166 lines
5.0 KiB
Python
166 lines
5.0 KiB
Python
"""
|
|
Training routes for gateway
|
|
"""
|
|
|
|
from fastapi import APIRouter, Request, HTTPException, Query
|
|
from fastapi.responses import JSONResponse
|
|
import httpx
|
|
import logging
|
|
from typing import Optional
|
|
|
|
from app.core.config import settings
|
|
|
|
logger = logging.getLogger(__name__)
|
|
router = APIRouter()
|
|
|
|
@router.post("/train")
|
|
async def start_training(request: Request):
|
|
"""Proxy training request to training service"""
|
|
try:
|
|
body = await request.body()
|
|
auth_header = request.headers.get("Authorization")
|
|
|
|
async with httpx.AsyncClient(timeout=30.0) as client:
|
|
response = await client.post(
|
|
f"{settings.TRAINING_SERVICE_URL}/train",
|
|
content=body,
|
|
headers={
|
|
"Content-Type": "application/json",
|
|
"Authorization": auth_header
|
|
}
|
|
)
|
|
|
|
return JSONResponse(
|
|
status_code=response.status_code,
|
|
content=response.json()
|
|
)
|
|
|
|
except httpx.RequestError as e:
|
|
logger.error(f"Training service unavailable: {e}")
|
|
raise HTTPException(
|
|
status_code=503,
|
|
detail="Training service unavailable"
|
|
)
|
|
except Exception as e:
|
|
logger.error(f"Training error: {e}")
|
|
raise HTTPException(status_code=500, detail="Internal server error")
|
|
|
|
@router.get("/status/{training_job_id}")
|
|
async def get_training_status(training_job_id: str, request: Request):
|
|
"""Get training job status"""
|
|
try:
|
|
auth_header = request.headers.get("Authorization")
|
|
|
|
async with httpx.AsyncClient(timeout=10.0) as client:
|
|
response = await client.get(
|
|
f"{settings.TRAINING_SERVICE_URL}/status/{training_job_id}",
|
|
headers={"Authorization": auth_header}
|
|
)
|
|
|
|
return JSONResponse(
|
|
status_code=response.status_code,
|
|
content=response.json()
|
|
)
|
|
|
|
except httpx.RequestError as e:
|
|
logger.error(f"Training service unavailable: {e}")
|
|
raise HTTPException(
|
|
status_code=503,
|
|
detail="Training service unavailable"
|
|
)
|
|
except Exception as e:
|
|
logger.error(f"Training status error: {e}")
|
|
raise HTTPException(status_code=500, detail="Internal server error")
|
|
|
|
@router.get("/models")
|
|
async def get_trained_models(request: Request):
|
|
"""Get trained models"""
|
|
try:
|
|
auth_header = request.headers.get("Authorization")
|
|
|
|
async with httpx.AsyncClient(timeout=10.0) as client:
|
|
response = await client.get(
|
|
f"{settings.TRAINING_SERVICE_URL}/models",
|
|
headers={"Authorization": auth_header}
|
|
)
|
|
|
|
return JSONResponse(
|
|
status_code=response.status_code,
|
|
content=response.json()
|
|
)
|
|
|
|
except httpx.RequestError as e:
|
|
logger.error(f"Training service unavailable: {e}")
|
|
raise HTTPException(
|
|
status_code=503,
|
|
detail="Training service unavailable"
|
|
)
|
|
except Exception as e:
|
|
logger.error(f"Get models error: {e}")
|
|
raise HTTPException(status_code=500, detail="Internal server error")
|
|
|
|
@router.get("/jobs")
|
|
async def get_training_jobs(
|
|
request: Request,
|
|
limit: Optional[int] = Query(10, ge=1, le=100),
|
|
offset: Optional[int] = Query(0, ge=0)
|
|
):
|
|
"""Get training jobs"""
|
|
try:
|
|
auth_header = request.headers.get("Authorization")
|
|
|
|
async with httpx.AsyncClient(timeout=10.0) as client:
|
|
response = await client.get(
|
|
f"{settings.TRAINING_SERVICE_URL}/jobs",
|
|
params={"limit": limit, "offset": offset},
|
|
headers={"Authorization": auth_header}
|
|
)
|
|
|
|
return JSONResponse(
|
|
status_code=response.status_code,
|
|
content=response.json()
|
|
)
|
|
|
|
except httpx.RequestError as e:
|
|
logger.error(f"Training service unavailable: {e}")
|
|
raise HTTPException(
|
|
status_code=503,
|
|
detail="Training service unavailable"
|
|
)
|
|
except Exception as e:
|
|
logger.error(f"Get training jobs error: {e}")
|
|
raise HTTPException(status_code=500, detail="Internal server error")
|
|
|
|
|
|
# gateway/Dockerfile
|
|
FROM python:3.11-slim
|
|
|
|
WORKDIR /app
|
|
|
|
# Install system dependencies
|
|
RUN apt-get update && apt-get install -y \
|
|
gcc \
|
|
curl \
|
|
&& rm -rf /var/lib/apt/lists/*
|
|
|
|
# Copy requirements
|
|
COPY requirements.txt .
|
|
|
|
# Install Python dependencies
|
|
RUN pip install --no-cache-dir -r requirements.txt
|
|
|
|
# Copy application code
|
|
COPY . .
|
|
|
|
# Add shared libraries to Python path
|
|
ENV PYTHONPATH="/app:/app/shared:$PYTHONPATH"
|
|
|
|
# Expose port
|
|
EXPOSE 8000
|
|
|
|
# Health check
|
|
HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
|
|
CMD curl -f http://localhost:8000/health || exit 1
|
|
|
|
# Run application
|
|
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"] |