Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
160 px This dataset consists of the Imagenette dataset {size} variant.
DatasetBuilder:tfds.image.imagenette.Imagenette
imagenette is configured with tfds.image.imagenette.ImagenetteConfig and has
the following configurations predefined (defaults to the first one):
full-size(v0.1.0) (Size: 1.45 GiB): Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
160 px This dataset consists of the Imagenette dataset full-size variant.
320px(v0.1.0) (Size: 325.48 MiB): Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
160 px This dataset consists of the Imagenette dataset 320px variant.
160px(v0.1.0) (Size: 94.18 MiB): Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
- 160 px This dataset consists of the Imagenette dataset 160px variant.
imagenette/full-size
Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
- 160 px This dataset consists of the Imagenette dataset full-size variant.
Versions:
0.1.0(default):
Statistics
| Split | Examples |
|---|---|
| ALL | 13,394 |
| TRAIN | 12,894 |
| VALIDATION | 500 |
Features
FeaturesDict({
'image': Image(shape=(None, None, 3), dtype=tf.uint8),
'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=10),
})
Homepage
Supervised keys (for as_supervised=True)
('image', 'label')
imagenette/320px
Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
- 160 px This dataset consists of the Imagenette dataset 320px variant.
Versions:
0.1.0(default):
Statistics
| Split | Examples |
|---|---|
| ALL | 13,394 |
| TRAIN | 12,894 |
| VALIDATION | 500 |
Features
FeaturesDict({
'image': Image(shape=(None, None, 3), dtype=tf.uint8),
'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=10),
})
Homepage
Supervised keys (for as_supervised=True)
('image', 'label')
imagenette/160px
Imagenette is a subset of 10 easily classified classes from the Imagenet dataset. It was originally prepared by Jeremy Howard of FastAI. The objective behind putting together a small version of the Imagenet dataset was mainly because running new ideas/algorithms/experiments on the whole Imagenet take a lot of time.
This version of the dataset allows researchers/practitioners to quickly try out ideas and share with others. The dataset comes in three variants:
- Full size
- 320 px
- 160 px This dataset consists of the Imagenette dataset 160px variant.
Versions:
0.1.0(default):
Statistics
| Split | Examples |
|---|---|
| ALL | 13,394 |
| TRAIN | 12,894 |
| VALIDATION | 500 |
Features
FeaturesDict({
'image': Image(shape=(None, None, 3), dtype=tf.uint8),
'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=10),
})
Homepage
Supervised keys (for as_supervised=True)
('image', 'label')
Citation