rnn

Main entry point

This is the main entry point. You can execute this file. See rnn.initConfig() for some arguments, or just run ./rnn.py --help. See Technological overview for a technical overview.

rnn.analyze_data(config)[source]
rnn.crnnGreeting(configFilename=None, commandLineOptions=None)[source]
rnn.executeMainTask()[source]
rnn.finalize()[source]
rnn.getCacheByteSizes()[source]
Return type:(int,int,int)

:returns cache size in bytes for (train,dev,eval)

rnn.init(configFilename=None, commandLineOptions=(), config_updates=None, extra_greeting=None)[source]
Parameters:
  • configFilename (str|None) –
  • commandLineOptions (tuple[str]|list[str]|None) –
  • config_updates (dict[str]|None) –
  • extra_greeting (str|None) –
rnn.initBackendEngine()[source]
rnn.initConfig(configFilename=None, commandLineOptions=())[source]

Initializes the global config.

rnn.initConfigJsonNetwork()[source]
rnn.initData()[source]

Initializes the globals train,dev,eval of type Dataset.

rnn.initDevices()[source]
Return type:list[Device]
rnn.initEngine(devices)[source]

Initializes global engine.

rnn.initLog()[source]
rnn.load_data(config, cache_byte_size, files_config_key, **kwargs)[source]
Parameters:
  • config (Config) –
  • cache_byte_size (int) –
  • files_config_key (str) – such as “train” or “dev”
  • kwargs – passed on to init_dataset() or init_dataset_via_str()
Return type:

(Dataset,int)

:returns the dataset, and the cache byte size left over if we cache the whole dataset.

rnn.main(argv)[source]
rnn.needData()[source]
rnn.printTaskProperties(devices=None)[source]