vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-10 09:04:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	.devops/full-rocm.Dockerfile
#	.devops/llama-cli-rocm.Dockerfile
#	.devops/llama-server-rocm.Dockerfile
#	.github/workflows/build.yml
#	.github/workflows/python-type-check.yml
#	CMakeLists.txt
#	CONTRIBUTING.md
#	README.md
#	ci/run.sh
#	examples/embedding/embedding.cpp
#	examples/server/README.md
#	flake.lock
#	ggml/include/ggml.h
#	ggml/src/ggml.c
#	requirements/requirements-convert_legacy_llama.txt
#	scripts/sync-ggml.last
#	src/llama-vocab.cpp
#	src/llama.cpp
#	tests/test-backend-ops.cpp
#	tests/test-grad0.cpp
#	tests/test-tokenizer-0.cpp
This commit is contained in:
Concedo 2024-10-02 01:00:57 +08:00
commit ce7f9c9a2c
39 changed files with 103400 additions and 102738 deletions
Download patch file Download diff file Expand all files Collapse all files

18
common/arg.cpp
View file

@ -285,6 +285,10 @@ static bool gpt_params_parse_ex(int argc, char ** argv, gpt_params_context & ctx
params.kv_overrides.back().key[0] = 0;
}
if (params.reranking && params.embedding) {
throw std::invalid_argument("error: either --embedding or --reranking can be specified, but not both");
}
return true;
}
@ -392,7 +396,7 @@ gpt_params_context gpt_params_parser_init(gpt_params & params, llama_example ex,
[](gpt_params & params) {
params.verbose_prompt = true;
}
).set_examples({LLAMA_EXAMPLE_MAIN}));
));
add_opt(llama_arg(
{"--no-display-prompt"},
format("don't print prompt at generation (default: %s)", !params.display_prompt ? "true" : "false"),
@ -1094,13 +1098,14 @@ gpt_params_context gpt_params_parser_init(gpt_params & params, llama_example ex,
}
).set_sparam());
add_opt(llama_arg(
{"--pooling"}, "{none,mean,cls,last}",
{"--pooling"}, "{none,mean,cls,last,rank}",
"pooling type for embeddings, use model default if unspecified",
[](gpt_params & params, const std::string & value) {
/**/ if (value == "none") { params.pooling_type = LLAMA_POOLING_TYPE_NONE; }
else if (value == "mean") { params.pooling_type = LLAMA_POOLING_TYPE_MEAN; }
else if (value == "cls") { params.pooling_type = LLAMA_POOLING_TYPE_CLS; }
else if (value == "cls") { params.pooling_type = LLAMA_POOLING_TYPE_CLS; }
else if (value == "last") { params.pooling_type = LLAMA_POOLING_TYPE_LAST; }
else if (value == "rank") { params.pooling_type = LLAMA_POOLING_TYPE_RANK; }
else { throw std::invalid_argument("invalid value"); }
}
).set_examples({LLAMA_EXAMPLE_EMBEDDING, LLAMA_EXAMPLE_RETRIEVAL, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_POOLING"));
@ -1750,6 +1755,13 @@ gpt_params_context gpt_params_parser_init(gpt_params & params, llama_example ex,
params.embedding = true;
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_EMBEDDINGS"));
add_opt(llama_arg(
{"--reranking", "--rerank"},
format("enable reranking endpoint on server (default: %s)", params.reranking ? "enabled" : "disabled"),
[](gpt_params & params) {
params.reranking = true;
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_RERANKING"));
add_opt(llama_arg(
{"--api-key"}, "KEY",
"API key to use for authentication (default: none)",