mindformers.generation¶

Mindformers generation.

mindformers.generation.generation_config¶

Class that holds a configuration for a generation task. Args: > Parameters that control the length of the output.

`mindformers.generation.logits_process.LogitsProcessor`	Abstract base class for all logit processors that can be applied during generation.
`mindformers.generation.logits_process.LogitsWarper`	Abstract base class for all logit warpers that can be applied during generation with multinomial sampling.
`mindformers.generation.logits_process.LogitsProcessorList`	This class can be used to create a list of [LogitsProcessor] or [LogitsWarper] to subsequently process a scores input tensor.
`mindformers.generation.logits_process.RepetitionPenaltyLogitsProcessor`	[LogitsProcessor] enforcing an exponential penalty on repeated sequences.
`mindformers.generation.logits_process.LogitNormalization`	[LogitsWarper] and [LogitsProcessor] for normalizing the scores using log-softmax.
`mindformers.generation.logits_process.TemperatureLogitsWarper`	[LogitsWarper] for temperature (exponential scaling output probability distribution).
`mindformers.generation.logits_process.TopKLogitsWarper`	[LogitsWarper] that performs top-k, i.e. restricting to the k highest probability elements.
`mindformers.generation.logits_process.TopPLogitsWarper`	[LogitsWarper] that performs top-p, i.e. restricting to top tokens summing to prob_cut_off <= prob_cut_off.

`mindformers.generation.streamers.BaseStreamer`	Base class from which .generate() streamers should inherit.
`mindformers.generation.streamers.TextStreamer`	Simple text streamer that prints the token(s) to stdout as soon as entire words are formed.
`mindformers.generation.streamers.TextIteratorStreamer`	Streamer that stores print-ready text in a queue, to be used by a downstream application as an iterator.

Generator For the nlp models