Merge pull request #2 from the023/fixdocker

Fix docker PWD and network
jasonacox · Feb 8, 2024 · c0af4b0 · c0af4b0
2 parents 542a0a8 + 938d96a
commit c0af4b0
Show file tree

Hide file tree

Showing 17 changed files with 23 additions and 21 deletions.
diff --git a/README.md b/README.md
@@ -64,7 +64,7 @@ python3 -m llama_cpp.server \
     --model ./models/mistral-7b-instruct-v0.1.Q5_K_M.gguf \
     --host localhost \
     --n_gpu_layers 99 \
-    -n_ctx 2048 \
+    --n_ctx 2048 \
     --chat_format llama-2
 ```
 
@@ -111,7 +111,7 @@ docker run \
     -e OPENAI_API_BASE="http://localhost:8000/v1" \
     -e LLM_MODEL="tinyllm" \
     -e USE_SYSTEM="false" \
-    -v ./prompts.json:/app/prompts.json \
+    -v $PWD/prompts.json:/app/prompts.json \
     --name chatbot \
     --restart unless-stopped \
     jasonacox/chatbot

diff --git a/chatbot/Dockerfile b/chatbot/Dockerfile
@@ -51,7 +51,8 @@ EXPOSE $PORT
 #     -e DEVICE="cpu" \
 #     -e RESULTS=1 \
 #     -e USE_SYSTEM="false" \
-#     -v ./prompts.json:/app/prompts.json \
+#     -v $PWD/prompts.json:/app/prompts.json \
+#     --network="host" \
 #     --name chatbot \
 #     --restart unless-stopped \
 #     chatbot:0.11.0
diff --git a/chatbot/Dockerfile.basic b/chatbot/Dockerfile.basic
@@ -51,7 +51,7 @@ EXPOSE $PORT
 #     -e DEVICE="cpu" \
 #     -e RESULTS=1 \
 #     -e USE_SYSTEM="false" \
-#     -v ./prompts.json:/app/prompts.json \
+#     -v $PWD/prompts.json:/app/prompts.json \
 #     --name chatbot \
 #     --restart unless-stopped \
 #     chatbot:0.11.0
diff --git a/chatbot/Dockerfile.rag b/chatbot/Dockerfile.rag
@@ -51,7 +51,7 @@ EXPOSE $PORT
 #     -e DEVICE="cpu" \
 #     -e RESULTS=1 \
 #     -e USE_SYSTEM="false" \
-#     -v ./prompts.json:/app/prompts.json \
+#     -v $PWD/prompts.json:/app/prompts.json \
 #     --name chatbot \
 #     --restart unless-stopped \
 #     chatbot:0.11.0
diff --git a/chatbot/README.md b/chatbot/README.md
@@ -20,7 +20,7 @@ docker run \
     -e OPENAI_API_BASE="http://localhost:8000/v1" \
     -e LLM_MODEL="tinyllm" \
     -e USE_SYSTEM="false" \
-    -v ./prompts.json:/app/prompts.json \
+    -v $PWD/prompts.json:/app/prompts.json \
     --name chatbot \
     --restart unless-stopped \
     jasonacox/chatbot

diff --git a/chatbot/run.sh b/chatbot/run.sh
@@ -51,7 +51,8 @@ docker run \
     -e MAXTOKENS=16384 \
     -e TEMPERATURE=0.0 \
     -e QDRANT_HOST="localhost" \
-    -v ./prompts.json:/app/prompts.json \
+    -v $PWD/prompts.json:/app/prompts.json \
+    --network="host" \
     --name chatbot \
     --restart unless-stopped \
     jasonacox/chatbot

diff --git a/llmserver/Dockerfile b/llmserver/Dockerfile
@@ -54,7 +54,7 @@ EXPOSE $PORT
 #     --runtime=nvidia --gpus all \
 #     -d \
 #     -p 8000:8000 \
-#     -v ./models:/app/models \
+#     -v $PWD/models:/app/models \
 #     -e MODEL=models/llama-2-7b-chat.Q5_K_M.gguf \
 #     -e N_GPU_LAYERS=32 \
 #     -e HOST=0.0.0.0 \

diff --git a/llmserver/README.md b/llmserver/README.md
@@ -35,7 +35,7 @@ python3 -m llama_cpp.server \
     --model ./models/mistral-7b-instruct-v0.1.Q5_K_M.gguf \
     --host localhost \
     --n_gpu_layers 99 \
-    -n_ctx 2048 \
+    --n_ctx 2048 \
     --chat_format llama-2
 
 ```
@@ -119,7 +119,7 @@ docker run \
     --runtime=nvidia --gpus all \
     -d \
     -p 8000:8000 \
-    -v ./models:/app/models \
+    -v $PWD/models:/app/models \
     -e MODEL=models/llama-2-7b-chat.Q5_K_M.gguf \
     -e N_GPU_LAYERS=32 \
     -e HOST=0.0.0.0 \

diff --git a/llmserver/models/Dockerfile b/llmserver/models/Dockerfile
@@ -56,7 +56,7 @@ EXPOSE $PORT
 #    docker run -d \
 #     --runtime=nvidia --gpus all \
 #     -p $PORT:$PORT \
-#     -v .:/app/models \
+#     -v $PWD:/app/models \
 #     -e MODEL=$MODEL \
 #     -e N_CTX=$CONTEXT_SIZE \
 #     -e CHAT_FORMAT=$CHAT_FORMAT \

diff --git a/llmserver/models/run.sh b/llmserver/models/run.sh
@@ -80,7 +80,7 @@ INT_REQ=${INT_REQ:-$DEFAULT_INT_REQ}
 docker run -d \
     --runtime=nvidia --gpus all \
     -p $PORT:$PORT \
-    -v .:/app/models \
+    -v $PWD:/app/models \
     -e MODEL=$MODEL \
     -e N_CTX=$CONTEXT_SIZE \
     -e CHAT_FORMAT=$CHAT_FORMAT \

diff --git a/llmserver/setup-docker.sh b/llmserver/setup-docker.sh
@@ -23,7 +23,7 @@ if [[ $start_server == "y" ]]; then
     --runtime=nvidia --gpus all \
     -d \
     -p 8000:8000 \
-    -v ./models:/app/models \
+    -v $PWD/models:/app/models \
     -e MODEL=models/llama-2-7b-chat.Q5_K_M.gguf \
     -e N_GPU_LAYERS=32 \
     -e HOST=0.0.0.0 \

diff --git a/rag/README.md b/rag/README.md
@@ -116,8 +116,8 @@ docker run \
     -e QDRANT_HOST="localhost" \
     -e RESULTS=1 \
     -e SENTENCE_TRANSFORMERS_HOME=/app/models \
-    -v ./models:/app/models \
-    -v ./prompts.json:/app/prompts.json \
+    -v $PWD/models:/app/models \
+    -v $PWD/prompts.json:/app/prompts.json \
     --name chatbot \
     --restart unless-stopped \
     jasonacox/chatbot:latest-rag

diff --git a/rag/run.sh b/rag/run.sh
@@ -52,8 +52,8 @@ docker run \
     -e QDRANT_HOST="localhost" \
     -e RESULTS=1 \
     -e SENTENCE_TRANSFORMERS_HOME=/app/models \
-    -v ./models:/app/models \
-    -v ./prompts.json:/app/prompts.json \
+    -v $PWD/models:/app/models \
+    -v $PWD/prompts.json:/app/prompts.json \
     --name chatbot \
     --restart unless-stopped \
     jasonacox/chatbot:latest-rag

diff --git a/vllm/README.md b/vllm/README.md
@@ -141,7 +141,7 @@ nvidia-docker run -d -p 8000:8000 --gpus=all --shm-size=10.24gb \
   -e HF_HOME=/app/models \
   -e NUM_GPU=4 \
   -e EXTRA_ARGS="--dtype float --max-model-len 20000" \
-  -v models:/app/models \
+  -v $PWD/models:/app/models \
   --name vllm \
   vllm 
 # Print Running Logs - ^C to Stop Viewing Logs

diff --git a/vllm/run-awq.sh b/vllm/run-awq.sh
@@ -30,7 +30,7 @@ docker run -d \
   -e NUM_GPU=1 \
   -e SERVED_MODEL_NAME=tinyllm \
   -e HF_HOME=/app/models \
-  -v ./models:/app/models \
+  -v $PWD/models:/app/models \
   --restart unless-stopped \
   --name $CONTAINER \
   vllm

diff --git a/vllm/run-pascal.sh b/vllm/run-pascal.sh
@@ -28,7 +28,7 @@ docker run -d \
   -e NUM_GPU=1 \
   -e SERVED_MODEL_NAME=tinyllm \
   -e HF_HOME=/app/models \
-  -v ./models:/app/models \
+  -v $PWD/models:/app/models \
   --restart unless-stopped \
   --name $CONTAINER \
   vllm

diff --git a/vllm/run.sh b/vllm/run.sh
@@ -46,7 +46,7 @@ docker run -d \
   -e NUM_GPU=1 \
   -e SERVED_MODEL_NAME=tinyllm \
   -e HF_HOME=/app/models \
-  -v ./models:/app/models \
+  -v $PWD/models:/app/models \
   --restart unless-stopped \
   --name $CONTAINER \
   vllm