Open
Description
Hello there! I want to determine the memory consumption when building a brotli.Compressor object utilizing the highest dictionary level (i.e. 11) and an empty dictionary (i.e. b"") for a float16 numpy array of 2k elements.
First things first, i'm approaching this via memory-profiler. An occuring problem is that, when i'm building the brotli.Compressor given the above attributes, i get an error that compressor obeject has no atribute process.
When i'm using compressed = brotli.compress(data,quality,mode,dict), the memory consumption seems to be significantly low.
Therefore:
- Do you know an approximation of the memory used building the above described object?
- How can i resolve the error compressor object has no attribute process?
Metadata
Assignees
Labels
No labels