Merge pull request #31 from beclab/feat/bertv3-embedding

feat: add r4userembedding with model image

.github/workflows/update_r4userembedding_with_model.yml

@@ -0,0 +1,65 @@
+name: Publish to Dockerhub ( userembedding )
+
+on:
+    workflow_dispatch:
+      inputs:
+        tags:
+          description: 'Release Tags'
+
+jobs:
+    publish_dockerhub_amd64:
+        runs-on: ubuntu-latest
+        steps:
+            - name: Check out the repo
+              uses: actions/checkout@v3
+
+            - name: Log in to Docker Hub
+              uses: docker/login-action@v2
+              with:
+                  username: ${{ secrets.DOCKERHUB_USERNAME }}
+                  password: ${{ secrets.DOCKERHUB_PASS }}
+
+            - name: Build r4userembeddingwithmodel and push Docker image
+              uses: docker/build-push-action@v3
+              with:
+                push: true
+                tags: beclab/r4userembeddingwithmodel:${{ github.event.inputs.tags }}-amd64
+                file: Dockerfile.r4userembeddingwithmodel
+                platforms: linux/amd64
+
+    publish_dockerhub_arm64:
+        runs-on: self-hosted
+        steps:
+            - name: Check out the repo
+              uses: actions/checkout@v3
+
+            - name: Log in to Docker Hub
+              uses: docker/login-action@v2
+              with:
+                  username: ${{ secrets.DOCKERHUB_USERNAME }}
+                  password: ${{ secrets.DOCKERHUB_PASS }}
+
+            - name: Build r4userembeddingwithmodel and push Docker image
+              uses: docker/build-push-action@v3
+              with:
+                push: true
+                tags: beclab/r4userembeddingwithmodel:${{ github.event.inputs.tags }}-arm64
+                file: Dockerfile.r4userembeddingwithmodel
+                platforms: linux/arm64
+
+    publish_manifest:
+        needs:
+        - publish_dockerhub_amd64
+        - publish_dockerhub_arm64
+        runs-on: ubuntu-latest
+        steps:
+          - name: Log in to Docker Hub
+            uses: docker/login-action@v2
+            with:
+              username: ${{ secrets.DOCKERHUB_USERNAME }}
+              password: ${{ secrets.DOCKERHUB_PASS }}
+
+          - name: Push manifest
+            run: |
+              docker manifest create beclab/r4userembeddingwithmodel:${{ github.event.inputs.tags }} --amend beclab/r4userembeddingwithmodel:${{ github.event.inputs.tags }}-amd64 --amend beclab/r4userembeddingwithmodel:${{ github.event.inputs.tags }}-arm64
+              docker manifest push beclab/r4userembeddingwithmodel:${{ github.event.inputs.tags }}

Dockerfile.r4userembeddingwithmodel

@@ -0,0 +1,35 @@
+FROM ubuntu:jammy As builder
+
+# RUN apk add openssl-dev musl-dev g++
+
+RUN mkdir -p /userembedding && \
+    apt update && \
+    apt install curl -y && \
+    curl https://sh.rustup.rs -sSf | bash -s -- -y && \
+    # echo 'source $HOME/.cargo/env' >> $HOME/.bashrc && \
+    apt install build-essential -y && \
+    apt-get install libssl-dev -y && \
+    apt-get install pkg-config -y 
+
+ENV PATH="/root/.cargo/bin:${PATH}"
+WORKDIR /userembedding
+COPY user-embedding/src  /userembedding/src
+COPY user-embedding/Cargo.toml /userembedding/Cargo.toml
+RUN cargo build --release 
+RUN  /userembedding/target/release/downloadmodel
+
+
+
+FROM  ubuntu:jammy
+
+# Import from builder.
+
+
+WORKDIR /userembedding
+
+# Copy our build
+COPY --from=builder /root/.cache/huggingface  /root/.cache/huggingface
+COPY --from=builder /userembedding/target/release/userembedding ./
+
+
+CMD ["/userembedding/userembedding"]

user-embedding/Cargo.toml

@@ -37,3 +37,7 @@ tracing-subscriber = "0.3.18"
 [[bin]]
 name = "userembedding"
 path = "src/userembedding.rs"
+
+[[bin]]
+name = "downloadmodel"
+path = "src/download_model.rs"

user-embedding/src/download_model.rs

@@ -0,0 +1,31 @@
+use hf_hub::api::sync::Api;
+
+use userembedding::{
+    common,
+    embedding_common::{self, MODEL_RELATED_INFO_MAP},
+};
+
+fn download_models() -> Result<(), Box<dyn std::error::Error>> {
+    for (_, model_related_info) in MODEL_RELATED_INFO_MAP.iter() {
+        tracing::info!("Downloading model {}...", model_related_info.model_name);
+        let default_model: String = model_related_info.hugging_face_model_name.to_string();
+        let default_revision: String = model_related_info.hugging_face_model_revision.to_string();
+        let (_, _) =
+            embedding_common::build_model_and_tokenizer(default_model, default_revision).unwrap();
+        tracing::info!(
+            "Model {} downloaded successfully",
+            model_related_info.model_name
+        );
+    }
+
+    Ok(())
+}
+
+fn main() {
+    // Call the function to download models
+    common::init_logger();
+
+    if let Err(e) = download_models() {
+        eprintln!("Error downloading models: {}", e);
+    }
+}