app.py

# Install necessary packages (only for local environment)
# !pip install pandas granite-tsfm

import pandas as pd
from granite_tsfm import TimeSeriesPreprocessor, TinyTimeMixerForPrediction, TimeSeriesForecastingPipeline

# Load dataset (Replace with actual dataset)
data = pd.read_csv('your_dataset.csv', parse_dates=['timestamp_column'])

# Preprocess the data
tsp = TimeSeriesPreprocessor(
    id_columns=[],
    timestamp_column='timestamp_column',
    target_columns=['value1', 'value2'],  # Replace with your target column names
    prediction_length=96,
    context_length=512,
    scaling=True
)
processed_data = tsp.fit_transform(data)

# Load the pre-trained model
model = TinyTimeMixerForPrediction.from_pretrained(
    'ibm-granite/granite-timeseries-ttm-r2',
    num_input_channels=tsp.num_input_channels
)

# Generate forecasts
pipeline = TimeSeriesForecastingPipeline(
    model=model,
    feature_extractor=tsp
)
forecasts = pipeline(data)

# Display the forecasts
print(forecasts)