Code
Hub
Workspaces
Connect
Indexed graphs
Engine
MCP
copy
hub
/
github.com/deepspeedai/DeepSpeed
/ ByteTensor
Method
ByteTensor
accelerator/hpu_accelerator.py:211–212 ·
view source on GitHub ↗
(self)
Source
from the content-addressed store, hash-verified
209
210
@property
211
def
ByteTensor(self):
212
return
functools.partial(torch.tensor, dtype=torch.uint8, device=
'hpu'
)
213
214
@property
215
def
DoubleTensor(self):
Callers
9
_get_test_write_file_and_device_buffer
Function · 0.45
_get_test_write_file_and_unpinned_tensor
Function · 0.45
_get_test_write_file_and_pinned_tensor
Function · 0.45
test_offset_write
Method · 0.45
has_overflow
Method · 0.45
_checkpoint_tag_validation
Method · 0.45
has_overflow
Method · 0.45
has_overflow
Method · 0.45
send_obj
Function · 0.45
Calls
no outgoing calls
Tested by
4
_get_test_write_file_and_device_buffer
Function · 0.36
_get_test_write_file_and_unpinned_tensor
Function · 0.36
_get_test_write_file_and_pinned_tensor
Function · 0.36
test_offset_write
Method · 0.36