A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 22:15 UTC
[CI/Build] add python-json-logger to requirements-common (#12842)
Roger Meier committed
Feb 24, 2025 at 15:10 UTC
7940d8a6a7841113c89b168080b51785946f0cdb
2 files changed
+3
-7
examples/other/logging_configuration.md
+2
-7
@@ -49,7 +49,8 @@ disabled, an error will occur while starting vLLM.
49
### Example 1: Customize vLLM root logger
50
51
For this example, we will customize the vLLM root logger to use
52
-[`python-json-logger`](https://github.com/madzak/python-json-logger) to log to
52
+[`python-json-logger`](https://github.com/nhairs/python-json-logger)
53
+(which is part of the container image) to log to
54
STDOUT of the console in JSON format with a log level of `INFO`.
55
56
To begin, first, create an appropriate JSON logging configuration file:
@@ -82,12 +83,6 @@ To begin, first, create an appropriate JSON logging configuration file:
83
}
84
```
85
85
-Next, install the `python-json-logger` package if it's not already installed:
86
-
87
-```bash
88
-pip install python-json-logger
89
-```
90
-
86
Finally, run vLLM with the `VLLM_LOGGING_CONFIG_PATH` environment variable set
87
to the path of the custom logging configuration JSON file:
88
requirements-common.txt
+1
@@ -38,3 +38,4 @@ compressed-tensors == 0.9.2 # required for compressed-tensors
38
depyf==0.18.0 # required for profiling and debugging with compilation config
39
cloudpickle # allows pickling lambda functions in model_executor/models/registry.py
40
watchfiles # required for http server to monitor the updates of TLS files
41
+python-json-logger # Used by logging as per examples/other/logging_configuration.md