flan

Dockerfile

@@ -4,7 +4,7 @@ FROM python:3.10.9
 WORKDIR /
 
 # Copy the current directory contents into the container at .
-COPY . .
+COPY ./requirements.txt /requirements.txt
 
 RUN pip install transformers
 
@@ -17,8 +17,8 @@ RUN pip install torch
 RUN pip install accelerate
 
 # Install hugging face hub to download llama2 model
-RUN pip install --upgrade huggingface_hub
-RUN CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install 'llama-cpp-python[server]' --upgrade --force-reinstall --no-cache-dir 
+#RUN pip install --upgrade huggingface_hub
+#RUN CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install 'llama-cpp-python[server]' --upgrade --force-reinstall --no-cache-dir 
 # Install requirements.txt 
 RUN pip install --no-cache-dir --upgrade -r /requirements.txt
 
@@ -32,7 +32,7 @@ ENV HOME=/home/user \
 
 WORKDIR $HOME/app
 
-RUN huggingface-cli download TheBloke/Llama-2-7b-Chat-GGUF llama-2-7b-chat.Q2_K.gguf --local-dir . --local-dir-use-symlinks False
+#RUN huggingface-cli download TheBloke/Llama-2-7b-Chat-GGUF llama-2-7b-chat.Q2_K.gguf --local-dir . --local-dir-use-symlinks False
 
 
 COPY --chown=user . $HOME/app

question_paper.py

@@ -1,10 +1,6 @@
-from fastapi import FastAPI
-from fastapi.encoders import jsonable_encoder
-from fastapi.responses import JSONResponse
+from transformers import pipeline
 
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-from fastapi.middleware.cors import CORSMiddleware
+pipe_flan = pipeline("text2text-generation", model="google/flan-t5-small")
 
 app = FastAPI()
 
@@ -30,7 +26,9 @@ def llama():
 
     # tresponse = TOKENIZER.decode(tokens[0], skip_special_tokens=False)
     # print(tresponse)
-    response_message = {"message": "hello world"}
+    input = "Generate 5 jokes"
+    output = pipe_flan(input)
+    response_message = {"message": output[0]["generated_text"]}
     json_response = jsonable_encoder(response_message)
     return JSONResponse(content=json_response)