Implementing text search and image search based on clip algorithm

• how to run

step 1:

  pip install -r requirements.txt

  python scripts/download_model.py  # download model

and the image data can be download from the internet.

step 2: you should select an vector db to save the embedding,such as milvus or faiss.

docker pull milvusdb/milvus:v2.4.14

###单独启动
docker run -d \
   --name milvus-standalone \
   --security-opt seccomp:unconfined \
   -e ETCD_USE_EMBED=true \
   -e ETCD_DATA_DIR=/var/lib/milvus/etcd \
   -e ETCD_CONFIG_PATH=/milvus/configs/embedEtcd.yaml \
   -e COMMON_STORAGETYPE=local \
   -v $(pwd)/volumes/milvus:/var/lib/milvus \
   -v $(pwd)/embedEtcd.yaml:/milvus/configs/embedEtcd.yaml \
   -v $(pwd)/user.yaml:/milvus/configs/user.yaml \
   -p 19530:19530 \
   -p 9091:9091 \
   -p 2379:2379 \
   --health-cmd="curl -f http://localhost:9091/healthz" \
   --health-interval=30s \
   --health-start-period=90s \
   --health-timeout=20s \
   --health-retries=3 \
   milvusdb/milvus:v2.4.14

 ###test connect
 python scripts/test_connect.py

step 3: get the embedding of datasets and save them to vector db

  python scripts/insert_search.py
  python scripts/main.py

• the result is shown on like this: