How many decimal places for support in the result set of fpgrowth?

[hpconnox]

As far as I can say, the support value in the result set (fpgrowth) is returning a maximum of 6 decimal places. Is there any posibility to increase the number of decimal places (to 8 or 10).

[shenxiangzhuang]

TLDR

You can set the display.precision by pd.set_option("display.precision", 8) before you print the DataFrame and you will get what you need.

The details

There is no round operation in source code:

mlxtend/mlxtend/frequent_patterns/fpcommon.py

Lines 64 to 78 in 63a2655

	def generate_itemsets(generator, num_itemsets, colname_map):
	itemsets = []
	supports = []
	for sup, iset in generator:
	itemsets.append(frozenset(iset))
	supports.append(sup / num_itemsets)
	res_df = pd.DataFrame({"support": supports, "itemsets": itemsets})
	if colname_map is not None:
	res_df["itemsets"] = res_df["itemsets"].apply(
	lambda x: frozenset([colname_map[i] for i in x])
	)
	return res_df

So it's just the display setting of pandas, you check pandas doc here.