beernet.py

discrete_columns = [
    'hop 1 form',
    'hop 2 form',
    'hop 3 form',
    'hop 4 form',
    'hop 1 phase',
    'hop 2 phase',
    'hop 3 phase',
    'hop 4 phase',
    'hop 1 name',
    'hop 2 name',
    'hop 3 name',
    'hop 4 name',
    'fermentable 1 name',
    'fermentable 2 name',
    'fermentable 3 name',
    'fermentable 4 name',
    'fermentable 5 name',
    'yeast'
]
df['batch'] = df['batch'].astype(float)

imp = SimpleImputer(strategy='most_frequent', missing_values='-', keep_empty_features=True)
cols = df.columns

imputed = imp.fit_transform(df)

df = pd.DataFrame(imputed, columns=cols)
df['batch'] = df['batch'].astype(float)

df['fermentable 1 weight'] = df['fermentable 1 weight'].astype(float)
df['fermentable 2 weight'] = df['fermentable 2 weight'].astype(float)
df['fermentable 3 weight'] = df['fermentable 3 weight'].astype(float)
df['fermentable 4 weight'] = df['fermentable 4 weight'].astype(float)
df['fermentable 5 weight'] = df['fermentable 5 weight'].astype(float)

df['hop 1 weight'] = df['hop 1 weight'].astype(float)
df['hop 2 weight'] = df['hop 2 weight'].astype(float)
df['hop 3 weight'] = df['hop 3 weight'].astype(float)
df['hop 4 weight'] = df['hop 4 weight'].astype(float)

df['hop 1 time'] = df['hop 1 time'].astype(float)
df['hop 2 time'] = df['hop 2 time'].astype(float)
df['hop 3 time'] = df['hop 3 time'].astype(float)
df['hop 4 time'] = df['hop 4 time'].astype(float)

df['yeast'] = df['yeast'].astype(str)

ctgan = CTGAN(epochs=45, cuda=True, verbose=True)
ctgan.fit(df, discrete_columns)

# Create synthetic data
synthetic_data = ctgan.sample(20)
synthetic_data.to_csv('./synthetic_recipes_45_epochs.csv')
print(synthetic_data)