feat: initialize project with Python 3.10 and feature retrieval module

This commit is contained in:
2026-02-01 12:11:11 +08:00
parent 9e9070bdb4
commit cf83c09165
7 changed files with 2792 additions and 324 deletions

1
.python-version Normal file
View File

@@ -0,0 +1 @@
3.10

View File

@@ -39,9 +39,7 @@ in {
# https://devenv.sh/basics/
enterShell = ''
eval "$(micromamba shell hook --shell bash)"
micromamba activate ./.venv
export PATH=$PATH:/home/sikongjueluo/Projects/Mini-Nav
source ./.venv/bin/activate
'';
# https://devenv.sh/tasks/

6
main.py Normal file
View File

@@ -0,0 +1,6 @@
def main():
print("Hello from mini-nav!")
if __name__ == "__main__":
main()

View File

@@ -0,0 +1,37 @@
from typing import cast
import torch
from tqdm.auto import tqdm
from datasets import Dataset, load_dataset
from transformers import AutoImageProcessor, AutoModel
@torch.no_grad()
def establish_database(processor, model, images, batch_size=64):
device = model.device
model.eval()
for i in tqdm(range(0, len(images), batch_size)):
batch = images[i : i + batch_size]
inputs = processor(images=batch, return_tensors="pt")
# 迁移数据到GPU
inputs.to(device, non_blocking=True)
outputs = model(**inputs)
feats = outputs.last_hidden_state # [B, N, D]
# 后处理 / 存库
if __name__ == "__main__":
train_dataset = load_dataset("uoft-cs/cifar10", split="train")
train_dataset = cast(Dataset, train_dataset)
processor = AutoImageProcessor.from_pretrained(
"facebook/dinov2-large", device_map="cuda"
)
model = AutoModel.from_pretrained("facebook/dinov2-large", device_map="cuda")
establish_database(processor, model, train_dataset["img"])

34
pyproject.toml Normal file
View File

@@ -0,0 +1,34 @@
[project]
name = "mini-nav"
version = "0.1.0"
description = "Add your description here"
readme = "README.md"
requires-python = ">=3.10"
dependencies = [
"accelerate>=1.12.0",
"datasets>=4.5.0",
"h5py>=3.15.1",
"httpx[socks]>=0.28.1",
"torch>=2.10.0",
"torchvision>=0.25.0",
"transformers>=5.0.0",
]
[dependency-groups]
dev = [
"pytest>=9.0.2",
"pytest-mpi>=0.6",
]
[[tool.uv.index]]
name = "pytorch-cu130"
url = "https://download.pytorch.org/whl/cu130"
explicit = true
[tool.uv.sources]
torch = [
{ index = "pytorch-cu130", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
torchvision = [
{ index = "pytorch-cu130", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]

2713
uv.lock generated Normal file

File diff suppressed because it is too large Load Diff

321
venv.yaml
View File

@@ -1,321 +0,0 @@
name:
channels:
- aihabitat
- conda-forge
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
dependencies:
- _libgcc_mutex=0.1
- _openmp_mutex=4.5
- aiodns=3.5.0
- aiohappyeyeballs=2.6.1
- aiohttp=3.12.15
- aiosignal=1.4.0
- alsa-lib=1.2.15.3
- aom=3.9.1
- arrow-cpp=19.0.0
- async-timeout=5.0.1
- attrs=25.3.0
- aws-c-auth=0.8.5
- aws-c-cal=0.8.5
- aws-c-common=0.11.1
- aws-c-compression=0.3.1
- aws-c-event-stream=0.5.4
- aws-c-http=0.9.3
- aws-c-io=0.17.0
- aws-c-mqtt=0.12.2
- aws-c-s3=0.7.11
- aws-c-sdkutils=0.2.3
- aws-checksums=0.2.3
- aws-crt-cpp=0.31.0
- aws-sdk-cpp=1.11.528
- brotli=1.0.9
- brotli-bin=1.0.9
- brotlicffi=1.0.9.2
- bzip2=1.0.8
- c-ares=1.34.6
- ca-certificates=2026.1.4
- cairo=1.18.4
- certifi=2025.8.3
- cffi=2.0.0
- charset-normalizer=3.4.4
- cmake=3.14.0
- colorama=0.4.6
- contourpy=1.3.0
- curl=8.8.0
- cycler=0.12.1
- cyrus-sasl=2.1.28
- datasets=3.3.2
- dav1d=1.2.1
- dbus=1.16.2
- dill=0.3.8
- expat=2.7.3
- filelock=3.17.0
- fontconfig=2.14.2
- fonttools=4.59.1
- freetype=2.13.3
- fribidi=1.0.16
- frozenlist=1.8.0
- fsspec=2024.12.0
- gflags=2.2.2
- gitdb=4.0.12
- gitpython=3.1.45
- glog=0.5.0
- graphite2=1.3.14
- habitat-sim=0.3.3
- habitat-sim-mutex=1.0
- harfbuzz=10.2.0
- hf-xet=1.1.8
- huggingface_hub=0.34.4
- icu=73.2
- idna=3.11
- imageio=2.37.0
- importlib-resources=6.5.2
- importlib_resources=6.5.2
- jpeg=9e
- keyutils=1.6.3
- kiwisolver=1.4.7
- krb5=1.21.3
- lcms2=2.17
- ld_impl_linux-64=2.45
- lerc=4.0.0
- libabseil=20250127.0
- libavif=1.3.0
- libavif16=1.3.0
- libblas=3.11.0
- libbrotlicommon=1.0.9
- libbrotlidec=1.0.9
- libbrotlienc=1.0.9
- libcblas=3.11.0
- libclang13=21.1.8
- libcups=2.4.15
- libcurl=8.8.0
- libdeflate=1.22
- libdrm=2.4.125
- libedit=3.1.20250104
- libegl=1.7.0
- libev=4.33
- libevent=2.1.12
- libexpat=2.7.3
- libffi=3.5.2
- libgcc=15.2.0
- libgcc-ng=15.2.0
- libgfortran=15.2.0
- libgfortran-ng=15.2.0
- libgfortran5=15.2.0
- libgl=1.7.0
- libglib=2.84.2
- libglvnd=1.7.0
- libglx=1.7.0
- libgomp=15.2.0
- libgrpc=1.71.0
- libiconv=1.18
- libkrb5=1.21.3
- liblapack=3.11.0
- libllvm14=14.0.6
- libllvm15=15.0.7
- libllvm21=21.1.8
- liblzma=5.8.2
- liblzma-devel=5.8.2
- libnghttp2=1.58.0
- libnsl=2.0.1
- libntlm=1.8
- libopenblas=0.3.30
- libopus=1.6.1
- libpciaccess=0.18
- libpng=1.6.43
- libpq=17.6
- libprotobuf=5.29.3
- libre2-11=2024.07.02
- libsqlite=3.46.0
- libssh2=1.11.0
- libstdcxx=15.2.0
- libstdcxx-ng=15.2.0
- libthrift=0.15.0
- libtiff=4.7.1
- libuuid=2.41.3
- libuv=1.51.0
- libvpx=1.13.1
- libwebp-base=1.6.0
- libxcb=1.17.0
- libxcrypt=4.4.36
- libxkbcommon=1.11.0
- libxkbfile=1.1.0
- libxml2=2.13.9
- libxslt=1.1.43
- libzlib=1.2.13
- llvmlite=0.43.0
- lmdb=0.9.31
- lz4-c=1.9.4
- matplotlib=3.9.4
- matplotlib-base=3.9.4
- mesalib=25.1.5
- multidict=6.7.0
- multiprocess=0.70.15
- munkres=1.1.4
- mysql-common=9.3.0
- mysql-libs=9.3.0
- ncurses=6.5
- nspr=4.38
- nss=3.100
- numba=0.60.0
- numpy=1.26.4
- openjpeg=2.5.2
- openldap=2.6.10
- openssl=3.6.0
- orc=2.1.1
- packaging=26.0
- pandas=1.2.4
- pcre2=10.42
- pillow=11.3.0
- pip=25.2
- pixman=0.46.4
- propcache=0.3.1
- protobuf=5.29.3
- pthread-stubs=0.4
- pyarrow=19.0.0
- pycares=4.10.0
- pycparser=2.23
- pyparsing=3.2.3
- pyside6=6.9.2
- pysocks=1.7.1
- python=3.9.19
- python-dateutil=2.9.0.post0
- python-xxhash=3.5.0
- python_abi=3.9
- pytz=2025.2
- pyyaml=6.0.2
- qhull=2020.2
- qtbase=6.9.2
- qtdeclarative=6.9.2
- qtshadertools=6.9.2
- qtsvg=6.9.2
- qttools=6.9.2
- qtwebchannel=6.9.2
- qtwebengine=6.9.1
- qtwebsockets=6.9.2
- quaternion=2023.0.4
- rav1e=0.7.1
- re2=2024.07.02
- readline=8.3
- regex=2025.9.1
- requests=2.32.5
- rhash=1.4.3
- s2n=1.5.14
- safetensors=0.5.3
- scipy=1.13.1
- setuptools=80.9.0
- six=1.17.0
- smmap=3.0.5
- snappy=1.2.2
- spirv-tools=2025.4
- sqlite=3.51.1
- svt-av1=3.1.2
- tk=8.6.13
- tokenizers=0.21.0
- tornado=6.5.2
- tqdm=4.67.1
- transformers=4.49.0
- typing-extensions=4.15.0
- typing_extensions=4.15.0
- tzdata=2025c
- unicodedata2=16.0.0
- urllib3=2.5.0
- utf8proc=2.6.1
- uvloop=0.22.1
- wheel=0.45.1
- withbullet=2.0
- xcb-util=0.4.1
- xcb-util-cursor=0.1.6
- xcb-util-image=0.4.0
- xcb-util-keysyms=0.4.1
- xcb-util-renderutil=0.3.10
- xcb-util-wm=0.4.2
- xkeyboard-config=2.46
- xorg-libice=1.1.2
- xorg-libsm=1.2.6
- xorg-libx11=1.8.12
- xorg-libxau=1.0.12
- xorg-libxcomposite=0.4.6
- xorg-libxcursor=1.2.3
- xorg-libxdamage=1.1.6
- xorg-libxdmcp=1.1.5
- xorg-libxext=1.3.6
- xorg-libxfixes=6.0.2
- xorg-libxi=1.8.2
- xorg-libxinerama=1.1.5
- xorg-libxrandr=1.5.4
- xorg-libxrender=0.9.12
- xorg-libxshmfence=1.3.3
- xorg-libxtst=1.2.5
- xorg-libxxf86vm=1.1.6
- xxhash=0.8.0
- xz=5.8.2
- xz-gpl-tools=5.8.2
- xz-tools=5.8.2
- yaml=0.2.5
- yarl=1.18.0
- zipp=3.23.0
- zlib=1.2.13
- zstd=1.5.6
- pip:
- Jinja2==3.1.6
- Markdown==3.9
- MarkupSafe==3.0.3
- Pygments==2.19.2
- Werkzeug==3.1.5
- absl-py==2.3.1
- antlr4-python3-runtime==4.9.3
- braceexpand==0.1.7
- click==8.1.8
- cloudpickle==3.1.2
- decorator==5.2.1
- faster-fifo==1.5.2
- grpcio==1.76.0
- gym==0.23.0
- gym-notices==0.1.0
- habitat-baselines==0.3.320250127
- habitat-lab==0.3.320250127
- hydra-core==1.3.2
- ifcfg==0.24
- imageio-ffmpeg==0.6.0
- importlib_metadata==8.7.1
- lmdb==1.7.5
- markdown-it-py==3.0.0
- mdurl==0.1.2
- moviepy==2.2.1
- mpmath==1.3.0
- msgpack==1.1.2
- networkx==3.2.1
- nvidia-cublas-cu12==12.8.4.1
- nvidia-cuda-cupti-cu12==12.8.90
- nvidia-cuda-nvrtc-cu12==12.8.93
- nvidia-cuda-runtime-cu12==12.8.90
- nvidia-cudnn-cu12==9.10.2.21
- nvidia-cufft-cu12==11.3.3.83
- nvidia-cufile-cu12==1.13.1.3
- nvidia-curand-cu12==10.3.9.90
- nvidia-cusolver-cu12==11.7.3.90
- nvidia-cusparse-cu12==12.5.8.93
- nvidia-cusparselt-cu12==0.7.1
- nvidia-nccl-cu12==2.27.3
- nvidia-nvjitlink-cu12==12.8.93
- nvidia-nvtx-cu12==12.8.90
- objectio==0.2.29
- omegaconf==2.3.0
- opencv-python==4.11.0.86
- pillow==10.4.0
- proglog==0.1.12
- python-dotenv==1.2.1
- rich==14.3.1
- shellingham==1.5.4
- simplejson==3.20.2
- sympy==1.14.0
- tensorboard==2.20.0
- tensorboard-data-server==0.7.2
- threadpoolctl==3.6.0
- torch==2.8.0
- torchvision==0.23.0
- triton==3.4.0
- typer==0.21.1
- webdataset==0.1.40