Informer model¶

In [1]:

#!pip install deepts_forecasting

#!pip install deepts_forecasting

Import libraries¶

In [1]:

import numpy as np
import torch
from torch.utils.data import DataLoader
import pytorch_lightning as pl
from pytorch_lightning.callbacks import EarlyStopping, LearningRateMonitor
from pytorch_lightning.loggers import TensorBoardLogger

from deepts_forecasting.utils.data import TimeSeriesDataSet
from deepts_forecasting.utils.data.encoders import TorchNormalizer
from deepts_forecasting.datasets import AirPassengersDataset
from deepts_forecasting.models.informer import Informer

import numpy as np import torch from torch.utils.data import DataLoader import pytorch_lightning as pl from pytorch_lightning.callbacks import EarlyStopping, LearningRateMonitor from pytorch_lightning.loggers import TensorBoardLogger from deepts_forecasting.utils.data import TimeSeriesDataSet from deepts_forecasting.utils.data.encoders import TorchNormalizer from deepts_forecasting.datasets import AirPassengersDataset from deepts_forecasting.models.informer import Informer

D:\Anaconda3\envs\DeepTS_Forecasting\lib\site-packages\tqdm\auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
  from .autonotebook import tqdm as notebook_tqdm

Dataset¶

In [2]:

data = AirPassengersDataset().load()
data['year'] = data['Month'].dt.year
data['month'] = data['Month'].dt.month
data['group'] = '0'
data['time_idx'] = np.arange(len(data))
data['Passengers'] = data['Passengers'].astype(float)
data['month'] = data['month'].astype('str')
data.head()

data = AirPassengersDataset().load() data['year'] = data['Month'].dt.year data['month'] = data['Month'].dt.month data['group'] = '0' data['time_idx'] = np.arange(len(data)) data['Passengers'] = data['Passengers'].astype(float) data['month'] = data['month'].astype('str') data.head()

Out[2]:

	Month	Passengers	year	month	group	time_idx
0	1949-01-01	112.0	1949	1	0	0
1	1949-02-01	118.0	1949	2	0	1
2	1949-03-01	132.0	1949	3	0	2
3	1949-04-01	129.0	1949	4	0	3
4	1949-05-01	121.0	1949	5	0	4

Split train/test sets¶

In [3]:

max_encoder_length = 18
max_prediction_length = 12

training_cutoff = data["time_idx"].max() - max_encoder_length - max_prediction_length

training = TimeSeriesDataSet(
    data[lambda x: x.time_idx <= training_cutoff],
    max_encoder_length= max_encoder_length,
    min_encoder_length=max_encoder_length,
    max_prediction_length=max_prediction_length,
    min_prediction_length=max_prediction_length,
    time_idx="time_idx",
    target="Passengers",
    group_ids=["group"],
    static_categoricals=[],
    static_reals=[],
    time_varying_known_categoricals=['month'],
    time_varying_known_reals=[],
    time_varying_unknown_reals=["Passengers"],
    time_varying_unknown_categoricals=[],
    target_normalizer=TorchNormalizer(method="standard",
                                      transformation=None),
    )

training.get_parameters()
validation = TimeSeriesDataSet.from_dataset(training,
                                            data[lambda x: x.time_idx > training_cutoff])

batch_size = 16
train_dataloader = DataLoader(training, batch_size=batch_size, shuffle=False, drop_last=False)
val_dataloader = DataLoader(validation, batch_size=batch_size, shuffle=False, drop_last=False)

max_encoder_length = 18 max_prediction_length = 12 training_cutoff = data["time_idx"].max() - max_encoder_length - max_prediction_length training = TimeSeriesDataSet( data[lambda x: x.time_idx <= training_cutoff], max_encoder_length= max_encoder_length, min_encoder_length=max_encoder_length, max_prediction_length=max_prediction_length, min_prediction_length=max_prediction_length, time_idx="time_idx", target="Passengers", group_ids=["group"], static_categoricals=[], static_reals=[], time_varying_known_categoricals=['month'], time_varying_known_reals=[], time_varying_unknown_reals=["Passengers"], time_varying_unknown_categoricals=[], target_normalizer=TorchNormalizer(method="standard", transformation=None), ) training.get_parameters() validation = TimeSeriesDataSet.from_dataset(training, data[lambda x: x.time_idx > training_cutoff]) batch_size = 16 train_dataloader = DataLoader(training, batch_size=batch_size, shuffle=False, drop_last=False) val_dataloader = DataLoader(validation, batch_size=batch_size, shuffle=False, drop_last=False)

Define model¶

In [8]:

pl.seed_everything(1234)
# create PyTorch Lighning Trainer with early stopping
early_stop_callback = EarlyStopping(monitor="val_loss", min_delta=1e-4,
                                    patience=60, verbose=False, mode="min")
lr_logger = LearningRateMonitor()

trainer = pl.Trainer(
    max_epochs=300,
    gpus=0,  # run on CPU, if on multiple GPUs, use accelerator="ddp"
    gradient_clip_val=0.1,
    limit_train_batches=30,  # 30 batches per epoch
    callbacks=[lr_logger, early_stop_callback],
    logger=TensorBoardLogger("lightning_logs")
)

model = Informer.from_dataset(training,
                                  d_model=16,
                                  d_ff=16,
                                  n_heads=1,
                                  num_layers=2,
                                      )
model.summarize

pl.seed_everything(1234) # create PyTorch Lighning Trainer with early stopping early_stop_callback = EarlyStopping(monitor="val_loss", min_delta=1e-4, patience=60, verbose=False, mode="min") lr_logger = LearningRateMonitor() trainer = pl.Trainer( max_epochs=300, gpus=0, # run on CPU, if on multiple GPUs, use accelerator="ddp" gradient_clip_val=0.1, limit_train_batches=30, # 30 batches per epoch callbacks=[lr_logger, early_stop_callback], logger=TensorBoardLogger("lightning_logs") ) model = Informer.from_dataset(training, d_model=16, d_ff=16, n_heads=1, num_layers=2, ) model.summarize

Global seed set to 1234
GPU available: False, used: False
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs

Out[8]:

<bound method LightningModule.summarize of Informer(
  (loss): L1Loss()
  (logging_metrics): ModuleList()
  (encoder_input_linear): Linear(in_features=7, out_features=16, bias=True)
  (decoder_input_linear): Linear(in_features=6, out_features=16, bias=True)
  (encoder_positional_encoding): PositionalEncoding(
    (dropout): Dropout(p=0.1, inplace=False)
  )
  (decoder_positional_encoding): PositionalEncoding(
    (dropout): Dropout(p=0.1, inplace=False)
  )
  (informer_encoder): InformerEncoder(
    (encoder_layers): ModuleList(
      (0): InformerEncoderLayer(
        (attention): AttentionLayer(
          (inner_attention): ProbAttention(
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (query_projection): Linear(in_features=16, out_features=16, bias=True)
          (key_projection): Linear(in_features=16, out_features=16, bias=True)
          (value_projection): Linear(in_features=16, out_features=16, bias=True)
          (out_projection): Linear(in_features=16, out_features=16, bias=True)
        )
        (conv1): Conv1d(16, 16, kernel_size=(1,), stride=(1,))
        (conv2): Conv1d(16, 16, kernel_size=(1,), stride=(1,))
        (norm1): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (norm2): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
      (1): InformerEncoderLayer(
        (attention): AttentionLayer(
          (inner_attention): ProbAttention(
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (query_projection): Linear(in_features=16, out_features=16, bias=True)
          (key_projection): Linear(in_features=16, out_features=16, bias=True)
          (value_projection): Linear(in_features=16, out_features=16, bias=True)
          (out_projection): Linear(in_features=16, out_features=16, bias=True)
        )
        (conv1): Conv1d(16, 16, kernel_size=(1,), stride=(1,))
        (conv2): Conv1d(16, 16, kernel_size=(1,), stride=(1,))
        (norm1): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (norm2): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
    )
    (conv_layers): ModuleList(
      (0): ConvLayer(
        (downConv): Conv1d(16, 16, kernel_size=(3,), stride=(1,), padding=(2,), padding_mode=circular)
        (norm): BatchNorm1d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (activation): ELU(alpha=1.0)
        (maxPool): MaxPool1d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
      )
    )
    (norm): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
  )
  (informer_decoder): InformerDecoder(
    (layers): ModuleList(
      (0): InformerDecoderLayer(
        (self_attention): AttentionLayer(
          (inner_attention): FullAttention(
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (query_projection): Linear(in_features=16, out_features=16, bias=True)
          (key_projection): Linear(in_features=16, out_features=16, bias=True)
          (value_projection): Linear(in_features=16, out_features=16, bias=True)
          (out_projection): Linear(in_features=16, out_features=16, bias=True)
        )
        (cross_attention): AttentionLayer(
          (inner_attention): FullAttention(
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (query_projection): Linear(in_features=16, out_features=16, bias=True)
          (key_projection): Linear(in_features=16, out_features=16, bias=True)
          (value_projection): Linear(in_features=16, out_features=16, bias=True)
          (out_projection): Linear(in_features=16, out_features=16, bias=True)
        )
        (conv1): Conv1d(16, 64, kernel_size=(1,), stride=(1,))
        (conv2): Conv1d(64, 16, kernel_size=(1,), stride=(1,))
        (norm1): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (norm2): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (norm3): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
      (1): InformerDecoderLayer(
        (self_attention): AttentionLayer(
          (inner_attention): FullAttention(
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (query_projection): Linear(in_features=16, out_features=16, bias=True)
          (key_projection): Linear(in_features=16, out_features=16, bias=True)
          (value_projection): Linear(in_features=16, out_features=16, bias=True)
          (out_projection): Linear(in_features=16, out_features=16, bias=True)
        )
        (cross_attention): AttentionLayer(
          (inner_attention): FullAttention(
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (query_projection): Linear(in_features=16, out_features=16, bias=True)
          (key_projection): Linear(in_features=16, out_features=16, bias=True)
          (value_projection): Linear(in_features=16, out_features=16, bias=True)
          (out_projection): Linear(in_features=16, out_features=16, bias=True)
        )
        (conv1): Conv1d(16, 64, kernel_size=(1,), stride=(1,))
        (conv2): Conv1d(64, 16, kernel_size=(1,), stride=(1,))
        (norm1): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (norm2): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (norm3): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
    )
    (norm): LayerNorm((16,), eps=1e-05, elementwise_affine=True)
  )
  (out_linear): Linear(in_features=16, out_features=1, bias=True)
  (embeddings): ModuleDict(
    (month): Embedding(12, 6)
  )
)>

In [9]:

model.hparams

model.hparams

Out[9]:

"activation":                        relu
"categorical_groups":                {}
"d_ff":                              16
"d_model":                           16
"dropout":                           0.1
"embedding_labels":                  {'month': array(['1', '10', '11', '12', '2', '3', '4', '5', '6', '7', '8', '9'],
      dtype=object)}
"embedding_paddings":                []
"embedding_sizes":                   {'month': [12, 6]}
"learning_rate":                     0.001
"log_interval":                      -1
"log_val_interval":                  None
"logging_metrics":                   ModuleList()
"loss":                              L1Loss()
"max_encoder_length":                18
"max_prediction_length":             12
"monotone_constaints":               {}
"n_heads":                           1
"num_layers":                        2
"output_size":                       1
"output_transformer":                TorchNormalizer()
"static_categoricals":               []
"static_reals":                      []
"time_varying_categoricals_decoder": ['month']
"time_varying_categoricals_encoder": ['month']
"time_varying_reals_decoder":        []
"time_varying_reals_encoder":        ['Passengers']
"x_categoricals":                    ['month']
"x_reals":                           ['Passengers']

Train model with early stopping¶

In [ ]:

trainer.fit(
    model, train_dataloader=train_dataloader, val_dataloaders=val_dataloader,
)

# (given that we use early stopping, this is not necessarily the last epoch)
best_model_path = trainer.checkpoint_callback.best_model_path
best_model = Informer.load_from_checkpoint(best_model_path)

# calcualte mean absolute error on validation set
actuals = torch.cat([model.transform_output(prediction=y, target_scale=x['target_scale'])
                     for x, y in iter(val_dataloader)])
predictions, x_index = best_model.predict(val_dataloader)
mae = (actuals - predictions).abs().mean()
# print('predictions shape is:', predictions.shape)
# print('actuals shape is:', actuals.shape)
print(torch.cat([actuals, predictions]))
print('MAE is:', mae)

trainer.fit( model, train_dataloader=train_dataloader, val_dataloaders=val_dataloader, ) # (given that we use early stopping, this is not necessarily the last epoch) best_model_path = trainer.checkpoint_callback.best_model_path best_model = Informer.load_from_checkpoint(best_model_path) # calcualte mean absolute error on validation set actuals = torch.cat([model.transform_output(prediction=y, target_scale=x['target_scale']) for x, y in iter(val_dataloader)]) predictions, x_index = best_model.predict(val_dataloader) mae = (actuals - predictions).abs().mean() # print('predictions shape is:', predictions.shape) # print('actuals shape is:', actuals.shape) print(torch.cat([actuals, predictions])) print('MAE is:', mae)