19. Appendix: Tools for Deep Learning

In this chapter, we will walk you through major tools for deep learning, from introducing Jupyter notebook in Section 19.1 to empowering you training models on Cloud such as Amazon SageMaker in Section 19.2, Amazon EC2 in Section 19.3 and Google Colab in Section 19.4. Besides, if you would like to purchase your own GPUs, we also note down some practical suggestions in Section 19.5. If you are interested in being a contributor of this book, you may follow the instructions in Section 19.6.

• 19.1. Using Jupyter
  • 19.1.1. Editing and Running the Code Locally
  • 19.1.2. Advanced Options
  • 19.1.3. Summary
  • 19.1.4. Exercises
• 19.2. Using Amazon SageMaker
  • 19.2.1. Registering and Logging In
  • 19.2.2. Creating a SageMaker Instance
  • 19.2.3. Running and Stopping an Instance
  • 19.2.4. Updating Notebooks
  • 19.2.5. Summary
  • 19.2.6. Exercises
• 19.3. Using AWS EC2 Instances
  • 19.3.1. Creating and Running an EC2 Instance
  • 19.3.2. Installing CUDA
  • 19.3.3. Installing MXNet and Downloading the D2L Notebooks
  • 19.3.4. Running Jupyter
  • 19.3.5. Closing Unused Instances
  • 19.3.6. Summary
  • 19.3.7. Exercises
• 19.4. Using Google Colab
  • 19.4.1. Summary
  • 19.4.2. Exercises
• 19.5. Selecting Servers and GPUs
  • 19.5.1. Selecting Servers
  • 19.5.2. Selecting GPUs
  • 19.5.3. Summary
• 19.6. Contributing to This Book
  • 19.6.1. Minor Text Changes
  • 19.6.2. Propose a Major Change
  • 19.6.3. Adding a New Section or a New Framework Implementation
  • 19.6.4. Submitting a Major Change
  • 19.6.5. Summary
  • 19.6.6. Exercises
• 19.7. d2l API Document
  • Accumulator
  • AddNorm
  • AdditiveAttention
  • Animator
  • AttentionDecoder
  • BERTEncoder
  • BERTModel
  • BPRLoss
  • BananasDataset
  • Benchmark
  • CTRDataset
  • Decoder
  • DotProductAttention
  • Encoder
  • EncoderBlock
  • EncoderDecoder
  • HingeLossbRec
  • MaskLM
  • MaskedSoftmaxCELoss
  • MultiHeadAttention
  • NextSentencePred
  • PositionWiseFFN
  • PositionalEncoding
  • RNNModel
  • RNNModelScratch
  • RandomGenerator
  • Residual
  • SNLIDataset
  • Seq2SeqEncoder
  • SeqDataLoader
  • Timer
  • TokenEmbedding
  • TransformerEncoder
  • VOCSegDataset
  • Vocab
  • accuracy()
  • annotate()
  • argmax()
  • assign_anchor_to_bbox()
  • astype()
  • batchify()
  • bbox_to_rect()
  • bleu()
  • box_center_to_corner()
  • box_corner_to_center()
  • box_iou()
  • build_array_nmt()
  • copyfile()
  • corr2d()
  • count_corpus()
  • download()
  • download_all()
  • download_extract()
  • evaluate_accuracy()
  • evaluate_accuracy_gpu()
  • evaluate_accuracy_gpus()
  • evaluate_loss()
  • evaluate_ranking()
  • get_centers_and_contexts()
  • get_data_ch11()
  • get_dataloader_workers()
  • get_fashion_mnist_labels()
  • get_negatives()
  • get_tokens_and_segments()
  • grad_clipping()
  • hit_and_auc()
  • linreg()
  • load_array()
  • load_corpus_time_machine()
  • load_data_bananas()
  • load_data_fashion_mnist()
  • load_data_imdb()
  • load_data_ml100k()
  • load_data_nmt()
  • load_data_ptb()
  • load_data_snli()
  • load_data_time_machine()
  • load_data_voc()
  • load_data_wiki()
  • masked_softmax()
  • multibox_detection()
  • multibox_prior()
  • multibox_target()
  • nms()
  • numpy()
  • offset_boxes()
  • offset_inverse()
  • plot()
  • predict_ch3()
  • predict_ch8()
  • predict_sentiment()
  • predict_seq2seq()
  • predict_snli()
  • preprocess_nmt()
  • read_csv_labels()
  • read_data_bananas()
  • read_data_ml100k()
  • read_data_nmt()
  • read_imdb()
  • read_ptb()
  • read_snli()
  • read_time_machine()
  • read_voc_images()
  • reduce_sum()
  • reorg_test()
  • reorg_train_valid()
  • reshape()
  • resnet18()
  • seq_data_iter_random()
  • seq_data_iter_sequential()
  • set_axes()
  • set_figsize()
  • sgd()
  • show_bboxes()
  • show_heatmaps()
  • show_images()
  • show_list_len_pair_hist()
  • show_trace_2d()
  • size()
  • split_and_load_ml100k()
  • split_batch()
  • split_batch_multi_inputs()
  • split_data_ml100k()
  • squared_loss()
  • subsample()
  • synthetic_data()
  • to()
  • tokenize()
  • tokenize_nmt()
  • train_2d()
  • train_batch_ch13()
  • train_ch11()
  • train_ch13()
  • train_ch3()
  • train_ch6()
  • train_ch8()
  • train_concise_ch11()
  • train_epoch_ch3()
  • train_epoch_ch8()
  • train_ranking()
  • train_recsys_rating()
  • train_seq2seq()
  • transpose()
  • transpose_output()
  • transpose_qkv()
  • truncate_pad()
  • try_all_gpus()
  • try_gpu()
  • update_D()
  • update_G()
  • use_svg_display()
  • voc_colormap2label()
  • voc_label_indices()
  • voc_rand_crop()
  • Accumulator
  • AddNorm
  • AdditiveAttention
  • Animator
  • AttentionDecoder
  • BERTEncoder
  • BERTModel
  • BananasDataset
  • Benchmark
  • Decoder
  • DotProductAttention
  • Encoder
  • EncoderBlock
  • EncoderDecoder
  • MaskLM
  • MaskedSoftmaxCELoss
  • MultiHeadAttention
  • NextSentencePred
  • PositionWiseFFN
  • PositionalEncoding
  • RNNModel
  • RNNModelScratch
  • RandomGenerator
  • Residual
  • SNLIDataset
  • Seq2SeqEncoder
  • SeqDataLoader
  • Timer
  • TokenEmbedding
  • TransformerEncoder
  • VOCSegDataset
  • Vocab
  • accuracy()
  • annotate()
  • argmax()
  • assign_anchor_to_bbox()
  • astype()
  • batchify()
  • bbox_to_rect()
  • bleu()
  • box_center_to_corner()
  • box_corner_to_center()
  • box_iou()
  • build_array_nmt()
  • copyfile()
  • corr2d()
  • count_corpus()
  • download()
  • download_all()
  • download_extract()
  • evaluate_accuracy()
  • evaluate_accuracy_gpu()
  • evaluate_loss()
  • get_centers_and_contexts()
  • get_data_ch11()
  • get_dataloader_workers()
  • get_fashion_mnist_labels()
  • get_negatives()
  • get_tokens_and_segments()
  • grad_clipping()
  • linreg()
  • load_array()
  • load_corpus_time_machine()
  • load_data_bananas()
  • load_data_fashion_mnist()
  • load_data_imdb()
  • load_data_nmt()
  • load_data_ptb()
  • load_data_snli()
  • load_data_time_machine()
  • load_data_voc()
  • load_data_wiki()
  • masked_softmax()
  • multibox_detection()
  • multibox_prior()
  • multibox_target()
  • nms()
  • numpy()
  • offset_boxes()
  • offset_inverse()
  • plot()
  • predict_ch3()
  • predict_ch8()
  • predict_sentiment()
  • predict_seq2seq()
  • predict_snli()
  • preprocess_nmt()
  • read_csv_labels()
  • read_data_bananas()
  • read_data_nmt()
  • read_imdb()
  • read_ptb()
  • read_snli()
  • read_time_machine()
  • read_voc_images()
  • reduce_sum()
  • reorg_test()
  • reorg_train_valid()
  • reshape()
  • resnet18()
  • seq_data_iter_random()
  • seq_data_iter_sequential()
  • sequence_mask()
  • set_axes()
  • set_figsize()
  • sgd()
  • show_bboxes()
  • show_heatmaps()
  • show_images()
  • show_list_len_pair_hist()
  • show_trace_2d()
  • size()
  • split_batch()
  • squared_loss()
  • subsample()
  • synthetic_data()
  • to()
  • tokenize()
  • tokenize_nmt()
  • train_2d()
  • train_batch_ch13()
  • train_ch11()
  • train_ch13()
  • train_ch3()
  • train_ch6()
  • train_ch8()
  • train_concise_ch11()
  • train_epoch_ch3()
  • train_epoch_ch8()
  • train_seq2seq()
  • transpose()
  • transpose_output()
  • transpose_qkv()
  • truncate_pad()
  • try_all_gpus()
  • try_gpu()
  • update_D()
  • update_G()
  • use_svg_display()
  • voc_colormap2label()
  • voc_label_indices()
  • voc_rand_crop()
  • Accumulator
  • AddNorm
  • AdditiveAttention
  • Animator
  • AttentionDecoder
  • Benchmark
  • Decoder
  • DotProductAttention
  • Encoder
  • EncoderBlock
  • EncoderDecoder
  • MaskedSoftmaxCELoss
  • MultiHeadAttention
  • PositionWiseFFN
  • PositionalEncoding
  • RNNModel
  • RNNModelScratch
  • Residual
  • Seq2SeqEncoder
  • SeqDataLoader
  • Timer
  • TrainCallback
  • TransformerEncoder
  • Updater
  • Vocab
  • accuracy()
  • annotate()
  • bbox_to_rect()
  • bleu()
  • box_center_to_corner()
  • box_corner_to_center()
  • build_array_nmt()
  • corr2d()
  • count_corpus()
  • download()
  • download_all()
  • download_extract()
  • evaluate_accuracy()
  • evaluate_loss()
  • get_data_ch11()
  • get_fashion_mnist_labels()
  • grad_clipping()
  • linreg()
  • load_array()
  • load_corpus_time_machine()
  • load_data_fashion_mnist()
  • load_data_nmt()
  • load_data_time_machine()
  • masked_softmax()
  • numpy()
  • plot()
  • predict_ch3()
  • predict_ch8()
  • predict_seq2seq()
  • preprocess_nmt()
  • read_data_nmt()
  • read_time_machine()
  • seq_data_iter_random()
  • seq_data_iter_sequential()
  • sequence_mask()
  • set_axes()
  • set_figsize()
  • sgd()
  • show_heatmaps()
  • show_images()
  • show_list_len_pair_hist()
  • show_trace_2d()
  • size()
  • squared_loss()
  • synthetic_data()
  • tokenize()
  • tokenize_nmt()
  • train_2d()
  • train_ch11()
  • train_ch3()
  • train_ch6()
  • train_ch8()
  • train_concise_ch11()
  • train_epoch_ch3()
  • train_epoch_ch8()
  • train_seq2seq()
  • transpose_output()
  • transpose_qkv()
  • truncate_pad()
  • try_all_gpus()
  • try_gpu()
  • update_D()
  • update_G()
  • use_svg_display()