Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.8/site-packages/keras/src/datasets/cifar100.py: 38%
24 statements
« prev ^ index » next coverage.py v7.4.0, created at 2024-01-03 07:57 +0000
« prev ^ index » next coverage.py v7.4.0, created at 2024-01-03 07:57 +0000
1# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
2#
3# Licensed under the Apache License, Version 2.0 (the "License");
4# you may not use this file except in compliance with the License.
5# You may obtain a copy of the License at
6#
7# http://www.apache.org/licenses/LICENSE-2.0
8#
9# Unless required by applicable law or agreed to in writing, software
10# distributed under the License is distributed on an "AS IS" BASIS,
11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12# See the License for the specific language governing permissions and
13# limitations under the License.
14# ==============================================================================
15"""CIFAR100 small images classification dataset."""
17import os
19import numpy as np
21from keras.src import backend
22from keras.src.datasets.cifar import load_batch
23from keras.src.utils.data_utils import get_file
25# isort: off
26from tensorflow.python.util.tf_export import keras_export
29@keras_export("keras.datasets.cifar100.load_data")
30def load_data(label_mode="fine"):
31 """Loads the CIFAR100 dataset.
33 This is a dataset of 50,000 32x32 color training images and
34 10,000 test images, labeled over 100 fine-grained classes that are
35 grouped into 20 coarse-grained classes. See more info at the
36 [CIFAR homepage](https://www.cs.toronto.edu/~kriz/cifar.html).
38 Args:
39 label_mode: one of "fine", "coarse". If it is "fine" the category labels
40 are the fine-grained labels, if it is "coarse" the output labels are the
41 coarse-grained superclasses.
43 Returns:
44 Tuple of NumPy arrays: `(x_train, y_train), (x_test, y_test)`.
46 **x_train**: uint8 NumPy array of grayscale image data with shapes
47 `(50000, 32, 32, 3)`, containing the training data. Pixel values range
48 from 0 to 255.
50 **y_train**: uint8 NumPy array of labels (integers in range 0-99)
51 with shape `(50000, 1)` for the training data.
53 **x_test**: uint8 NumPy array of grayscale image data with shapes
54 `(10000, 32, 32, 3)`, containing the test data. Pixel values range
55 from 0 to 255.
57 **y_test**: uint8 NumPy array of labels (integers in range 0-99)
58 with shape `(10000, 1)` for the test data.
60 Example:
62 ```python
63 (x_train, y_train), (x_test, y_test) = keras.datasets.cifar100.load_data()
64 assert x_train.shape == (50000, 32, 32, 3)
65 assert x_test.shape == (10000, 32, 32, 3)
66 assert y_train.shape == (50000, 1)
67 assert y_test.shape == (10000, 1)
68 ```
69 """
70 if label_mode not in ["fine", "coarse"]:
71 raise ValueError(
72 '`label_mode` must be one of `"fine"`, `"coarse"`. '
73 f"Received: label_mode={label_mode}."
74 )
76 dirname = "cifar-100-python"
77 origin = "https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz"
78 path = get_file(
79 dirname,
80 origin=origin,
81 untar=True,
82 file_hash=( # noqa: E501
83 "85cd44d02ba6437773c5bbd22e183051d648de2e7d6b014e1ef29b855ba677a7"
84 ),
85 )
87 fpath = os.path.join(path, "train")
88 x_train, y_train = load_batch(fpath, label_key=label_mode + "_labels")
90 fpath = os.path.join(path, "test")
91 x_test, y_test = load_batch(fpath, label_key=label_mode + "_labels")
93 y_train = np.reshape(y_train, (len(y_train), 1))
94 y_test = np.reshape(y_test, (len(y_test), 1))
96 if backend.image_data_format() == "channels_last":
97 x_train = x_train.transpose(0, 2, 3, 1)
98 x_test = x_test.transpose(0, 2, 3, 1)
100 return (x_train, y_train), (x_test, y_test)