Function mindspore::dataset::WikiText

Function Documentation

inline std::shared_ptr<WikiTextDataset> mindspore::dataset::WikiText(const std::string &dataset_dir, const std::string &usage = "all", int64_t num_samples = 0, ShuffleMode shuffle = ShuffleMode::kGlobal, int32_t num_shards = 1, int32_t shard_id = 0, const std::shared_ptr<DatasetCache> &cache = nullptr)

Function to create a WikiText Dataset.

Note

The generated dataset has one column [‘text’].

Parameters

  • dataset_dir – [in] Path to the root directory that contains the dataset.

  • usage – [in] One of “all”, “train” , ‘valid’ or “test” (default = “all”).

  • num_samples – [in] The number of samples to be included in the dataset (Default = 0, means all samples).

  • shuffle – [in] The mode for shuffling data every epoch (Default=ShuffleMode.kGlobal). Can be any of: ShuffleMode.kFalse - No shuffling is performed. ShuffleMode.kFiles - Shuffle files only. ShuffleMode.kGlobal - Shuffle both the files and samples.

  • num_shards – [in] Number of shards that the dataset should be divided into (Default = 1).

  • shard_id – [in] The shard ID within num_shards. This argument should be specified only when num_shards is also specified (Default = 0).

  • cache – [in] Tensor cache to use (default=nullptr, which means no cache is used).

Returns

Shared pointer to the WikiTextDataset.

样例
/* Define dataset path and MindData object */
std::string folder_path = "/path/to/wiki_dataset_directory";
std::shared_ptr<Dataset> ds = WikiText(folder_path, "all");

/* Create iterator to read dataset */
std::shared_ptr<Iterator> iter = ds->CreateIterator();
std::unordered_map<std::string, mindspore::MSTensor> row;
iter->GetNextRow(&row);

/* Note: In WikiText dataset, each dictionary has key "text" */
auto text = row["text"];