-
Notifications
You must be signed in to change notification settings - Fork 3.8k
Description
With v0.6 adding quantization support, I think it is good time to add documentation on our quantization story.
There have been many questions on the forum, some of which are listed at the bottom. I myself have recently become interested in the topic, but I'm having hard time digging through the forum, github issues, PRs etc.
It would be great if we could add an end to end quantization usage tutorial or design document to our developer guide.
Current state of quantization effort
Is there a complete quantization tutorial including the use of various quantization algorithms and the addition of operators?
Quantization Configuration Documentation?
How to choose different quantization methods?
Int8 tutorial request
Understanding the quantization passes
Which operators are quantized in TVM?
Does tvm support quantization