Raghuraman Krishnamoorthi 904949382e Ensure that histogram observers have zero-point of zero for post ReLU activations (#37107) ... Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/37107 Currently histogram observers relax both the min and max values of the activations for performance speedup reasons. This causes an issue for glow where there is a slow down if the zero-point is not zero for post ReLU activations. ghstack-source-id: 102768017 Test Plan: buck test caffe2/test:quantization -- 'test_histogram_observer_one_sided \(quantization\.test_quantization\.RecordHistogramObserverTest\)' --print-passing-details Differential Revision: D21187636 fbshipit-source-id: 8d616b9e9caf2979a26a215e99434f71025e3d8b		2020-04-24 20:57:34 -07:00
..
__init__.py	Ignore F401 in all __init__.py without putting noqa (#25823)	2019-10-23 15:28:13 -07:00
_numeric_suite.py	Revert D21045393: [PyTorch Numeric Suite] Add module level comparison	2020-04-24 07:03:04 -07:00
_quantize_script.py	Fix type annotations and make MyPy run on torch/ (#36584)	2020-04-22 14:17:08 -07:00
default_mappings.py	quantized layer norm: add to static quant (#36690)	2020-04-16 18:18:02 -07:00
fake_quantize.py	Per channel quantization performance improvement (#33772)	2020-02-26 10:19:25 -08:00
fuse_modules.py	[quant] Enable fusion for conv modules with bias (#36173)	2020-04-08 15:53:32 -07:00
observer.py	Ensure that histogram observers have zero-point of zero for post ReLU activations (#37107)	2020-04-24 20:57:34 -07:00
qconfig.py	[quant][graph] Add a new observer type for dynamic quantization (#35455)	2020-03-26 17:38:21 -07:00
quantize.py	Add the more fusion (conv3d and batchnorm)support in pytorch quantization flow (#33540)	2020-03-23 20:36:03 -07:00
stubs.py	Factored out the default mappings	2019-10-03 11:52:21 -07:00