diff --git a/README.md b/README.md index 6230e2c..1b9f03c 100644 --- a/README.md +++ b/README.md @@ -202,6 +202,9 @@ curl -sSL https://install.python-poetry.org | python3 - ```bash # Install dependencies poetry install + +# Install flash-attn for Qwen3-Embedding-4B, it does not support PEP-517 builds, so no Poetry +pip install flash-attn --no-build-isolation ``` ### Optional: Weights and Biases for Experiment Tracking @@ -286,10 +289,10 @@ wandb login # Using FP16 mixed precision for faster training # Using a sample fraction of 0.5 to use half the dataset nohup eridu train \ - --model "intfloat/multilingual-e5-base" \ + --model "Qwen/Qwen3-Embedding-4B" \ --use-gpu \ --fp16 \ - --batch-size 32 \ + --batch-size 128 \ --epochs 1 \ --patience 1 \ --weight-decay 0.01 \ diff --git a/poetry.lock b/poetry.lock index 1c7be5d..be97112 100644 --- a/poetry.lock +++ b/poetry.lock @@ -3533,6 +3533,36 @@ extra = ["lxml (>=4.6)", "pydot (>=3.0.1)", "pygraphviz (>=1.14)", "sympy (>=1.1 test = ["pytest (>=7.2)", "pytest-cov (>=4.0)", "pytest-xdist (>=3.0)"] test-extras = ["pytest-mpl", "pytest-randomly"] +[[package]] +name = "ninja" +version = "1.13.0" +description = "Ninja is a small build system with a focus on speed" +optional = false +python-versions = ">=3.8" +groups = ["main"] +markers = "platform_machine == \"aarch64\" or platform_machine == \"x86_64\" or platform_machine != \"aarch64\" and platform_machine != \"x86_64\"" +files = [ + {file = "ninja-1.13.0-py3-none-macosx_10_9_universal2.whl", hash = "sha256:fa2a8bfc62e31b08f83127d1613d10821775a0eb334197154c4d6067b7068ff1"}, + {file = "ninja-1.13.0-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3d00c692fb717fd511abeb44b8c5d00340c36938c12d6538ba989fe764e79630"}, + {file = "ninja-1.13.0-py3-none-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:be7f478ff9f96a128b599a964fc60a6a87b9fa332ee1bd44fa243ac88d50291c"}, + {file = "ninja-1.13.0-py3-none-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:60056592cf495e9a6a4bea3cd178903056ecb0943e4de45a2ea825edb6dc8d3e"}, + {file = "ninja-1.13.0-py3-none-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:1c97223cdda0417f414bf864cfb73b72d8777e57ebb279c5f6de368de0062988"}, + {file = "ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fb46acf6b93b8dd0322adc3a4945452a4e774b75b91293bafcc7b7f8e6517dfa"}, + {file = "ninja-1.13.0-py3-none-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:4be9c1b082d244b1ad7ef41eb8ab088aae8c109a9f3f0b3e56a252d3e00f42c1"}, + {file = "ninja-1.13.0-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:6739d3352073341ad284246f81339a384eec091d9851a886dfa5b00a6d48b3e2"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:11be2d22027bde06f14c343f01d31446747dbb51e72d00decca2eb99be911e2f"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:aa45b4037b313c2f698bc13306239b8b93b4680eb47e287773156ac9e9304714"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_i686.whl", hash = "sha256:5f8e1e8a1a30835eeb51db05cf5a67151ad37542f5a4af2a438e9490915e5b72"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_ppc64le.whl", hash = "sha256:3d7d7779d12cb20c6d054c61b702139fd23a7a964ec8f2c823f1ab1b084150db"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_riscv64.whl", hash = "sha256:d741a5e6754e0bda767e3274a0f0deeef4807f1fec6c0d7921a0244018926ae5"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_s390x.whl", hash = "sha256:e8bad11f8a00b64137e9b315b137d8bb6cbf3086fbdc43bf1f90fd33324d2e96"}, + {file = "ninja-1.13.0-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:b4f2a072db3c0f944c32793e91532d8948d20d9ab83da9c0c7c15b5768072200"}, + {file = "ninja-1.13.0-py3-none-win32.whl", hash = "sha256:8cfbb80b4a53456ae8a39f90ae3d7a2129f45ea164f43fadfa15dc38c4aef1c9"}, + {file = "ninja-1.13.0-py3-none-win_amd64.whl", hash = "sha256:fb8ee8719f8af47fed145cced4a85f0755dd55d45b2bddaf7431fa89803c5f3e"}, + {file = "ninja-1.13.0-py3-none-win_arm64.whl", hash = "sha256:3c0b40b1f0bba764644385319028650087b4c1b18cdfa6f45cb39a3669b81aa9"}, + {file = "ninja-1.13.0.tar.gz", hash = "sha256:4a40ce995ded54d9dc24f8ea37ff3bf62ad192b547f6c7126e7e25045e76f978"}, +] + [[package]] name = "nodeenv" version = "1.9.1" @@ -7328,4 +7358,4 @@ files = [ [metadata] lock-version = "2.1" python-versions = "^3.12" -content-hash = "49ea2b403635f79d7c52a31aa892737d4385ee68c4c33e6b991492a39e6de1f4" +content-hash = "29eae4df3e739afa60f7a82eb0b333b3c5c53f831c8068abb5ae224106f0a779" diff --git a/pyproject.toml b/pyproject.toml index e912ee5..97f1071 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -88,6 +88,8 @@ cuml-cu12 = { version = "^25.6.0", source = "rapids" } cleanco = "^2.3" stop-words = "^2018.7.23" langdetect = "^1.0.9" +packaging = "^25.0" +ninja = "^1.13.0" [tool.poetry.group.dev.dependencies] black = { version = "^25.1.0", extras = ["jupyter"] }