Merge branch 'main' into dpo

opea-project · Nov 8, 2024 · c44a54f · c44a54f
2 parents 058c9d1 + 786cabe
commit c44a54f
Show file tree

Hide file tree

Showing 42 changed files with 1,343 additions and 1,535 deletions.
diff --git a/.github/workflows/docker/compose/embeddings-compose-cd.yaml b/.github/workflows/docker/compose/embeddings-compose-cd.yaml
@@ -22,3 +22,7 @@ services:
     build:
       dockerfile: comps/embeddings/predictionguard/Dockerfile
     image: ${REGISTRY:-opea}/embedding-predictionguard:${TAG:-latest}
+  embedding-reranking-local:
+    build:
+      dockerfile: comps/embeddings/tei/langchain/Dockerfile.dynamic_batching
+    image: ${REGISTRY:-opea}/embedding-reranking-local:${TAG:-latest}
diff --git a/.github/workflows/docker/compose/llms-compose-cd.yaml b/.github/workflows/docker/compose/llms-compose-cd.yaml
@@ -23,10 +23,6 @@ services:
     build:
       dockerfile: comps/llms/text-generation/vllm/llama_index/Dockerfile
     image: ${REGISTRY:-opea}/llm-vllm-llamaindex:${TAG:-latest}
-  llm-vllm-llamaindex-hpu:
-    build:
-      dockerfile: comps/llms/text-generation/vllm/llama_index/dependency/Dockerfile.intel_hpu
-    image: ${REGISTRY:-opea}/llm-vllm-llamaindex-hpu:${TAG:-latest}
   llm-predictionguard:
     build:
       dockerfile: comps/llms/text-generation/predictionguard/Dockerfile

diff --git a/.github/workflows/docker/compose/llms-compose.yaml b/.github/workflows/docker/compose/llms-compose.yaml
@@ -24,10 +24,6 @@ services:
     build:
       dockerfile: comps/llms/text-generation/vllm/langchain/Dockerfile
     image: ${REGISTRY:-opea}/llm-vllm:${TAG:-latest}
-  llm-vllm-hpu:
-    build:
-      dockerfile: comps/llms/text-generation/vllm/langchain/dependency/Dockerfile.intel_hpu
-    image: ${REGISTRY:-opea}/llm-vllm-hpu:${TAG:-latest}
   llm-vllm-ray:
     build:
       dockerfile: comps/llms/text-generation/vllm/ray/Dockerfile

diff --git a/.github/workflows/scripts/freeze_images.sh b/.github/workflows/scripts/freeze_images.sh
@@ -5,8 +5,7 @@
 
 declare -A dict
 dict["langchain/langchain"]="docker://docker.io/langchain/langchain"
-# dict["vault.habana.ai/gaudi-docker/1.16.1/ubuntu22.04/habanalabs/pytorch-installer-2.2.2"]="docker://vault.habana.ai/gaudi-docker/1.16.1/ubuntu22.04/habanalabs/pytorch-installer-2.2.2"
-dict["opea/habanalabs:1.16.1-pytorch-installer-2.2.2"]="docker://docker.io/opea/habanalabs:1.16.1-pytorch-installer-2.2.2"
+dict["vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.0"]="docker://vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.0"
 
 function get_latest_version() {
     repo_image=$1

diff --git a/comps/animation/wav2lip/dependency/Dockerfile.intel_hpu b/comps/animation/wav2lip/dependency/Dockerfile.intel_hpu
@@ -1,6 +1,6 @@
 # Use a base image
 # FROM python:3.11-slim
-FROM vault.habana.ai/gaudi-docker/1.16.2/ubuntu22.04/habanalabs/pytorch-installer-2.2.2:latest AS hpu
+FROM vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.0 AS hpu
 
 # Set environment variables
 ENV LANG=en_US.UTF-8

diff --git a/comps/asr/whisper/dependency/Dockerfile.intel_hpu b/comps/asr/whisper/dependency/Dockerfile.intel_hpu
@@ -2,8 +2,7 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # HABANA environment
-# FROM vault.habana.ai/gaudi-docker/1.16.1/ubuntu22.04/habanalabs/pytorch-installer-2.2.2:latest as hpu
-FROM opea/habanalabs:1.16.1-pytorch-installer-2.2.2 as hpu
+FROM vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.0 AS hpu
 
 RUN useradd -m -s /bin/bash user && \
     mkdir -p /home/user && \

diff --git a/comps/cores/mega/cli.py b/comps/cores/mega/cli.py
@@ -4,16 +4,17 @@
 import argparse
 
 from .exporter import convert_to_docker_compose
+from .manifests_exporter import convert_to_manifests
 
 
-def export_kubernetes_manifests(mega_yaml, output_dir, device="cpu"):
-    print(f"Generating Kubernetes manifests from {mega_yaml} to {output_dir}")
-    # Add your logic to convert the YAML to Kubernetes manifest here
+def export_kubernetes_manifests(mega_yaml, output_file):
+    print(f"Generating Kubernetes manifests from {mega_yaml} to {output_file}")
+    convert_to_manifests(mega_yaml, output_file)
 
 
-def export_docker_compose(mega_yaml, output_file, device="cpu"):
+def export_docker_compose(mega_yaml, output_file):
     print(f"Generating Docker Compose file from {mega_yaml} to {output_file}")
-    convert_to_docker_compose(mega_yaml, output_file, device)
+    convert_to_docker_compose(mega_yaml, output_file)
 
 
 def opea_execute():
@@ -30,9 +31,6 @@ def opea_execute():
     compose_parser = export_subparsers.add_parser("docker-compose", help="Export to Docker Compose")
     compose_parser.add_argument("mega_yaml", help="Path to the mega YAML file")
     compose_parser.add_argument("output_file", help="Path to the Docker Compose file")
-    compose_parser.add_argument(
-        "--device", choices=["cpu", "gaudi", "xpu", "gpu"], default="cpu", help="Device type to use (default: cpu)"
-    )
 
     # Export to Kubernetes
     kube_parser = export_subparsers.add_parser("kubernetes", help="Export to Kubernetes")
@@ -48,7 +46,7 @@ def opea_execute():
     # Execute appropriate command
     if args.command == "export":
         if args.export_command == "docker-compose":
-            export_docker_compose(args.mega_yaml, args.output_file, args.device)
+            export_docker_compose(args.mega_yaml, args.output_file)
         elif args.export_command == "kubernetes":
             export_kubernetes_manifests(args.mega_yaml, args.output_dir, args.device)
         else: