What is the default dataset available in Keras?


  • 0
  • 0

๐Ÿš€ Hey TensorFlow enthusiasts! ๐Ÿค–

Did you know that tf.keras.datasets has some awesome built-in datasets to kickstart yourย #MachineLearningย journey? ๐ŸŽ“๐ŸŒ Here’s a quick run-through:

1. CIFAR-10 and CIFAR-100 ๐ŸŽจ: These are popular image datasets containing 60,000 32×32 color images in 10 and 100 different classes, respectively. Great for image classification tasks! ๐Ÿ–ผ๏ธ๐Ÿค–

tf.keras.datasets.cifar10.load_data()
tf.keras.datasets.cifar100.load_data()

2. MNIST and Fashion MNIST ๐Ÿ‘•๐Ÿ”ข: Classic datasets for handwritten digit (0-9) recognition and clothing item classification. Both contain 70,000 28×28 grayscale images. Perfect for beginners! ๐Ÿš€

tf.keras.datasets.fashion_mnist.load_data()

3. IMDB Movie Reviews ๐ŸŽฌ๐Ÿ“: This dataset has 25,000 highly polar movie reviews for sentiment analysis. Test your NLP skills and classify positive or negative reviews! ๐Ÿ’ฌ๐Ÿง 

tf.keras.datasets.imdb.load_data()

4. Reuters News Topics ๐Ÿ“ฐ: A collection of 11,228 newswires from Reuters, labeled with 46 topics. Ideal for multi-class text classification tasks. ๐ŸŒ๐Ÿ“š

tf.keras.datasets.reuters.load_data()

Dive into these datasets and level up yourย #DeepLearningย game! ๐Ÿš€๐Ÿ‘ฉโ€๐Ÿ’ป๐Ÿ‘จโ€๐Ÿ’ป Happy coding! ๐Ÿ’ป๐Ÿ˜ƒ

๐Ÿ’ฏ Feel free to follow me here:ย Shrijayan Rajendranย ๐Ÿ‘จโ€๐Ÿ”ฌ

Leave a Reply

Your email address will not be published. Required fields are marked *