Simple Regression Model¶

In [1]:

from river import (
    metrics,
    compose,
    preprocessing,
    datasets,
    stats,
    feature_extraction,
)
from deep_river.regression import Regressor
from torch import nn
from pprint import pprint
from tqdm import tqdm

from river import ( metrics, compose, preprocessing, datasets, stats, feature_extraction, ) from deep_river.regression import Regressor from torch import nn from pprint import pprint from tqdm import tqdm

In [2]:

dataset = datasets.Bikes()

for x, y in dataset:
    pprint(x)
    print(f"Number of available bikes: {y}")
    break

dataset = datasets.Bikes() for x, y in dataset: pprint(x) print(f"Number of available bikes: {y}") break

{'clouds': 75,
 'description': 'light rain',
 'humidity': 81,
 'moment': datetime.datetime(2016, 4, 1, 0, 0, 7),
 'pressure': 1017.0,
 'station': 'metro-canal-du-midi',
 'temperature': 6.54,
 'wind': 9.3}
Number of available bikes: 1

In [3]:

class MyModule(nn.Module):
    def __init__(self, n_features):
        super(MyModule, self).__init__()
        self.dense0 = nn.Linear(n_features, 5)
        self.nonlin = nn.ReLU()
        self.dense1 = nn.Linear(5, 1)
        self.softmax = nn.Softmax(dim=-1)

    def forward(self, X, **kwargs):
        X = self.nonlin(self.dense0(X))
        X = self.nonlin(self.dense1(X))
        X = self.softmax(X)
        return X


def get_hour(x):
    x["hour"] = x["moment"].hour
    return x

class MyModule(nn.Module): def __init__(self, n_features): super(MyModule, self).__init__() self.dense0 = nn.Linear(n_features, 5) self.nonlin = nn.ReLU() self.dense1 = nn.Linear(5, 1) self.softmax = nn.Softmax(dim=-1) def forward(self, X, **kwargs): X = self.nonlin(self.dense0(X)) X = self.nonlin(self.dense1(X)) X = self.softmax(X) return X def get_hour(x): x["hour"] = x["moment"].hour return x

In [4]:

metric = metrics.MAE()

model_pipeline = compose.Select(
    "clouds", "humidity", "pressure", "temperature", "wind"
)
model_pipeline += get_hour | feature_extraction.TargetAgg(
    by=["station", "hour"], how=stats.Mean()
)
model_pipeline |= preprocessing.StandardScaler()
model_pipeline |= Regressor(module=MyModule, loss_fn="mse", optimizer_fn="sgd")
model_pipeline

metric = metrics.MAE() model_pipeline = compose.Select( "clouds", "humidity", "pressure", "temperature", "wind" ) model_pipeline += get_hour | feature_extraction.TargetAgg( by=["station", "hour"], how=stats.Mean() ) model_pipeline |= preprocessing.StandardScaler() model_pipeline |= Regressor(module=MyModule, loss_fn="mse", optimizer_fn="sgd") model_pipeline

Out[4]:

['clouds', [...]

Select ( clouds humidity pressure temperature wind )

get_hour

def get_hour(x): x["hour"] = x["moment"].hour return x

y_mean_by_station_and_hour

TargetAgg ( by=['station', 'hour'] how=Mean () target_name="y" )

StandardScaler

StandardScaler ( with_std=True )

Regressor

Regressor ( module=None loss_fn="mse" optimizer_fn="sgd" lr=0.001 is_feature_incremental=False device="cpu" seed=42 )

In [5]:

for x, y in tqdm(dataset.take(5000)):
    y_pred = model_pipeline.predict_one(x)
    metric.update(y_true=y, y_pred=y_pred)
    model_pipeline.learn_one(x=x, y=y)
print(f"MAE: {metric.get():.2f}")

for x, y in tqdm(dataset.take(5000)): y_pred = model_pipeline.predict_one(x) metric.update(y_true=y, y_pred=y_pred) model_pipeline.learn_one(x=x, y=y) print(f"MAE: {metric.get():.2f}")

0it [00:00, ?it/s]

1it [00:00,  1.23it/s]

129it [00:00, 192.04it/s]

260it [00:01, 388.31it/s]

385it [00:01, 560.43it/s]

508it [00:01, 708.11it/s]

638it [00:01, 848.91it/s]

758it [00:01, 938.58it/s]

888it [00:01, 1034.05it/s]

1017it [00:01, 1103.51it/s]

1142it [00:01, 1138.26it/s]

1271it [00:01, 1181.32it/s]

1402it [00:01, 1216.95it/s]

1534it [00:02, 1243.86it/s]

1665it [00:02, 1262.07it/s]

1794it [00:02, 1234.51it/s]

1925it [00:02, 1254.09it/s]

2053it [00:02, 1258.22it/s]

2183it [00:02, 1270.42it/s]

2311it [00:02, 1264.28it/s]

2441it [00:02, 1272.27it/s]

2569it [00:02, 1261.41it/s]

2696it [00:02, 1249.11it/s]

2827it [00:03, 1265.79it/s]

2954it [00:03, 1260.24it/s]

3082it [00:03, 1265.05it/s]

3210it [00:03, 1267.21it/s]

3341it [00:03, 1279.19it/s]

3472it [00:03, 1287.47it/s]

3605it [00:03, 1297.92it/s]

3735it [00:03, 1281.67it/s]

3864it [00:03, 1281.18it/s]

3994it [00:03, 1286.05it/s]

4130it [00:04, 1306.16it/s]

4264it [00:04, 1314.29it/s]

4396it [00:04, 1315.94it/s]

4528it [00:04, 1305.63it/s]

4659it [00:04, 1298.81it/s]

4795it [00:04, 1316.10it/s]

4929it [00:04, 1322.14it/s]

5000it [00:04, 1061.07it/s]

MAE: 6.83