Let's support more operators in circle-quantizer
Q8: q8 quantization logic
Q16: q16 quantization logic
Verifier: Verifies the type/granularity of the quantized Op
Operator | Q8 | Q16 | Verifier
-- | -- | -- | --
sigmoid | o | o | 聽o
softmax | o | o | 聽o
tanh | o | o | o聽
permute (transpose) | o | o | o
reshape | o | o | o
slice | o聽| o聽| o聽
argmax | o | o | o聽
Operator | Q8 | Q16 | Verifier | Assignee
-- | -- | -- | -- | --
floor | o | o | o | @dayo09 聽
div | o | o | o聽| @dayo09 聽
floor_div | o | o | o聽| @dayo09 聽
greater | o聽 | o聽 | o聽| @jinevening
greater_equal | o聽 | o聽 | o聽| @jinevening
logical_or | o聽 | o | o聽| @jinevening
not_equal | o | o | o聽| @jinevening
concat | o | o | o聽| @jinevening
pad | o | o | o聽| @jinevening
Operator | Q8 | Q16 | Verifier| Assignee
-- | -- | -- | -- | --
elu | o | o | o聽| @dayo09 聽
rsqrt | o | o | o聽| @dayo09
sqrt | o | o | o聽| @dayo09
pow | o | o | o聽| @dayo09 聽
resize_bilinear | o | o | o聽| @dayo09 聽
resize_nearestneighbor | o | o | 聽| @dayo09 聽
yuv_rgb_scale (FC+Reshape) | - | - | -聽| -
instance_normalize | o | o | o聽| @jinevening
local_response_normalize | 聽 | 聽 | 聽| need discussion
batch_to_space | o聽 | o聽 | o聽| @jinevening
depth_to_space | 聽o | o聽 | o聽| @jinevening
space_to_batch | o聽 | o聽 | o聽| @jinevening
space_to_depth | o聽 | o聽 | o聽| @jinevening
split | o | o | o聽| @jinevening
strided_slice | o聽 | o聽 | o聽| @jinevening
CC @lemmaa
@jinevening I will start with not_equal 馃槃
@jinevening It seems compiler/luci/pass/src/test/TestIOGraph.h is staled and moved to compiler/luci/testhelper. https://github.com/Samsung/ONE/pull/6335
May I replace it?
May I replace it?
Of course :)
May I replace it?
If you like please go ahead :)
It would be nice to add an issue for this task.
By the offline conversation with @jinevening
lwq, cwq) doesn't mean all the nodes in the graph are the attributes.lwqlwq: most of the node's quantization type are lwqcwqcwq: only the nodes supports cwq quantization type are cwqlwq: most of the node's quantization type are lwq(1) Positive Test
| Type | Granularity |
| ---- | ----------- |
| U8 | LayerWise |
| U8 | ChannelWise |
| S16 | ChannelWise |
(2) Negative Test
F32 model to Type by QuantizeWithMinMaxPassWrongTypethe Verifier throws exceptionType | Granularity | WrongType
-- | -- | --
U8 | LayerWise | S16
U8 | ChannelWise | S16
S16 | ChannelWise | S16
F32 model to Type by QuantizeWithMinMaxPassthe Verifier throws exception| Type | Granularity |
| ---- | ----------- |
| U8 | LayerWise |
| U8 | ChannelWise |
| S16 | ChannelWise |
record-minmax currently does not support bool-type input.
Real world model's input is hardly bool type. But in the viewpoint of CI, we will need bool-type input for testing.
I made a draft to support bool type input in record-minmax #6526.
All done. Thanks @dayo09 :)
Most helpful comment
All done. Thanks @dayo09 :)