Skyrim Alchemy Optimizer

This Jupyter notebook can be used to maximize alchemy profitability using the ingredients you have on hand in The Elder Scrolls V: Skyrim.

It uses integer linear programming from scipy.optimize.milp to determine which potions to make, and in what quantities, to maximize total value. It needs a csv file of the ingredients you have with their quantities, and a csv file of all possible potions you could make with your ingredients. I used this helpful spreadsheet to create my csvs, which are available here as examples.

Python Setup

import numpy as np
import pandas as pd
from scipy.optimize import milp, Bounds, LinearConstraint

Read in Ingredients

ingredients = pd.read_csv('ingredients_have.csv')

Read in Recipes

recipes = pd.read_csv('recipes_can_make.csv')
recipes = recipes[recipes['Magnitude'] > 0]

Create recipe matrix A in Ax <= b

One row for each ingredient, one column for each potion. “1” indicates the ingredient is used in the potion.

A = pd.DataFrame(0, index=range(len(ingredients)),columns=range(len(recipes)))
for i in range(len(recipes)):
  if ingredients.iloc[ingredients["Ingredient"].str.find(recipes.loc[i, "Ingredient 1"]).idxmax()]["Quantity"] > 0:
    A.iloc[ingredients["Ingredient"].str.find(recipes.loc[i, "Ingredient 1"]).idxmax(), i] = 1
  if ingredients.iloc[ingredients["Ingredient"].str.find(recipes.loc[i, "Ingredient 2"]).idxmax()]["Quantity"] > 0:
    A.iloc[ingredients["Ingredient"].str.find(recipes.loc[i, "Ingredient 2"]).idxmax(), i] = 1
  if not pd.isnull(recipes.loc[i, "Ingredient 3"]):
    if ingredients.iloc[ingredients["Ingredient"].str.find(recipes.loc[i, "Ingredient 3"]).idxmax()]["Quantity"] > 0:
      A.iloc[ingredients["Ingredient"].str.find(recipes.loc[i, "Ingredient 3"]).idxmax(), i] = 1
A

Optimization Setup

f = np.array(-1 * recipes['Magnitude'], dtype=int)
b_max = np.array(ingredients['Quantity'], dtype=int)
x_lb = np.zeros(shape=len(recipes))
bounds = Bounds(lb=x_lb)
constraint = LinearConstraint(A, ub=b_max)
integrality = np.ones(shape=len(recipes), dtype=int)

Perform Optimization

res = milp(c=f, integrality=integrality, bounds=bounds, constraints=constraint)

Display Recommended Potions

total_magnitude = int(-res.fun)
num_potions = int(sum(res.x))
indices_to_make = np.nonzero(res.x > 0)
to_make_df = recipes.iloc[indices_to_make].copy()
to_make_df['QtyToMake'] = res.x[indices_to_make].astype(int)
to_make_df[['Magnitude','Type','Ingredient 1','Ingredient 2','Ingredient 3','MyPotionID','QtyToMake']].head(50)

print(f"To maximize magnitude and therefore value, create {num_potions} potions of the {len(to_make_df)} unique types listed above for a total magnitude of {total_magnitude}.")

View on GitHub