vasiliy 2d01f87737 Enable torch.empty for float8 dtypes + deterministic mode + cpu (#128744) ... Summary: Enables creating empty float8 tensors for: * cuda when `torch.use_deterministic_algorithms` is set to True * cpu for all settings of `torch.use_deterministic_algorithms` Context for NaN values of float8_e4m3fn and float8_e5m2: https://arxiv.org/pdf/2209.05433, Section 3, Table 1 Context for NaN values of float8_e4m3fnuz and float8_e5m2fnuz: https://arxiv.org/pdf/2206.02915, Section 3.2, "instead of reserving one exponent field to represent Inf and NaN, we reserve only a single codeword (corresponding to negative zero)" Test Plan: ``` python test/test_quantization.py -k test_empty ``` Reviewers: Subscribers: Tasks: Tags: Fixes https://github.com/pytorch/pytorch/issues/128733 Pull Request resolved: https://github.com/pytorch/pytorch/pull/128744 Approved by: https://github.com/malfet, https://github.com/drisspg		2024-06-15 02:05:30 +00:00
..
experimental	Enable torch.empty for float8 dtypes + deterministic mode + cpu (#128744)	2024-06-15 02:05:30 +00:00
__init__.py
test_backend_config.py
test_docs.py
test_quantized_functional.py
test_quantized_module.py	Add testing and fix weights_only load for quantized types and nn.Parameters with python attrs (#124330)	2024-04-23 04:13:26 +00:00
test_quantized_op.py	Revert "[cuDNN][Quantization] Don't print when plan finalization fails in cuDNN quantization backend (#128177)"	2024-06-12 02:20:15 +00:00
test_quantized_tensor.py	[quant] Make per_group and per_token quant match torch.fake_quantize (#125781)	2024-05-14 18:18:54 +00:00
test_top_level_apis.py
test_utils.py
test_workflow_module.py	Add testing and fix weights_only load for quantized types and nn.Parameters with python attrs (#124330)	2024-04-23 04:13:26 +00:00
test_workflow_ops.py	Add testing and fix weights_only load for quantized types and nn.Parameters with python attrs (#124330)	2024-04-23 04:13:26 +00:00