Added multiple regressors

config.py

@@ -0,0 +1,18 @@
+import logging
+import yaml
+
+logging.basicConfig(
+    format="%(asctime)s.%(msecs)03d %(message)s",
+    datefmt="%Y-%m-%d,%H:%M:%S",
+    level=logging.INFO,
+)
+logger = logging.getLogger()
+logger.setLevel(logging.INFO)
+
+
+def load_config(path):
+
+    with open(path) as f:
+        config = yaml.load(f, Loader=yaml.Loader)
+
+    return config

config.yaml

@@ -0,0 +1,8 @@
+model:
+  batch_size: 128
+  max_epochs: 150
+  gpus: 0
+  learning_rate: 0.001
+  hidden_size: 16
+  dropout: 0.1
+  hidden_continuous_size: 8

predict_regressors.py

@@ -1,15 +1,35 @@
 import torch
 import pandas as pd
+import matplotlib.pyplot as plt
 
 from pytorch_forecasting.metrics import SMAPE, PoissonLoss, QuantileLoss
 from pytorch_forecasting import Baseline, TemporalFusionTransformer, TimeSeriesDataSet
 from pytorch_forecasting.data import GroupNormalizer
 
-import matplotlib.pyplot as plt
+from config import load_config
+
+spec = load_config("config.yaml")
+BATCH_SIZE = spec["model"]["batch_size"]
+MAX_EPOCHS = spec["model"]["max_epochs"]
+GPUS = spec["model"]["gpus"]
+LEARNING_RATE = spec["model"]["learning_rate"]
+HIDDEN_SIZE = spec["model"]["hidden_size"]
+DROPOUT = spec["model"]["dropout"]
+HIDDEN_CONTINUOUS_SIZE = spec["model"]["hidden_continuous_size"]
 
 data = pd.read_csv("data/MERCHANT_NUMBER_OF_TRX.csv")
-data = data[["MERCHANT_1_NUMBER_OF_TRX", "date"]]
-data["id"] = "M1"
+data = data[[
+    "MERCHANT_1_NUMBER_OF_TRX",
+    "MERCHANT_2_NUMBER_OF_TRX",
+    "date"
+]]
+data = data.set_index("date").stack().reset_index()
+data = data.rename(
+    columns={
+        'level_1': 'id',
+        0: 'trx'
+    }
+)
 
 # add time index
 data["time_idx"] = pd.to_datetime(data.date).astype(int)
@@ -29,15 +49,15 @@
     .astype("category")
 
 # cut atypical values at the end of the sample
-train_data = data[:3200]
+train_data = data[:3200*2]
 max_prediction_length = 24
 max_encoder_length = 72
-training_cutoff = train_data["time_idx"].max() - max_prediction_length
+training_cutoff = data["time_idx"].max() - max_prediction_length
 
 training = TimeSeriesDataSet(
     train_data[lambda x: x.time_idx <= training_cutoff],
     time_idx="time_idx",
-    target="MERCHANT_1_NUMBER_OF_TRX",
+    target="trx",
     group_ids=["id"],
     min_encoder_length=max_encoder_length // 2,  # keep encoder length long (as it is in the validation set)
     max_encoder_length=max_encoder_length,
@@ -47,7 +67,7 @@
     time_varying_known_reals=["time_idx"],
     time_varying_known_categoricals=["hour", "month", "day_of_week"],
     time_varying_unknown_categoricals=[],
-    time_varying_unknown_reals=["MERCHANT_1_NUMBER_OF_TRX"],
+    time_varying_unknown_reals=["trx"],
     target_normalizer=GroupNormalizer(
         groups=["id"], transformation="softplus"
     ),
@@ -59,11 +79,11 @@
 
 model = TemporalFusionTransformer.from_dataset(
     training,
-    learning_rate=0.001,
-    hidden_size=16,
+    learning_rate=LEARNING_RATE,
+    hidden_size=HIDDEN_SIZE,
     attention_head_size=1,
-    dropout=0.1,
-    hidden_continuous_size=8,
+    dropout=DROPOUT,
+    hidden_continuous_size=HIDDEN_CONTINUOUS_SIZE,
     output_size=7,  # 7 quantiles by default
     loss=QuantileLoss(),
     log_interval=10,  # uncomment for learning rate finder and otherwise, e.g. to 10 for logging every 10 batches
@@ -72,31 +92,37 @@
 
 model.load_state_dict(torch.load("model/tft_regressor.pt"))
 
-for start in range(3200, 3700, 72):
-    test_data = data[start:(start + max_encoder_length)]
-    y_obs = data[(start + max_encoder_length): (start + max_encoder_length + max_prediction_length)]
-
-    y_hat = model.predict(
-        test_data,
-        mode="prediction",
-        return_x=True
-    )[0][0].tolist()
-
-    fig, ax = plt.subplots()
-
-    ax.plot(
-        pd.Series(data=y_hat,
-                  index=pd.to_datetime(y_obs.date)),
-        label='forecast'
-    )
-
-    ax.plot(
-        y_obs.set_index(pd.to_datetime(y_obs.date))["MERCHANT_1_NUMBER_OF_TRX"],
-        color='orange',
-        alpha=0.8,
-        label='observed',
-        linestyle='--',
-        linewidth=1.2
-    )
-
-    plt.show()
+for target in ["MERCHANT_1_NUMBER_OF_TRX", "MERCHANT_2_NUMBER_OF_TRX"]:
+
+    df = data[data["id"] == target].reset_index(drop=True)
+
+    for start in range(3200, 3500, 72):
+        test_data = df[start:(start + max_encoder_length)]
+        y_obs = df[(start + max_encoder_length): (start + max_encoder_length + max_prediction_length)]
+
+        y_hat = model.predict(
+            test_data,
+            mode="prediction",
+            return_x=True
+        )[0][0].tolist()
+
+        fig, ax = plt.subplots()
+
+        ax.plot(
+            pd.Series(data=y_hat,
+                      index=pd.to_datetime(y_obs.date)),
+            label='forecast'
+        )
+
+        ax.plot(
+            y_obs.set_index(pd.to_datetime(y_obs.date))["trx"],
+            color='orange',
+            alpha=0.8,
+            label='observed',
+            linestyle='--',
+            linewidth=1.2
+        )
+
+        plt.title(f'{target}')
+        plt.legend()
+        plt.show()

requirements.txt

@@ -3,4 +3,5 @@ pyarrow==3.0.0
 pandas==1.2.5
 numpy==1.21.3
 torch==1.10.0
-matplotlib~=3.4.3
+matplotlib~=3.4.3
+PyYAML~=6.0

train_regressors.py

@@ -1,12 +1,5 @@
-import copy
-from pathlib import Path
-import os
 import warnings
-import pickle
 
-warnings.filterwarnings("ignore")  # avoid printing out absolute paths
-
-import numpy as np
 import pandas as pd
 import pytorch_lightning as pl
 from pytorch_lightning.callbacks import EarlyStopping, LearningRateMonitor
@@ -19,9 +12,33 @@
 from pytorch_forecasting.models.temporal_fusion_transformer.tuning import optimize_hyperparameters
 from pytorch_forecasting.data.examples import get_stallion_data
 
+from config import load_config
+
+warnings.filterwarnings("ignore")  # avoid printing out absolute paths
+
+spec = load_config("config.yaml")
+BATCH_SIZE = spec["model"]["batch_size"]
+MAX_EPOCHS = spec["model"]["max_epochs"]
+GPUS = spec["model"]["gpus"]
+LEARNING_RATE = spec["model"]["learning_rate"]
+HIDDEN_SIZE = spec["model"]["hidden_size"]
+DROPOUT = spec["model"]["dropout"]
+HIDDEN_CONTINUOUS_SIZE = spec["model"]["hidden_continuous_size"]
+
+
 data = pd.read_csv("data/MERCHANT_NUMBER_OF_TRX.csv")
-data = data[["MERCHANT_1_NUMBER_OF_TRX", "date"]]
-data["id"] = "M1"
+data = data[[
+    "MERCHANT_1_NUMBER_OF_TRX",
+    "MERCHANT_2_NUMBER_OF_TRX",
+    "date"
+]]
+data = data.set_index("date").stack().reset_index()
+data = data.rename(
+    columns={
+        'level_1': 'id',
+        0: 'trx'
+    }
+)
 
 # add time index
 data["time_idx"] = pd.to_datetime(data.date).astype(int)
@@ -41,19 +58,15 @@
     .astype("category")
 
 # cut atypical values at the end of the sample
-# data = data[:3840]
-data = data[:3200]
-
+train_data = data[:3200*2]
 max_prediction_length = 24
 max_encoder_length = 72
 training_cutoff = data["time_idx"].max() - max_prediction_length
 
-test_data = data[lambda x: x.time_idx > x.time_idx.max() - max_encoder_length]
-
 training = TimeSeriesDataSet(
-    data[lambda x: x.time_idx <= training_cutoff],
+    train_data[lambda x: x.time_idx <= training_cutoff],
     time_idx="time_idx",
-    target="MERCHANT_1_NUMBER_OF_TRX",
+    target="trx",
     group_ids=["id"],
     min_encoder_length=max_encoder_length // 2,  # keep encoder length long (as it is in the validation set)
     max_encoder_length=max_encoder_length,
@@ -63,7 +76,7 @@
     time_varying_known_reals=["time_idx"],
     time_varying_known_categoricals=["hour", "month", "day_of_week"],
     time_varying_unknown_categoricals=[],
-    time_varying_unknown_reals=["MERCHANT_1_NUMBER_OF_TRX"],
+    time_varying_unknown_reals=["trx"],
     target_normalizer=GroupNormalizer(
         groups=["id"], transformation="softplus"
     ),
@@ -75,12 +88,11 @@
 
 # create validation set (predict=True) which means to predict the last max_prediction_length points in time
 # for each series
-validation = TimeSeriesDataSet.from_dataset(training, data, predict=True, stop_randomization=True)
+validation = TimeSeriesDataSet.from_dataset(training, train_data, predict=True, stop_randomization=True)
 
 # create dataloaders for model
-batch_size = 128  # set this between 32 to 128
-train_dataloader = training.to_dataloader(train=True, batch_size=batch_size, num_workers=0)
-val_dataloader = validation.to_dataloader(train=False, batch_size=batch_size * 10, num_workers=0)
+train_dataloader = training.to_dataloader(train=True, batch_size=BATCH_SIZE, num_workers=0)
+val_dataloader = validation.to_dataloader(train=False, batch_size=BATCH_SIZE * 10, num_workers=0)
 
 # calculate baseline mean absolute error, i.e. predict next value as the last available value from the history
 actuals = torch.cat([y for x, (y, weight) in iter(val_dataloader)])
@@ -96,24 +108,23 @@
 logger = TensorBoardLogger("lightning_logs")  # logging results to a tensorboard
 
 trainer = pl.Trainer(
-    max_epochs=150,
-    gpus=0,
+    max_epochs=MAX_EPOCHS,
+    gpus=GPUS,
     weights_summary="top",
     gradient_clip_val=0.1,
-    # limit_train_batches=30,  # comment in for training, running valiation every 30 batches
-    # fast_dev_run=True,  # comment in to check that networkor dataset has no serious bugs
+    # limit_train_batches=30,  # comment in for training, running validation every 30 batches
+    # fast_dev_run=True,  # comment in to check that network or dataset has no serious bugs
     callbacks=[lr_logger, early_stop_callback],
     logger=logger,
 )
 
-
 tft = TemporalFusionTransformer.from_dataset(
     training,
-    learning_rate=0.001,
-    hidden_size=16,
+    learning_rate=LEARNING_RATE,
+    hidden_size=HIDDEN_SIZE,
     attention_head_size=1,
-    dropout=0.1,
-    hidden_continuous_size=8,
+    dropout=DROPOUT,
+    hidden_continuous_size=HIDDEN_CONTINUOUS_SIZE,
     output_size=7,  # 7 quantiles by default
     loss=QuantileLoss(),
     log_interval=10,  # uncomment for learning rate finder and otherwise, e.g. to 10 for logging every 10 batches
@@ -129,5 +140,3 @@
 )
 
 torch.save(tft.state_dict(), "model/tft_regressor.pt")
-
-