aistudio@jupyter-12521213-9537736:~/Paddle/test/legacy_test$ python test_flash_attention.py
/opt/conda/envs/python35-paddle120-env/lib/python3.10/site-packages/paddle/utils/cpp_extension/extension_utils.py:718: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md
warnings.warn(warning_message)
W0930 08:54:10.102687 4045 gpu_resources.cc:114] Please NOTE: device: 0, GPU Compute Capability: 8.0, Driver API Version: 12.8, Runtime API Version: 12.6
/opt/conda/envs/python35-paddle120-env/lib/python3.10/site-packages/paddle/tensor/creation.py:1088: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach(), rather than paddle.to_tensor(sourceTensor).
return tensor(
/opt/conda/envs/python35-paddle120-env/lib/python3.10/site-packages/paddle/utils/decorator_utils.py:420: Warning:
Non compatible API. Please refer to https://www.paddlepaddle.org.cn/documentation/docs/en/develop/guides/model_convert/convert_from_pytorch/api_difference/torch/torch.max.html first.
warnings.warn(
/opt/conda/envs/python35-paddle120-env/lib/python3.10/subprocess.py:767: ResourceWarning: unclosed file <_io.TextIOWrapper name=17 encoding='UTF-8'>
_cleanup()
ResourceWarning: Enable tracemalloc to get the object allocation traceback
.I0930 08:54:10.467125 4045 pir_interpreter.cc:1524] New Executor is Running ...
I0930 08:54:10.467581 4045 pir_interpreter.cc:1547] pir interpreter is running by multi-thread mode ...
...Test case shape (2, 128, 8, 16) dtype float16 causal False
.Test unpadded case shape (2, 128, 8, 16) dtype float16 causal False
W0930 08:54:10.710698 4045 eager_utils.cc:3441] Paddle static graph(PIR) not support input out tensor for now!!!!!
/opt/conda/envs/python35-paddle120-env/lib/python3.10/site-packages/paddle/base/executor.py:1478: UserWarning: The value k is not found in program. It is not declared or is pruned.
warnings.warn(
/opt/conda/envs/python35-paddle120-env/lib/python3.10/site-packages/paddle/base/executor.py:1478: UserWarning: The value v is not found in program. It is not declared or is pruned.
warnings.warn(
.Test case shape (2, 128, 8, 16) dtype float16 causal False
.Test unpadded case shape (2, 128, 8, 16) dtype float16 causal False
.Test case shape (2, 256, 8, 16) dtype float16 causal False
.Test unpadded case shape (2, 256, 8, 16) dtype float16 causal False
.Test case shape (2, 512, 8, 16) dtype float16 causal True
.Test unpadded case shape (2, 512, 8, 16) dtype float16 causal True
.Test case shape (8, 1024, 16, 128) dtype float16 causal False
.Test unpadded case shape (8, 1024, 16, 128) dtype float16 causal False
.Test case shape (8, 1024, 16, 256) dtype float16 causal False
.Test unpadded case shape (8, 1024, 16, 256) dtype float16 causal False
..../opt/conda/envs/python35-paddle120-env/lib/python3.10/site-packages/paddle/utils/decorator_utils.py:420: Warning:
Non compatible API. Please refer to https://www.paddlepaddle.org.cn/documentation/docs/en/develop/guides/model_convert/convert_from_pytorch/api_difference/torch/torch.split.html first.
warnings.warn(
.................Test case shape (8, 1024, 16, 128) dtype float16 causal False
.Test unpadded case shape (8, 1024, 16, 128) dtype float16 causal False
.Test case shape (8, 1024, 16, 128) dtype float16 causal False
.Test unpadded case shape (8, 1024, 16, 128) dtype float16 causal False
...s
----------------------------------------------------------------------
Ran 43 tests in 66.735s
OK (skipped=1)