture is based on backward and forward residual links and a deep stack of fully connected layers.

There are two types of models in the library. The NBeatsGenericModel class implements a generic deep learning model, while the NBeatsInterpretableModel is augmented with certain inductive biases to be interpretable (trend and seasonality).

[46]:

from etna.models.nn import NBeatsGenericModel
from etna.models.nn import NBeatsInterpretableModel

[47]:

HORIZON = 7
metrics = [SMAPE(), MAPE(), MAE()]

[64]:

model_nbeats_generic = NBeatsGenericModel(
    input_size=2 * HORIZON,
    output_size=HORIZON,
    loss="smape",
    stacks=30,
    layers=4,
    layer_size=256,
    trainer_params=dict(max_epochs=1000),
    lr=1e-3,
)

pipeline_nbeats_generic = Pipeline(
    model=model_nbeats_generic,
    horizon=HORIZON,
    transforms=[],
)

[65]:

metrics_nbeats_generic, forecast_nbeats_generic, _ = pipeline_nbeats_generic.backtest(
    ts, metrics=metrics, n_folds=3, n_jobs=1
)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name  | Type        | Params
--------------------------------------
0 | model | NBeats      | 206 K
1 | loss  | NBeatsSMAPE | 0
--------------------------------------
206 K     Trainable params
0         Non-trainable params
206 K     Total params
0.826     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=1000` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  1.1min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name  | Type        | Params
--------------------------------------
0 | model | NBeats      | 206 K
1 | loss  | NBeatsSMAPE | 0
--------------------------------------
206 K     Trainable params
0         Non-trainable params
206 K     Total params
0.826     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=1000` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:  2.2min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name  | Type        | Params
--------------------------------------
0 | model | NBeats      | 206 K
1 | loss  | NBeatsSMAPE | 0
--------------------------------------
206 K     Trainable params
0         Non-trainable params
206 K     Total params
0.826     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=1000` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  3.3min remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  3.3min finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[66]:

score = metrics_nbeats_generic["SMAPE"].mean()
print(f"Average SMAPE for N-BEATS Generic: {score:.3f}")

Average SMAPE for N-BEATS Generic: 5.254

[67]:

plot_backtest(forecast_nbeats_generic, ts, history_len=20)
../_images/tutorials_NN_examples_81_0.png 
[76]:

model_nbeats_interp = NBeatsInterpretableModel(
    input_size=4 * HORIZON,
    output_size=HORIZON,
    loss="smape",
    trend_layer_size=64,
    seasonality_layer_size=256,
    trainer_params=dict(max_epochs=2000),
    lr=1e-3,
)

pipeline_nbeats_interp = Pipeline(
    model=model_nbeats_interp,
    horizon=HORIZON,
    transforms=[],
)

[77]:

metrics_nbeats_interp, forecast_nbeats_interp, _ = pipeline_nbeats_interp.backtest(
    ts, metrics=metrics, n_folds=3, n_jobs=1
)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name  | Type        | Params
--------------------------------------
0 | model | NBeats      | 224 K
1 | loss  | NBeatsSMAPE | 0
--------------------------------------
223 K     Trainable params
385       Non-trainable params
224 K     Total params
0.896     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=2000` reached.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  1.4min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name  | Type        | Params
--------------------------------------
0 | model | NBeats      | 224 K
1 | loss  | NBeatsSMAPE | 0
--------------------------------------
223 K     Trainable params
385       Non-trainable params
224 K     Total params
0.896     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=2000` reached.
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:  3.0min remaining:    0.0s
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs

  | Name  | Type        | Params
--------------------------------------
0 | model | NBeats      | 224 K
1 | loss  | NBeatsSMAPE | 0
--------------------------------------
223 K     Trainable params
385       Non-trainable params
224 K     Total params
0.896     Total estimated model params size (MB)

`Trainer.fit` stopped: `max_epochs=2000` reached.
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  4.4min remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  4.4min finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished

[78]:

score = metrics_nbeats_interp["SMAPE"].mean()
print(f"Average SMAPE for N-BEATS Interpretable: {score:.3f}")

Average SMAPE for N-BEATS Interpretable: 4.987

[79]:

plot_backtest(forecast_nbeats_interp, ts, history_len=20)
../_images/tutorials_NN_examples_85_0.png