From 6e57cab13f296b2f84ab09b6f77a154b27cf3092 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?D=C3=A1vid=20Omasta?= <david.omasta@student.tuke.sk>
Date: Sat, 17 Feb 2024 19:00:00 +0000
Subject: [PATCH] Upload files to 'Backend'

update backend
---
 Backend/Dockerfile       | 55 ++++++++++++++++++++++++++++++++++++++
 Backend/api.py           | 57 ++++++++++++++++++++++++++++++++++++++++
 Backend/requirements.txt |  5 ++++
 3 files changed, 117 insertions(+)
 create mode 100644 Backend/Dockerfile
 create mode 100644 Backend/api.py
 create mode 100644 Backend/requirements.txt

diff --git a/Backend/Dockerfile b/Backend/Dockerfile
new file mode 100644
index 0000000..0d8a369
--- /dev/null
+++ b/Backend/Dockerfile
@@ -0,0 +1,55 @@
+FROM python:3.10-slim-bullseye AS base
+
+WORKDIR /app
+
+# Set utf-8 encoding for Python et al
+ENV LANG=C.UTF-8 \
+	# Turn off writing .pyc files
+	PYTHONDONTWRITEBYTECODE=1 \
+	# Reduce the OS system calls for this tool it makes a difference
+	PYTHONUNBUFFERED=1 \
+	# Disables cache dir in pip
+	PIP_NO_CACHE_DIR=1 \
+	# Virtual environment
+	VENV="/opt/venv" \
+	# Add new user
+	APPUSER=appuser \
+	# Ensure that the python and pip executables used in the image
+	PATH="${VENV}/bin:$PATH"
+
+
+FROM base as builder
+
+COPY requirements.txt .
+RUN apt-get update \
+	&& apt-get install -y git build-essential
+
+
+RUN python -m venv ${VENV} \
+	&& . ${VENV}/bin/activate \
+    && pip install --upgrade pip \
+	&& pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu \
+    && pip install -r requirements.txt
+
+FROM base as runner
+
+
+COPY api.py .
+
+COPY --from=builder ${VENV} ${VENV}
+ENV PATH="${VENV}/bin:$PATH"
+
+# Update permissions & change user to not run as root
+RUN chgrp -R 0 /app \
+    && chmod -R g=u /app \
+    && groupadd -r ${APPUSER} \
+    && useradd -r -g ${APPUSER} ${APPUSER} \
+    && chown -R ${APPUSER}:${APPUSER} /app \
+    && usermod -d /app ${APPUSER}
+
+
+CMD ["/opt/venv/bin/uvicorn", "api:app", "--host", "0.0.0.0"]
+
+
+
+
diff --git a/Backend/api.py b/Backend/api.py
new file mode 100644
index 0000000..e1ea552
--- /dev/null
+++ b/Backend/api.py
@@ -0,0 +1,57 @@
+import torch
+import uvicorn
+from fastapi import FastAPI
+from pydantic import BaseModel
+from transformers import  MT5Tokenizer,AutoTokenizer, AutoModel ,T5ForConditionalGeneration
+import warnings
+import json
+import random
+import torch.nn.functional as F
+import os
+from dotenv import load_dotenv
+#from ece import compute_ECE
+from torch.utils.data import DataLoader
+from functools import reduce
+
+warnings.filterwarnings("ignore")
+DEVICE ='cpu'
+
+load_dotenv()
+host = os.getenv("HOST")
+port = os.getenv("PORT")
+
+model_dir = os.getenv("QA_MODEL")
+#model_dir = "C:/Users/david/Desktop/T5_JUPYTER/qa_model"
+tokenizer_dir = os.getenv("QA_TOKENIZER")
+#tokenizer_dir = "C:/Users/david/Desktop/T5_JUPYTER/qa_tokenizer"
+MODEL = T5ForConditionalGeneration.from_pretrained(model_dir, from_tf=False, return_dict=True).to(DEVICE)
+print("Model succesfully loaded!")
+TOKENIZER = AutoTokenizer.from_pretrained(tokenizer_dir, use_fast=True)
+print("Tokenizer succesfully loaded!")
+Q_LEN = 512
+TOKENIZER.add_tokens('<sep>')
+
+
+print('model loaded')
+
+app = FastAPI()
+
+# BASE MODEL
+class InputData(BaseModel):
+    context: str
+    question: str
+
+@app.post("/predict")
+async def predict(input_data: InputData):
+    inputs = TOKENIZER(input_data.question, input_data.context, max_length=512, padding="max_length", truncation=True, add_special_tokens=True)
+    input_ids = torch.tensor(inputs["input_ids"], dtype=torch.long).to(DEVICE).unsqueeze(0)
+    attention_mask = torch.tensor(inputs["attention_mask"], dtype=torch.long).to(DEVICE).unsqueeze(0)
+    outputs = MODEL.generate(input_ids=input_ids, attention_mask=attention_mask, return_dict_in_generate=True,output_scores=True,max_length=512)
+    predicted_ids = outputs.sequences.numpy()
+    predicted_text = TOKENIZER.decode(predicted_ids[0], skip_special_tokens=True)
+
+    return {'prediction':predicted_text}
+
+
+if __name__ == "__main__":
+    uvicorn.run(app, host=host, port=port)
\ No newline at end of file
diff --git a/Backend/requirements.txt b/Backend/requirements.txt
new file mode 100644
index 0000000..28a207c
--- /dev/null
+++ b/Backend/requirements.txt
@@ -0,0 +1,5 @@
+uvicorn==0.23.2
+fastapi==0.103.2
+transformers==4.34.0
+rank_bm25==0.2.2
+python-dotenv