-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathSpaceshipDataset.py
81 lines (65 loc) · 3.82 KB
/
SpaceshipDataset.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
import numpy as np
import torch
from torch.utils.data import Dataset
import polars as pl
class SpaceshipDataset(Dataset):
def __init__(self, file_path, transform=None, target_transform=None):
# Set labels based on if the passenger was transported
self.labels = pl.read_csv(file_path).get_column("Transported").apply(lambda s: [1.0] if s else [0.0])
# Import dataset and remove 'Transported' column (used for labels) and 'Name,' because it is not relevant
features_tmp = pl.read_csv(file_path).drop("Transported").drop("Name")
features_tmp = (features_tmp.drop("PassengerId").with_columns(
features_tmp.get_column("PassengerId").apply(lambda s: int(s.split("_")[0])).alias("GroupId"),
features_tmp.get_column("PassengerId").apply(lambda s: int(s.split("_")[1])).alias("IntraGroupId"),
# One hot encode
*features_tmp.get_column("Destination").to_dummies().get_columns(),
*features_tmp.get_column("HomePlanet").to_dummies().get_columns(),
*features_tmp.get_column("Cabin").apply(
lambda s: "ABCDEFGT".index(s.split("/")[0])
).alias("RoomDeck").to_dummies().get_columns(),
features_tmp.get_column("Cabin").apply(lambda s: int(s.split("/")[1])).alias("RoomNumber"),
# One hot encode
*features_tmp.get_column("Cabin").apply(lambda s: s.split("/")[2]).alias("RoomSide").to_dummies().get_columns(),
*features_tmp.get_column("CryoSleep").to_dummies().get_columns(),
*features_tmp.get_column("VIP").to_dummies().get_columns(),
).drop("Cabin").drop("Destination").drop("Destination_null").drop("HomePlanet").drop("HomePlanet_null")
.drop("RoomDeck_null").drop("RoomSide_null").drop("CryoSleep").drop("CryoSleep_null")
.drop("VIP").drop("VIP_null"))
# Fill any remaining null values with 0
features_tmp = features_tmp.fill_null(strategy="zero")
self.features = features_tmp
def __len__(self):
return len(self.labels)
def __getitem__(self, idx):
label = torch.tensor([self.labels[idx]], dtype=torch.float32)
feature = torch.from_numpy(np.float32(self.features[idx].to_numpy()))
return feature, label
class SubmissionDataset(Dataset):
def __init__(self, file_path, transform=None, target_transform=None):
# Set labels based on if the passenger was transported
self.labels = pl.read_csv(file_path)
# Import dataset and remove 'Transported' column (used for labels) and 'Name,' because it is not relevant
features_tmp = pl.read_csv(file_path).drop("Name")
features_tmp = features_tmp.drop("PassengerId").with_columns(
features_tmp.get_column("PassengerId").apply(lambda s: int(s.split("_")[0])).alias("GroupId"),
features_tmp.get_column("PassengerId").apply(lambda s: int(s.split("_")[1])).alias("IntraGroupId"),
# One hot encode
*features_tmp.get_column("Destination").to_dummies().get_columns(),
*features_tmp.get_column("HomePlanet").to_dummies().get_columns(),
*features_tmp.get_column("Cabin").apply(
lambda s: "ABCDEFGT".index(s.split("/")[0])
).alias("RoomDeck").to_dummies().get_columns(),
features_tmp.get_column("Cabin").apply(lambda s: int(s.split("/")[1])).alias("RoomNumber"),
# One hot encode
*features_tmp.get_column("Cabin").apply(lambda s: s.split("/")[2]).alias("RoomSide").to_dummies().get_columns(),
*features_tmp.get_column("CryoSleep").to_dummies().get_columns(),
*features_tmp.get_column("VIP").to_dummies().get_columns(),
).drop("Cabin").drop("Destination").drop("Destination_null").drop("HomePlanet").drop("HomePlanet_null").drop("RoomDeck_null").drop("RoomSide_null").drop("CryoSleep").drop("CryoSleep_null").drop("VIP").drop("VIP_null")
# Fill any remaining null values with 0
features_tmp = features_tmp.fill_null(strategy="zero")
self.features = features_tmp
def __len__(self):
return len(self.features.get_column("GroupId"))
def __getitem__(self, idx):
feature = torch.from_numpy(np.float32(self.features[idx].to_numpy()))
return feature