text-generation-webui/modules/callbacks.py

from queue import Queue
from threading import Thread

import torch
import transformers


# Copied from https://github.com/PygmalionAI/gradio-ui/
class _SentinelTokenStoppingCriteria(transformers.StoppingCriteria):

    def __init__(self, sentinel_token_ids: list[torch.LongTensor], starting_idx: int):
        transformers.StoppingCriteria.__init__(self)
        self.sentinel_token_ids = sentinel_token_ids
        self.starting_idx = starting_idx

    def __call__(self, input_ids: torch.LongTensor, _scores: torch.FloatTensor) -> bool:
        for sample in input_ids:
            trimmed_sample = sample[self.starting_idx:]

            for i in range(len(self.sentinel_token_ids)):
                # Can't unfold, output is still too tiny. Skip.
                if trimmed_sample.shape[-1] < self.sentinel_token_ids[i].shape[-1]:
                    continue
                for window in trimmed_sample.unfold(0, self.sentinel_token_ids[i].shape[-1], 1):
                    if torch.all(torch.eq(self.sentinel_token_ids[i], window)):
                        return True
        return False

class Stream(transformers.StoppingCriteria):
    def __init__(self, callback_func=None):
        self.callback_func = callback_func

    def __call__(self, input_ids, scores) -> bool:
        if self.callback_func is not None:
            self.callback_func(input_ids[0])
        return False

class Iteratorize:

    """
    Transforms a function that takes a callback
    into a lazy iterator (generator).
    """

    def __init__(self, func, kwargs={}, callback=None):
        self.mfunc=func
        self.c_callback=callback
        self.q = Queue()
        self.sentinel = object()
        self.kwargs = kwargs
        self.stop_now = False

        def _callback(val):
            if self.stop_now:
                raise ValueError
            self.q.put(val)

        def gentask():
            try:
                ret = self.mfunc(callback=_callback, **self.kwargs)
            except ValueError:
                pass
            clear_torch_cache()
            self.q.put(self.sentinel)
            if self.c_callback:
                self.c_callback(ret)

        self.thread = Thread(target=gentask)
        self.thread.start()

    def __iter__(self):
        return self

    def __next__(self):
        obj = self.q.get(True,None)
        if obj is self.sentinel:
            raise StopIteration
        else:
            return obj

    def __del__(self):
        clear_torch_cache()

    def __enter__(self):
        return self

    def __exit__(self, exc_type, exc_val, exc_tb):
        self.stop_now = True
        clear_torch_cache()

def clear_torch_cache():
    gc.collect()
    if not shared.args.cpu:
        torch.cuda.empty_cache()