predict_week.py

from time import time
from predict import get_predictor_values, predict
from tqdm import tqdm


def predict_week():
    predictions = list()
    use_predicted = list()
    hour = (((int(time() // 86400)) * 86400 + 4 * 60 * 60) * 1000) - (24*60*60*1000)
    print(hour)
    feat_temp = get_predictors()
    feat_dewpt = get_predictors(dewpt=True)
    for i in tqdm(range(0, 169)):
        use_predicted.append(i)
        df_temp = get_predictor_values(
            hour, feat_temp, use_predicted, predictions)
        df_dewpt = get_predictor_values(
            hour, feat_dewpt, use_predicted, predictions)
        pred_temp = predict(df_temp)[0]
        pred_dewpt = predict(df_dewpt, dewpt=True)[0]

        predictions.append({
            'date_value': hour,
            'predicted': {
                'temp': round(pred_temp, 1),
                'dew_point': round(pred_dewpt, 1)
            }
        })
        hour += 3600000
    return {'data': predictions}


def get_predictors(dewpt=False):
    predictors = list()
    if dewpt:
        with open('training_data/dewpt_final_features.txt', 'r') as f:
            predictors = [line.strip() for line in f]
    else:
        with open('training_data/temp_final_features_2.txt', 'r') as f:
            predictors = [line.strip() for line in f]
    return predictors