batche | Batch cache decorator

batche is a Python library. batche has no bugs, it has no vulnerabilities, it has a Permissive License and it has low support. However batche build file is not available. You can download it from GitHub.

batche provides a Python decorator analogous to lru_cache but for functions that transform list objects. Where lru_cache caches the results of a function call for a given set of arguments naively - for instance treating the batch inputs (1, 2, 3) and (1, 3, 2) as entirely different. batche introduces cache_batch_variable to cache on a per-item basis for a given batch variable. It helps in reducing computation by caching the outputs of each previously processed input in the batch. The cache is maintained using a dictionary or an LRU (Least Recently Used) Cache, based on the maxsize parameter. If maxsize is not provided, the cache will be a simple dictionary. This is useful when you have a costly function that takes a batch of items as input and returns a list of predictions. If you call this function multiple times with overlapping batches, the results for the overlapping examples will be fetched from the cache, improving the performance.