.. |
activation_checkpointing
|
d7ca3d8373
reduce setting global variables to reduce torch compile graph breaks (#6541)
|
1 周之前 |
checkpoint_engine
|
c5edc91ecb
change partititon_name to partition_name (#3700)
|
1 年之前 |
comm
|
11a62a0635
Add Compressedbackend for Onebit optimizers (#5473)
|
4 月之前 |
compression
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
data_pipeline
|
64defe65b7
Parallel map step for `DistributedDataAnalyzer` map-reduce (#5291)
|
6 月之前 |
fp16
|
11a62a0635
Add Compressedbackend for Onebit optimizers (#5473)
|
4 月之前 |
pipe
|
7622cd9e68
Use msgpack for p2p comm (#6547)
|
3 周之前 |
swap_tensor
|
b65ea50631
GDS Swapping Fix (#6386)
|
2 月之前 |
zero
|
65ab64481f
Add API for updating ZeRO gradients (#6590)
|
1 周之前 |
__init__.py
|
c56a4b9e0d
Improve universal checkpoint (#5289)
|
6 月之前 |
base_optimizer.py
|
c56a4b9e0d
Improve universal checkpoint (#5289)
|
6 月之前 |
bf16_optimizer.py
|
8fa6b50bfe
Revert "BF16 optimizer: Clear lp grads after updating hp grads in hook" (#6508)
|
1 月之前 |
compiler.py
|
2a0c0e3c27
Remove compile wrapper to simplify access to model attributes (#5581)
|
4 月之前 |
config.py
|
8b191d7ccf
Long sequence parallelism (Ulysses) integration with HuggingFace (#5774)
|
2 月之前 |
config_utils.py
|
0a4457cc48
Pydantic v2 migration (#5167)
|
2 月之前 |
constants.py
|
f74ea69abf
Improve DS logging control (#6602)
|
1 周之前 |
dataloader.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
eigenvalue.py
|
862aff37a5
Use `torch.nan_to_num` replace numpy wrapper one (#5877)
|
2 月之前 |
engine.py
|
5c4b97f109
apply fp16 autocast only to floating point values
|
1 周之前 |
hybrid_engine.py
|
645639bcf8
Rearrange inference OPS and stop using builder.load (#5490)
|
1 周之前 |
lr_schedules.py
|
85b7469ea0
Add first Step in LR Schedulers (#6597)
|
1 周之前 |
progressive_layer_drop.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
quantize.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
sparse_tensor.py
|
c84c28d23b
Support cpu tensors without direct device invocation (#3842)
|
9 月之前 |
state_dict_factory.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
utils.py
|
adec99121b
Add API to get devices of offload states (#6586)
|
1 周之前 |
weight_quantizer.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |