From 8ec1b87f16f85f110bb3e7e6d6871525b571dbf9 Mon Sep 17 00:00:00 2001 From: Nicolas Patry Date: Wed, 4 Oct 2023 12:57:21 +0200 Subject: [PATCH] Adding titles to CLI doc. (#1094) # What does this PR do? Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. --- docs/source/basic_tutorials/launcher.md | 122 +++++++++++++++++++++++- update_doc.py | 28 +++++- 2 files changed, 148 insertions(+), 2 deletions(-) diff --git a/docs/source/basic_tutorials/launcher.md b/docs/source/basic_tutorials/launcher.md index 08a6ed86cc3..62abe8c6edb 100644 --- a/docs/source/basic_tutorials/launcher.md +++ b/docs/source/basic_tutorials/launcher.md @@ -8,34 +8,52 @@ Text Generation Launcher Usage: text-generation-launcher [OPTIONS] Options: +``` +## MODEL_ID +```shell --model-id The name of the model to load. Can be a MODEL_ID as listed on like `gpt2` or `OpenAssistant/oasst-sft-1-pythia-12b`. Or it can be a local directory containing the necessary files as saved by `save_pretrained(...)` methods of transformers [env: MODEL_ID=] [default: bigscience/bloom-560m] +``` +## REVISION +```shell --revision The actual revision of the model if you're referring to a model on the hub. You can use a specific commit id or a branch like `refs/pr/2` [env: REVISION=] +``` +## VALIDATION_WORKERS +```shell --validation-workers The number of tokenizer workers used for payload validation and truncation inside the router [env: VALIDATION_WORKERS=] [default: 2] +``` +## SHARDED +```shell --sharded Whether to shard the model across multiple GPUs By default text-generation-inference will use all available GPUs to run the model. Setting it to `false` deactivates `num_shard` [env: SHARDED=] [possible values: true, false] +``` +## NUM_SHARD +```shell --num-shard The number of shards to use if you don't want to use all GPUs on a given machine. You can use `CUDA_VISIBLE_DEVICES=0,1 text-generation-launcher... --num_shard 2` and `CUDA_VISIBLE_DEVICES=2,3 text-generation-launcher... --num_shard 2` to launch 2 copies with 2 shard each on a given machine with 4 GPUs for instance [env: NUM_SHARD=] +``` +## QUANTIZE +```shell --quantize Whether you want the model to be quantized @@ -49,53 +67,80 @@ Options: - bitsandbytes-nf4: Bitsandbytes 4bit. Can be applied on any model, will cut the memory requirement by 4x, but it is known that the model will be much slower to run than the native f16 - bitsandbytes-fp4: Bitsandbytes 4bit. nf4 should be preferred in most cases but maybe this one has better perplexity performance for you model +``` +## DTYPE +```shell --dtype The dtype to be forced upon the model. This option cannot be used with `--quantize` [env: DTYPE=] [possible values: float16, bfloat16] +``` +## TRUST_REMOTE_CODE +```shell --trust-remote-code Whether you want to execute hub modelling code. Explicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision [env: TRUST_REMOTE_CODE=] +``` +## MAX_CONCURRENT_REQUESTS +```shell --max-concurrent-requests The maximum amount of concurrent requests for this particular deployment. Having a low limit will refuse clients requests instead of having them wait for too long and is usually good to handle backpressure correctly [env: MAX_CONCURRENT_REQUESTS=] [default: 128] +``` +## MAX_BEST_OF +```shell --max-best-of This is the maximum allowed value for clients to set `best_of`. Best of makes `n` generations at the same time, and return the best in terms of overall log probability over the entire generated sequence [env: MAX_BEST_OF=] [default: 2] +``` +## MAX_STOP_SEQUENCES +```shell --max-stop-sequences This is the maximum allowed value for clients to set `stop_sequences`. Stop sequences are used to allow the model to stop on more than just the EOS token, and enable more complex "prompting" where users can preprompt the model in a specific way and define their "own" stop token aligned with their prompt [env: MAX_STOP_SEQUENCES=] [default: 4] +``` +## MAX_TOP_N_TOKENS +```shell --max-top-n-tokens This is the maximum allowed value for clients to set `top_n_tokens`. `top_n_tokens is used to return information about the the `n` most likely tokens at each generation step, instead of just the sampled token. This information can be used for downstream tasks like for classification or ranking [env: MAX_TOP_N_TOKENS=] [default: 5] +``` +## MAX_INPUT_LENGTH +```shell --max-input-length This is the maximum allowed input length (expressed in number of tokens) for users. The larger this value, the longer prompt users can send which can impact the overall memory required to handle the load. Please note that some models have a finite range of sequence they can handle [env: MAX_INPUT_LENGTH=] [default: 1024] +``` +## MAX_TOTAL_TOKENS +```shell --max-total-tokens This is the most important value to set as it defines the "memory budget" of running clients requests. Clients will send input sequences and ask to generate `max_new_tokens` on top. with a value of `1512` users can send either a prompt of `1000` and ask for `512` new tokens, or send a prompt of `1` and ask for `1511` max_new_tokens. The larger this value, the larger amount each request will be in your RAM and the less effective batching can be [env: MAX_TOTAL_TOKENS=] [default: 2048] +``` +## WAITING_SERVED_RATIO +```shell --waiting-served-ratio This represents the ratio of waiting queries vs running queries where you want to start considering pausing the running queries to include the waiting ones into the same batch. `waiting_served_ratio=1.2` Means when 12 queries are waiting and there's only 10 queries left in the current batch we check if we can fit those 12 waiting queries into the batching strategy, and if yes, then batching happens delaying the 10 running queries by a `prefill` run. @@ -104,12 +149,18 @@ Options: [env: WAITING_SERVED_RATIO=] [default: 1.2] +``` +## MAX_BATCH_PREFILL_TOKENS +```shell --max-batch-prefill-tokens Limits the number of tokens for the prefill operation. Since this operation take the most memory and is compute bound, it is interesting to limit the number of requests that can be sent [env: MAX_BATCH_PREFILL_TOKENS=] [default: 4096] +``` +## MAX_BATCH_TOTAL_TOKENS +```shell --max-batch-total-tokens **IMPORTANT** This is one critical control to allow maximum usage of the available hardware. @@ -123,6 +174,9 @@ Options: [env: MAX_BATCH_TOTAL_TOKENS=] +``` +## MAX_WAITING_TOKENS +```shell --max-waiting-tokens This setting defines how many tokens can be passed before forcing the waiting queries to be put on the batch (if the size of the batch allows for it). New queries require 1 `prefill` forward, which is different from `decode` and therefore you need to pause the running batch in order to run `prefill` to create the correct values for the waiting queries to be able to join the batch. @@ -135,57 +189,87 @@ Options: [env: MAX_WAITING_TOKENS=] [default: 20] +``` +## HOSTNAME +```shell --hostname The IP address to listen on [env: HOSTNAME=] [default: 0.0.0.0] +``` +## PORT +```shell -p, --port The port to listen on [env: PORT=] [default: 3000] +``` +## SHARD_UDS_PATH +```shell --shard-uds-path The name of the socket for gRPC communication between the webserver and the shards [env: SHARD_UDS_PATH=] [default: /tmp/text-generation-server] +``` +## MASTER_ADDR +```shell --master-addr The address the master shard will listen on. (setting used by torch distributed) [env: MASTER_ADDR=] [default: localhost] +``` +## MASTER_PORT +```shell --master-port The address the master port will listen on. (setting used by torch distributed) [env: MASTER_PORT=] [default: 29500] +``` +## HUGGINGFACE_HUB_CACHE +```shell --huggingface-hub-cache The location of the huggingface hub cache. Used to override the location if you want to provide a mounted disk for instance [env: HUGGINGFACE_HUB_CACHE=] +``` +## WEIGHTS_CACHE_OVERRIDE +```shell --weights-cache-override The location of the huggingface hub cache. Used to override the location if you want to provide a mounted disk for instance [env: WEIGHTS_CACHE_OVERRIDE=] +``` +## DISABLE_CUSTOM_KERNELS +```shell --disable-custom-kernels For some models (like bloom), text-generation-inference implemented custom cuda kernels to speed up inference. Those kernels were only tested on A100. Use this flag to disable them if you're running on different hardware and encounter issues [env: DISABLE_CUSTOM_KERNELS=] +``` +## CUDA_MEMORY_FRACTION +```shell --cuda-memory-fraction Limit the CUDA available memory. The allowed value equals the total visible memory multiplied by cuda-memory-fraction [env: CUDA_MEMORY_FRACTION=] [default: 1.0] +``` +## ROPE_SCALING +```shell --rope-scaling Rope scaling will only be used for RoPE models and allow rescaling the position rotary to accomodate for larger prompts. @@ -198,50 +282,86 @@ Options: [env: ROPE_SCALING=] [possible values: linear, dynamic] +``` +## ROPE_FACTOR +```shell --rope-factor Rope scaling will only be used for RoPE models See `rope_scaling` [env: ROPE_FACTOR=] +``` +## JSON_OUTPUT +```shell --json-output Outputs the logs in JSON format (useful for telemetry) [env: JSON_OUTPUT=] +``` +## OTLP_ENDPOINT +```shell --otlp-endpoint [env: OTLP_ENDPOINT=] +``` +## CORS_ALLOW_ORIGIN +```shell --cors-allow-origin [env: CORS_ALLOW_ORIGIN=] +``` +## WATERMARK_GAMMA +```shell --watermark-gamma [env: WATERMARK_GAMMA=] +``` +## WATERMARK_DELTA +```shell --watermark-delta [env: WATERMARK_DELTA=] +``` +## NGROK +```shell --ngrok Enable ngrok tunneling [env: NGROK=] +``` +## NGROK_AUTHTOKEN +```shell --ngrok-authtoken ngrok authentication token [env: NGROK_AUTHTOKEN=] +``` +## NGROK_EDGE +```shell --ngrok-edge ngrok edge [env: NGROK_EDGE=] +``` +## ENV +```shell -e, --env Display a lot of information about your runtime environment +``` +## HELP +```shell -h, --help Print help (see a summary with '-h') +``` +## VERSION +```shell -V, --version Print version -``` \ No newline at end of file +``` diff --git a/update_doc.py b/update_doc.py index 1fa398b0e7f..6206e2111a5 100644 --- a/update_doc.py +++ b/update_doc.py @@ -11,8 +11,34 @@ def main(): output = subprocess.check_output(["text-generation-launcher", "--help"]).decode( "utf-8" ) + wrap_code_blocks_flag = "" - final_doc = f"# Text-generation-launcher arguments\n\n{wrap_code_blocks_flag}\n\n```shell\n{output}\n```" + final_doc = f"# Text-generation-launcher arguments\n\n{wrap_code_blocks_flag}\n\n" + + lines = output.split("\n") + + header = "" + block = [] + for line in lines: + if line.startswith(" -") or line.startswith(" -"): + rendered_block = '\n'.join(block) + if header: + final_doc += f"## {header}\n```shell\n{rendered_block}\n```\n" + else: + final_doc += f"```shell\n{rendered_block}\n```\n" + block = [] + tokens = line.split("<") + if len(tokens)>1: + header = tokens[-1][:-1] + else: + header = line.split("--")[-1] + header = header.upper().replace("-", "_") + + block.append(line) + + rendered_block = '\n'.join(block) + final_doc += f"## {header}\n```shell\n{rendered_block}\n```\n" + block = [] filename = "docs/source/basic_tutorials/launcher.md" if args.check: