26 lines
778 B
Python
26 lines
778 B
Python
|
import pandas as pd
|
||
|
|
|
||
|
|
train_in = pd.read_csv(
|
||
|
|
"../train/in.tsv",
|
||
|
|
header=None,
|
||
|
|
names=["id", "nazwa", "typ_zbioru", "rok", "miesiąc"],
|
||
|
|
sep="\t",
|
||
|
|
)
|
||
|
|
train_exp = pd.read_csv("../train/expected.tsv", header=None, names=["opad"], sep="\t")
|
||
|
|
|
||
|
|
train = train_in.copy()
|
||
|
|
train["opad"] = train_exp
|
||
|
|
|
||
|
|
for dataset in ("dev-0", "test-A"):
|
||
|
|
with open(f"../{dataset}/out.tsv", "w") as f_out, open(
|
||
|
|
f"../{dataset}/in.tsv", "r"
|
||
|
|
) as f_in:
|
||
|
|
for line in f_in:
|
||
|
|
id, _, _, rok, miesiac = line.rstrip().split("\t")
|
||
|
|
rok_2010 = train[
|
||
|
|
(train["id"] == int(id))
|
||
|
|
& (train["rok"] == 2010)
|
||
|
|
& (train["miesiąc"] == int(miesiac))
|
||
|
|
]["opad"].item()
|
||
|
|
f_out.write(str(rok_2010) + "\n")
|