Merge pull request #100 from ibm-granite/inference_service_dir_refactor

Inference service directory refactor

.github/workflows/inference-services-test.yml

@@ -1,7 +1,7 @@
 # This workflow will install Python dependencies, run tests and lint with a variety of Python versions
 # For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
 
-name: Inference Service Tests
+name: Tsfminference Service Tests
 
 on:
   push:
@@ -32,7 +32,7 @@ jobs:
         pip install poetry
         cd services/inference
         poetry install -n --with dev
-    - name: Test local server inference service with pytest
+    - name: Test local server tsfminference service with pytest
       run: |
         source .venv/bin/activate
         cd services/inference

services/inference/.dockerignore

@@ -1,5 +1,5 @@
 **
-!tsfmservices
+!tsfminference
 !poetry.lock
 !pyproject.toml
 

services/inference/Dockerfile

@@ -10,10 +10,11 @@ RUN microdnf -y update && \
 
 ENV POETRY_VIRTUALENVS_IN_PROJECT=1
 
-COPY tsfmservices* /tsfmservices/tsfmservices
-COPY pyproject.toml /tsfmservices/
-COPY poetry.lock /tsfmservices/
-WORKDIR /tsfmservices
+RUN mkdir /inference
+COPY tsfminference/* /inference/tsfminference/
+COPY pyproject.toml /inference/
+COPY poetry.lock /inference/
+WORKDIR /inference
 RUN pip3.11 install poetry && poetry install
 
 FROM registry.access.redhat.com/ubi9/ubi-minimal:latest AS deploy
@@ -22,20 +23,19 @@ RUN microdnf -y update && \
         shadow-utils python3.11 && \
     microdnf clean all
 
-WORKDIR /tsfmservices
+WORKDIR /inference
 
-COPY --from=builder /tsfmservices/ /tsfmservices/
+COPY --from=builder /inference /inference
 
-ENV VIRTUAL_ENV=/tsfmservices/.venv
+ENV VIRTUAL_ENV=/inference/.venv
 ENV PATH="$VIRTUAL_ENV/bin:$PATH"
 ENV HF_HOME=/tmp
 
-RUN groupadd --system tsfmservices --gid 1001 && \
-    adduser --system --uid 1001 --gid 0 --groups tsfmservices \
-    --create-home --home-dir /tsfmservices --shell /sbin/nologin \
-    --comment "TSFMServices User" tsfmservices
+RUN groupadd --system tsfminference --gid 1001 && \
+    adduser --system --uid 1001 --gid 0 --groups tsfminference \
+    --create-home --home-dir /inference --shell /sbin/nologin \
+    --comment "tsfminference User" tsfminference
 
-USER tsfmservices
+USER tsfminference
 
-# CMD ["python",  "-m", "tsfmservices.inference.main"]
-CMD ["python", "-m", "uvicorn","tsfmservices.inference.main:app", "--host", "0.0.0.0", "--port", "8000" ]
+CMD ["python", "-m", "uvicorn","tsfminference.main:app", "--host", "0.0.0.0", "--port", "8000" ]

services/inference/Makefile

@@ -1,18 +1,18 @@
 CONTAINER_BUILDER ?= docker
 
-# starts the inference service (used mainly for test cases)
+# starts the tsfminference service (used mainly for test cases)
 start_service_local:
-	python -m tsfmservices.inference.main &
+	python -m tsfminference.main &
 	sleep 10
 stop_service_local:
-	pkill  -f 'python.*tsfmservices.*'
+	pkill  -f 'python.*tsfminference.*'
 	sleep 10
 
 image:
-	$(CONTAINER_BUILDER) build -t tsfmservices -f Dockerfile .
+	$(CONTAINER_BUILDER) build -t tsfminference -f Dockerfile .
 
 start_service_image: image
-	$(CONTAINER_BUILDER) run -p 8000:8000 -d --rm --name tsfmserver tsfmservices
+	$(CONTAINER_BUILDER) run -p 8000:8000 -d --rm --name tsfmserver tsfminference
 	sleep 10
 stop_service_image:
 	$(CONTAINER_BUILDER) stop tsfmserver

services/inference/README.md

@@ -45,11 +45,11 @@ CONTAINER_BUILDER=<docker|podmain> make image
 # e.g, CONTAINER_BUILDER=docker make image
 ```
 
-After a successful build you should have a local image named `tsfmservices:latest`
+After a successful build you should have a local image named `tsfminference:latest`
 
 ```sh
-(py311) ➜  tsfm-services git:(revised-build-system) ✗ docker images | grep tsfmservices | head -n 1
-tsfmservices                                             latest               df592dcb0533   46 seconds ago      1.49GB
+(py311) ➜  tsfm-services git:(revised-build-system) ✗ docker images | grep tsfminference | head -n 1
+tsfminference                                             latest               df592dcb0533   46 seconds ago      1.49GB
 # some of the numeric and hash values on your machine could be different
 ```
 

services/inference/pyproject.toml

@@ -1,18 +1,18 @@
 [tool.poetry]
-name = "tsfmservices"
+name = "tsfminference"
 version = "0.0.1"
 description = "Service layer for TSFM granite models."
 authors = ["IBM"]
 license = "https://github.com/ibm-granite/granite-tsfm/blob/main/LICENSE"
-packages = [{ include = "tsfmservices/**/*.py" }]
+packages = [{ include = "tsfminference/**/*.py" }]
 
 [tool.poetry.dependencies]
 # including 3.9 causes poetry lock to run forever
 python = ">=3.10,<3.12"
 numpy = { version = "<2" }
 tsfm_public = { git = "https://github.com/IBM/tsfm.git", tag = "v0.2.5" }
 
-# trying to pick up cpu version for inference
+# trying to pick up cpu version for tsfminference
 # to make image smaller
 torch = { version = ">2,<3", source = "pytorch" }
 

services/inference/tsfmservices/__init__.py → services/inference/tsfminference/__init__.py

@@ -26,23 +26,9 @@
     "%(levelname)s:p-%(process)d:t-%(thread)d:%(filename)s:%(funcName)s:%(message)s",
 )
 
-# Do we have redis?
-
-try:
-    # Third Party
-    import redis
-    import rq
-
-    HAVE_REDIS_BACKEND = True
-except ImportError:
-    HAVE_REDIS_BACKEND = False
-
-TSFM_USE_KFTO_ASYNC_BACKEND = int(os.getenv("TSFM_USE_KFTO_ASYNC_BACKEND", "0")) == 1
-USE_REDIS_ASYNC_BACKEND = HAVE_REDIS_BACKEND and not TSFM_USE_KFTO_ASYNC_BACKEND
-
 TSFM_ALLOW_LOAD_FROM_HF_HUB = int(os.getenv("TSFM_ALLOW_LOAD_FROM_HF_HUB", "1")) == 1
 
 TSFM_CONFIG_FILE = os.getenv(
     "TSFM_CONFIG_FILE",
-    os.path.realpath(os.path.join(os.path.dirname(__file__), "config", "default_config.yml")),
+    os.path.realpath(os.path.join(os.path.dirname(__file__), "default_config.yml")),
 )

services/inference/tsfmservices/inference/inference.py → services/inference/tsfminference/inference.py

@@ -1,6 +1,6 @@
 # Copyright contributors to the TSFM project
 #
-"""Inference Runtime"""
+"""Tsfminference Runtime"""
 
 import datetime
 import logging
@@ -11,11 +11,11 @@
 from fastapi import APIRouter, HTTPException
 
 from tsfm_public import TimeSeriesForecastingPipeline, TimeSeriesPreprocessor
-from tsfmservices import TSFM_ALLOW_LOAD_FROM_HF_HUB
 
-from ..common.constants import API_VERSION
-from ..common.util import load_config, load_model, register_config
+from . import TSFM_ALLOW_LOAD_FROM_HF_HUB
+from .constants import API_VERSION
 from .payloads import ForecastingInferenceInput, PredictOutput
+from .util import load_config, load_model, register_config
 
 
 LOGGER = logging.getLogger(__file__)

services/inference/tsfmservices/inference/main.py → services/inference/tsfminference/main.py

@@ -7,13 +7,13 @@
 import yaml
 from fastapi import FastAPI
 
-from tsfmservices import (
+from . import (
     TSFM_CONFIG_FILE,
     TSFM_PYTHON_LOGGING_FORMAT,
     TSFM_PYTHON_LOGGING_LEVEL,
 )
-from tsfmservices.common.constants import API_VERSION
-from tsfmservices.inference import InferenceRuntime
+from .constants import API_VERSION
+from .inference import InferenceRuntime
 
 
 logging.basicConfig(

services/inference/tsfmservices/inference/payloads.py → services/inference/tsfminference/payloads.py

@@ -1,6 +1,6 @@
 # Copyright contributors to the TSFM project
 #
-"""Payload definitions for inference"""
+"""Payload definitions for tsfminference"""
 
 from typing import Annotated, Any, Dict, List, Optional