microsoft DeepSpeedExamples issues

microsoft / DeepSpeedExamples

Example models using DeepSpeed

Apache License 2.0

5.83k stars 987 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Consult the first phase.

#909 csxrzhang opened 2 days ago
0
an error with gradient checkpointing in DeepspeedChat step 3

#908 wangyuwen1999 opened 5 days ago
0
单机多卡进行RLHF在第三步中使用Qwen模型作Actor Model报错

#907 Dakai798 opened 1 week ago
0
DeepSpeed-Chat step-1 hanging for a long time

#906 lemon-little opened 1 week ago
0
Enable cpu/xpu support for the benchmarking suite

#905 louie-tsai opened 1 month ago
7
CPU OOM when inferencing Llama3-70B-Chinese-Chat

#904 GORGEOUSLCX opened 1 month ago
0
cannot pickle 'Stream' object

#903 teis-e opened 1 month ago
0
can not run the test-gpt.sh because of assertionError

#902 leachee99 opened 1 month ago
0
请问fastgen 是否支持长文本和序列并行推理

#901 AceCoder0 opened 1 month ago
0
Add --client-only arg to mii benchmark

#900 delock closed 3 weeks ago
0
Refactored LLM benchmark code

#899 mrwyattii closed 1 week ago
0
fix bug with queue.empty not being reliable

#898 mrwyattii closed 2 months ago
0
Update tokens_per_sec calculation to work w/ stream and non-stream cases

#897 lekurile closed 2 months ago
0
run-example.sh fails with urllib3.exceptions.ProtocolError: Response ended prematurely

#896 awan-10 opened 2 months ago
8
updating tokens per second to include the token count of generated tokens.

#895 guptha23 closed 2 months ago
0
[Error] AutoTune: `connect to host localhost port 22: Connection refused`

#894 wqw547243068 opened 2 months ago
0
How to use deepspeed for multi-node and multi-card task in slurm cluster

#893 dshwei opened 2 months ago
0
Does Zero-Inference support TP?

#892 preminstrel opened 2 months ago
11
extend max_prompt_length and input text for 128k evaluation

#891 HeyangQin opened 2 months ago
0
Deepspeed support finetune extra model with lora ?

#890 wanghongqu opened 2 months ago
1
不同机器上python环境变量路径不同，deepspeed启动后发现找不到其他机器的python环境，如何解决

#889 liqwertyu opened 2 months ago
0
when calculating actor loss, why the mask is "action_mask[:, start: ] "

#888 fancghit closed 2 months ago
0
The actor constantly generates ['</s>'] or ['<|endoftext|></s>'] after 200 steps in RLHF with hybrid engine disabled

#887 mousewu opened 2 months ago
0
About multiple-thread attention computation on CPU using zero-inference example.

#886 luckyq opened 2 months ago
0
Suggested GPU to run the demo code of step2_reward_model_finetuning (DeepSpeed-Chat)

#885 wenbozhangjs opened 3 months ago
0
[REQUEST] More fine-grained distributed strategies for RLHF training

#884 youshaox opened 3 months ago
0
The reward value did not increase.

#883 Sun-Shiqi opened 3 months ago
1
Fix response check in call_aml function

#882 HeyangQin closed 3 months ago
0
Update throughput-latency plot script

#881 lekurile closed 2 months ago
0
[Inference Benchmark] set `num_requests` based on `num_clients`

#880 mrwyattii closed 3 months ago
0
Confusion about Deepspeed Inference

#879 ZekaiGalaxy opened 3 months ago
1
`AttributeError: readonly attribute` while trying to run training/HelloDeepSpeed

#878 htjain opened 3 months ago
0
Benchmark mii stalled and crashed

#877 Albert-Zhao-2020 opened 3 months ago
0
[inference benchmark] update AML kwargs to match vLLM kwargs

#876 mrwyattii closed 3 months ago
0
Improve robustness of infernece AML benchmark

#875 HeyangQin closed 3 months ago
0
Fix AML benchmark E2E measurment

#874 mrwyattii closed 3 months ago
0
Add LoRA optimization to the SD training example

#873 PareesaMS opened 3 months ago
0
Replace deprecated transformers.deepspeed module

#872 HollowMan6 opened 3 months ago
0
Xiaoxia/fp v1

#871 xiaoxiawu-microsoft closed 3 months ago
0
Remove AML key from args dict when saving results

#870 lekurile closed 3 months ago
0
Inference Benchmark: Catch AML error response

#869 mrwyattii closed 4 months ago
0
Update Inference Benchmarking Scripts - Support AML

#868 lekurile closed 3 months ago
1
[Bug] DeepSpeed Inference Does not Work with LLaMA (Latest verison)

#867 allanj opened 4 months ago
3
[BUG in Stable Diffusion inference] There's an error on CUDAGraph when using deepspeed inference. How to fix it?

#866 foin6 opened 4 months ago
2
Extend FastGen benchmark to use AML endpoints

#865 mrwyattii closed 4 months ago
0
zero3 and enable hybrid engine are not suitable for llama2, how to solve it?

#864 terence1023 opened 4 months ago
2
<fill-mask>Modify codes so that different accelerators can be called according to specific device conditions

#863 foin6 closed 4 months ago
1
Fix path in human-eval example README

#862 lekurile closed 4 months ago
0
RLHF problems when using Qwen model

#861 128Ghe980 opened 4 months ago
1
Codellama finetune

#860 nani1149 opened 4 months ago
0