Datasets:
mickylan2367
commited on
Commit
•
0bbb43a
1
Parent(s):
591f58b
Update README (#3)
Browse files- Update README (12144fd7ab4ed55436aa315f444729e47d1ffa6a)
README.md
CHANGED
@@ -1,11 +1,11 @@
|
|
1 |
---
|
2 |
-
task_categories:
|
3 |
-
- text-generation
|
4 |
license: cc-by-sa-4.0
|
5 |
language:
|
6 |
- en
|
7 |
tags:
|
8 |
- music
|
|
|
|
|
9 |
---
|
10 |
# Google/Music-Capsの音声データをスペクトログラム化したデータ。
|
11 |
|
@@ -26,13 +26,14 @@ data = data["train"]
|
|
26 |
```
|
27 |
|
28 |
### 1: データローダーへ
|
29 |
-
* まだテストデータと検証データは用意していないので、コメントアウトしています
|
30 |
* こんな感じの関数で、データローダーにできます。
|
31 |
```py
|
32 |
from torchvision import transforms
|
33 |
from torch.utils.data import DataLoader
|
34 |
BATCH_SIZE = ??? # 自分で設定
|
35 |
IMAGE_SIZE = ???
|
|
|
|
|
36 |
|
37 |
def load_datasets():
|
38 |
data_transforms = [
|
@@ -42,26 +43,24 @@ def load_datasets():
|
|
42 |
]
|
43 |
data_transform = transforms.Compose(data_transforms)
|
44 |
|
45 |
-
|
46 |
-
|
47 |
-
|
|
|
48 |
|
49 |
for idx in range(len(train["image"])):
|
50 |
train["image"][idx] = data_transform(train["image"][idx])
|
51 |
-
|
52 |
-
# validation["image"][idx] = data_transform(validation["image"][idx])
|
53 |
|
54 |
train = Dataset.from_dict(train)
|
55 |
-
# test = Dataset.from_dict(test)
|
56 |
-
# validation = Dataset.from_dict(validation)
|
57 |
-
|
58 |
train = train.with_format("torch") # リスト型回避
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
#
|
63 |
-
|
64 |
-
|
|
|
65 |
|
66 |
```
|
67 |
|
|
|
1 |
---
|
|
|
|
|
2 |
license: cc-by-sa-4.0
|
3 |
language:
|
4 |
- en
|
5 |
tags:
|
6 |
- music
|
7 |
+
size_categories:
|
8 |
+
- 1K<n<10K
|
9 |
---
|
10 |
# Google/Music-Capsの音声データをスペクトログラム化したデータ。
|
11 |
|
|
|
26 |
```
|
27 |
|
28 |
### 1: データローダーへ
|
|
|
29 |
* こんな感じの関数で、データローダーにできます。
|
30 |
```py
|
31 |
from torchvision import transforms
|
32 |
from torch.utils.data import DataLoader
|
33 |
BATCH_SIZE = ??? # 自分で設定
|
34 |
IMAGE_SIZE = ???
|
35 |
+
TRAIN_SIZE = ??? # 訓練に使用したいデータセット数
|
36 |
+
TEST_SIZE = ??? # テストに使用したいデータセット数
|
37 |
|
38 |
def load_datasets():
|
39 |
data_transforms = [
|
|
|
43 |
]
|
44 |
data_transform = transforms.Compose(data_transforms)
|
45 |
|
46 |
+
data = load_dataset("mickylan2367/spectrogram")
|
47 |
+
data = data["train"]
|
48 |
+
train = data[slice(0, TRAIN_SIZE, None)]
|
49 |
+
test = data[slice(TRAIN_SIZE, TRAIN_SIZE + TEST_SIZE, 0)]
|
50 |
|
51 |
for idx in range(len(train["image"])):
|
52 |
train["image"][idx] = data_transform(train["image"][idx])
|
53 |
+
test["image"][idx] = data_transform(test["image"][idx])
|
|
|
54 |
|
55 |
train = Dataset.from_dict(train)
|
|
|
|
|
|
|
56 |
train = train.with_format("torch") # リスト型回避
|
57 |
+
test = Dataset.from_dict(train)
|
58 |
+
test = test.with_format("torch") # リスト型回避
|
59 |
+
|
60 |
+
# or
|
61 |
+
train_loader = DataLoader(train, batch_size=BATCH_SIZE, shuffle=True, drop_last=True)
|
62 |
+
test_loader = DataLoader(test, batch_size=BATCH_SIZE, shuffle=True, drop_last=True)
|
63 |
+
return train_loader, test_loader
|
64 |
|
65 |
```
|
66 |
|