Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / deepjavalibrary/djl-serving issues and pull requests
#1123 - Upgrade flash attention v2 version to 2.3.0
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1122 - Integrate with seq_scheduler wheel
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
#1121 - Add dependency on stop-runners for lora correctness test
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
#1120 - [fix] Device_and_search_config_issue
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
#1119 - [Docker] upgrade dependencies version
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1118 - Add support for stopwords in huggingface handler
Pull Request -
State: closed - Opened by ydm-amazon about 1 year ago
- 5 comments
#1117 - [INF2][Handler] fix none type check
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1116 - Handling requests concurrently: AWS Sagemaker endpoint
Issue -
State: closed - Opened by sneha5gsm about 1 year ago
- 3 comments
#1115 - [Handler] add neuron int8 quantization
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1114 - Unmerged lora correctness test
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
- 1 comment
#1113 - [RollingBatch] add active requests and pending requests for skip tokens
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1112 - [vLLM] capture max_rolling_batch settting issues
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1111 - add flash2 support for huggingface accelerate
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1110 - Add unmerged lora integration test
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
- 1 comment
#1109 - Creates auto increment ID for models
Pull Request -
State: closed - Opened by zachgk about 1 year ago
#1108 - Upgrade lmi_dist
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1107 - [INF2][Handler] added optimization level per Neuron instruction
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1106 - Does deepspeed engine support Llama2 question-answering task?
Issue -
State: open - Opened by YunTaoYoung about 1 year ago
- 1 comment
Labels: bug
#1105 - fix deepspeed bugs and have better logging
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1104 - Fix typo
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
#1103 - Use custom peft wheel
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
#1102 - [CI] fix the inf2 container build failure
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1101 - [feature] Test llama-7b-gptq on scheduler_rolling_batch
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
- 3 comments
#1100 - Cleans some unused pieces of PyProcess
Pull Request -
State: closed - Opened by zachgk about 1 year ago
#1099 - Fix setting adapters arg
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
#1098 - fix some bugs in handler
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1097 - add no code testing for rollingbatch
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1096 - Allow DJLServing to take override engine name
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1095 - Removes quick abort of python reader threads
Pull Request -
State: closed - Opened by zachgk about 1 year ago
- 4 comments
#1094 - Creates adapters by directory
Pull Request -
State: closed - Opened by zachgk about 1 year ago
- 2 comments
#1093 - [vLLM] add pyarrow dependency
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1092 - tensor parallelism across multiple GPU's
Issue -
State: closed - Opened by samanthvishwas about 1 year ago
- 2 comments
#1091 - [INF2] grant write permission
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1090 - [INF2] fix some bugs and remove old tests
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1089 - [Neuron] upgrade to Neuron 2.14.0 SDK
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1088 - Add unmerged lora support in HF handler
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
- 4 comments
#1087 - [INF2] don't install linux headers
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1086 - [feature] Enable model sharding on seq_scheduler tested on gpt_neox_20B
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
#1085 - [demo] PR
Pull Request -
State: closed - Opened by tosterberg about 1 year ago
#1084 - How to allow parallel requests
Issue -
State: closed - Opened by thoth42 about 1 year ago
- 2 comments
#1083 - [serving] Fixes log rotation issue
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1082 - Adds adapter support
Pull Request -
State: closed - Opened by zachgk about 1 year ago
#1081 - [handler] fix a few issues
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1080 - Fixes API responses
Pull Request -
State: closed - Opened by zachgk about 1 year ago
#1079 - Fix neuron cores auto-scaling
Pull Request -
State: open - Opened by fafriat about 1 year ago
- 18 comments
#1078 - add Neuron RollingBatch implementation
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1077 - [Handler] formalize all engines with same settings
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1076 - [Handler] add dynamic batching to transformers neuronx
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1075 - What backend does djl-serving default to for rolling_batch?
Issue -
State: open - Opened by yudhiesh about 1 year ago
- 3 comments
Labels: bug
#1074 - [fix] Fix device map
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
#1073 - [docker] Change default OMP_NUM_THREADS back to 1 for GPU
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
- 4 comments
#1072 - [CI] add vllm tests
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
- 1 comment
#1071 - [CherryPick] use tag for test and not hardcode
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1070 - [RollingBatch][CI] use tag for test and not hardcode
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1069 - Revert "Upgrade lmi_dist"
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1068 - Revert "Upgrade lmi_dist"
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1067 - [serving] Fixes wrong device mapping for non-tp mode
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1066 - update lmi_dist package to 0.9.4
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1065 - [Handler] fix device mapping issues
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1064 - [Handler] make huggingface streamer default
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1063 - [ci] Fixes gradle deprecation warnings
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1062 - [ci] Fixes PMD warning
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1061 - [serving] Make sure extracting jni from jar file in deps folder
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1060 - Upgrade lmi_dist
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1059 - Upgrade lmi_dist
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1058 - [INF2] disable checker for saved model
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1057 - [INF2] Disable checker for load saved model
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1056 - add revision to handler
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1055 - [fix] Device and format and implementation optimization
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
#1054 - [0.23.0 cherry pick] #1034 #1035 #1036 #1041 vLLM changes
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
- 1 comment
#1053 - [Cherrypick 0.23.0][python] Avoid holding the lock while running inference (#1045)
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1052 - [docker] Upgrades to inf2 2.13.2 version
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1050 - [RollingBatch] create request simulator to batch
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1049 - [DeepSpeed] upgrade dependencies
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1048 - Revert "Change rolling batch generations order (#1046)"
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1047 - add data collection and some inf2 bug fixes
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1046 - Change rolling batch generations order
Pull Request -
State: closed - Opened by xyang16 about 1 year ago
#1045 - [python] Avoid holding the lock while running inference
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
- 1 comment
#1044 - update inf2 dependencies to 2.13.1
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1043 - [fix] Two major bugs in rolling_batch_scheduler
Pull Request -
State: closed - Opened by KexinFeng about 1 year ago
- 1 comment
#1042 - How to use inference of multiple models
Issue -
State: open - Opened by faquir-sun about 1 year ago
- 5 comments
#1041 - [rollingbatch] add standalone script to run
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1040 - [docker] add version labels for sagemaker
Pull Request -
State: closed - Opened by tosterberg about 1 year ago
- 2 comments
#1039 - [docs] Document Python engine alias
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1038 - Fixes OOM checker bug
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1037 - [python] Fixes tokenizer bug when using hugging pipeline
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1036 - [VLLM] add option to set batched tokens
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1035 - [VLLM] use more complex logic to ensure all result are captured
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1034 - fix logging bug in vllm
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1033 - Fixed vllm rolling batcher bugs
Pull Request -
State: closed - Opened by lanking520 about 1 year ago
#1032 - [docs] Updates TENSOR_PARALLEL_DEGREE description
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1031 - [benchmark] Adds safetensors support
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1030 - [docs] Adds log4j configuration document
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1029 - [ci] Upgrades gradle to 8.3
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1028 - Add FT llama integration test
Pull Request -
State: closed - Opened by rohithkrn about 1 year ago
#1027 - [docker] Updates cache directory
Pull Request -
State: closed - Opened by frankfliu about 1 year ago
#1026 - Adding docs for llm tuning params
Pull Request -
State: closed - Opened by maaquib about 1 year ago
#1025 - How to use vLLM with djl-serving?
Issue -
State: closed - Opened by yudhiesh about 1 year ago
- 6 comments
#1024 - Adding LLM tuning guide
Pull Request -
State: closed - Opened by maaquib about 1 year ago
- 2 comments
#1023 - add mpt and starcoder tests
Pull Request -
State: closed - Opened by lanking520 about 1 year ago