ci: add Ruff config, pre-commit hooks, and GitHub Actions lint workflow

.github/workflows/lint.yml

@@ -0,0 +1,17 @@
+name: Lint
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+
+jobs:
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v5
+      - run: uv python install 3.10
+      - run: uv pip install ruff
+      - run: uv run ruff check .
+      - run: uv run ruff format --check .

.pre-commit-config.yaml

@@ -0,0 +1,7 @@
+repos:
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.12.2
+    hooks:
+      - id: ruff
+        args: [--fix]
+      - id: ruff-format

agents/game_agent/app.py

@@ -5,15 +5,15 @@
 # Jianwei Yang (jianwyan@microsoft.com)
 # --------------------------------------------------------
 
-import pygame
-import numpy as np
+import random
+import re
+
 import gradio as gr
-import time
+import numpy as np
+import pygame
 import torch
 from PIL import Image
 from transformers import AutoModelForCausalLM, AutoProcessor
-import re
-import random
 
 pygame.mixer.quit()  # Disable sound
 

agents/libero/eval_magma_libero.py

@@ -1,22 +1,13 @@
 import os
-import numpy as np
-import draccus
 from dataclasses import dataclass
-from typing import Optional, Tuple
+from typing import Tuple
+
+import draccus
 import tqdm
 from libero.libero import benchmark
-from libero_env_utils import (
-    get_libero_env, 
-    get_libero_dummy_action,
-    get_libero_obs,
-    get_max_steps,
-    set_seed_everywhere
-)
-from libero_magma_utils import (
-    get_magma_model,
-    get_magma_prompt,
-    get_magma_action
-)
+from libero_env_utils import get_libero_dummy_action, get_libero_env, get_libero_obs, get_max_steps, set_seed_everywhere
+from libero_magma_utils import get_magma_action, get_magma_model, get_magma_prompt
+
 
 @dataclass
 class LiberoConfig:

agents/libero/libero_env_utils.py

@@ -2,14 +2,16 @@
 
 import math
 import os
-import torch
 import random
-from PIL import Image
+
 import imageio
 import numpy as np
 import tensorflow as tf
+import torch
 from libero.libero import get_libero_path
 from libero.libero.envs import OffScreenRenderEnv
+from PIL import Image
+
 
 def resize_image(img, resize_size):
     """
@@ -91,7 +93,7 @@ def quat2axisangle(quat):
 
 def save_rollout_video(replay_images, success, task_description):
     """Saves a video replay of a rollout in libero."""
-    save_dir = f"./libero_videos"
+    save_dir = "./libero_videos"
     os.makedirs(save_dir, exist_ok=True)
     processed_task_description = task_description.lower().replace(" ", "_").replace("\n", "_").replace(".", "_")[:50]
     video_path = f"{save_dir}/quick_eval-success={success}--task={processed_task_description}.mp4"

agents/libero/libero_magma_utils.py

@@ -1,10 +1,12 @@
-import os
 import json
-import torch
+import os
+
 import numpy as np
-from magma.image_processing_magma import MagmaImageProcessor
-from magma.processing_magma import MagmaProcessor
+import torch
+
 from magma.modeling_magma import MagmaForConditionalGeneration
+from magma.processing_magma import MagmaProcessor
+
 
 def get_magma_model(model_name):
     processor = MagmaProcessor.from_pretrained(model_name, trust_remote_code=True) 

agents/robot_traj/app.py

@@ -5,17 +5,14 @@
 # Jianwei Yang (jianwyan@microsoft.com)
 # --------------------------------------------------------
 
-import os
-import warnings
-from utils.visualizer import Visualizer
-from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple
+import ast
 import random
-import gradio as gr
-import ast, re
 
+import gradio as gr
 import torch
 import torchvision
 from transformers import AutoModelForCausalLM, AutoProcessor
+from utils.visualizer import Visualizer
 
 '''
 build model

agents/robot_traj/app.pyi

@@ -5,17 +5,15 @@
 # Jianwei Yang (jianwyan@microsoft.com)
 # --------------------------------------------------------
 
-import os
-import warnings
-from utils.visualizer import Visualizer
-from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple
+import ast
 import random
-import gradio as gr
-import ast, re
+from typing import TYPE_CHECKING
 
+import gradio as gr
 import torch
 import torchvision
 from transformers import AutoModelForCausalLM, AutoProcessor
+from utils.visualizer import Visualizer
 
 '''
 build model
@@ -132,7 +130,6 @@ def inference(image, task, *args, **kwargs):
     except Exception as e:
         print(e)
         return None
-from gradio.events import Dependency
 
 class ImageMask(gr.components.Image):
     """
@@ -146,7 +143,8 @@ class ImageMask(gr.components.Image):
 
     def preprocess(self, x):
         return super().preprocess(x)
-    from typing import Callable, Literal, Sequence, Any, TYPE_CHECKING
+    from typing import Literal, Sequence
+
     from gradio.blocks import Block
     if TYPE_CHECKING:
         from gradio.components import Timer
@@ -163,7 +161,8 @@ class Video(gr.components.Video):
 
     def preprocess(self, x):
         return super().preprocess(x)
-    from typing import Callable, Literal, Sequence, Any, TYPE_CHECKING
+    from typing import Literal, Sequence
+
     from gradio.blocks import Block
     if TYPE_CHECKING:
         from gradio.components import Timer

agents/robot_traj/utils/visualizer.py

@@ -4,14 +4,14 @@
 # This source code is licensed under the license found in the
 # LICENSE file in the root directory of this source tree.
 import os
-import numpy as np
+
 import imageio
+import matplotlib.pyplot as plt
+import numpy as np
 import torch
-
-from matplotlib import cm
 import torch.nn.functional as F
 import torchvision.transforms as transforms
-import matplotlib.pyplot as plt
+from matplotlib import cm
 from PIL import Image, ImageDraw
 
 

agents/ui_agent/app.py

@@ -5,27 +5,19 @@
 # Jianwei Yang (jianwyan@microsoft.com)
 # --------------------------------------------------------
 
+import base64
+import io
 from typing import Optional
-import spaces
+
 import gradio as gr
-import numpy as np
+import spaces
 import torch
+from huggingface_hub import snapshot_download
 from PIL import Image
-import io
-import re
-
-import base64, os
-from util.utils import check_ocr_box, get_yolo_model, get_caption_model_processor, get_som_labeled_img
+from transformers import AutoModelForCausalLM, AutoProcessor
+from util.process_utils import extract_bbox, extract_mark_id, pred_2_point
 from util.som import MarkHelper, plot_boxes_with_marks, plot_circles_with_marks
-from util.process_utils import pred_2_point, extract_bbox, extract_mark_id
-
-import torch
-from PIL import Image
-
-from huggingface_hub import snapshot_download
-import torch
-from transformers import AutoModelForCausalLM
-from transformers import AutoProcessor 
+from util.utils import check_ocr_box, get_caption_model_processor, get_som_labeled_img, get_yolo_model
 
 # Define repository and local directory
 repo_id = "microsoft/OmniParser-v2.0"  # HF repo

agents/ui_agent/util/box_annotator.py

@@ -1,8 +1,7 @@
-from typing import List, Optional, Union, Tuple
+from typing import List, Optional, Tuple, Union
 
 import cv2
 import numpy as np
-
 from supervision.detection.core import Detections
 from supervision.draw.color import Color, ColorPalette
 

agents/ui_agent/util/omniparser.py

@@ -1,9 +1,13 @@
-from util.utils import get_som_labeled_img, get_caption_model_processor, get_yolo_model, check_ocr_box
-import torch
-from PIL import Image
-import io
 import base64
+import io
 from typing import Dict
+
+import torch
+from PIL import Image
+
+from util.utils import check_ocr_box, get_caption_model_processor, get_som_labeled_img, get_yolo_model
+
+
 class Omniparser(object):
     def __init__(self, config: Dict):
         self.config = config

agents/ui_agent/util/process_utils.py

@@ -1,5 +1,6 @@
 import re
 
+
 # is instruction English
 def is_english_simple(text):
     try:

agents/ui_agent/util/som.py

@@ -1,13 +1,11 @@
-import torch
-from ultralytics import YOLO
 from PIL import Image
-import io
-import base64
+
 device = 'cuda'
 
-from PIL import Image, ImageDraw, ImageFont
-import numpy as np
 import networkx as nx
+import numpy as np
+from PIL import ImageDraw, ImageFont
+
 # import cv2
 
 font_path = "agents/ui_agent/util/arial.ttf"

agents/ui_agent/util/utils.py

@@ -1,23 +1,18 @@
 # from ultralytics import YOLO
-import os
-import io
 import base64
+import io
 import time
-from PIL import Image, ImageDraw, ImageFont
-import json
-import requests
-# utility function
-import os
 
-import json
-import sys
-import os
+# utility function
 import cv2
+import easyocr
 import numpy as np
+
 # %matplotlib inline
 from matplotlib import pyplot as plt
-import easyocr
 from paddleocr import PaddleOCR
+from PIL import Image
+
 reader = easyocr.Reader(['en'])
 paddle_ocr = PaddleOCR(
     lang='en',  # other lang also available
@@ -28,26 +23,23 @@
     use_dilation=True,  # improves accuracy
     det_db_score_mode='slow',  # improves accuracy
     rec_batch_num=1024)
-import time
-import base64
 
-import os
-import ast
+from typing import List, Tuple, Union
+
+import supervision as sv
 import torch
-from typing import Tuple, List, Union
+import torchvision.transforms as T
 from torchvision.ops import box_convert
-import re
 from torchvision.transforms import ToPILImage
-import supervision as sv
-import torchvision.transforms as T
-from util.box_annotator import BoxAnnotator 
+
+from util.box_annotator import BoxAnnotator
 
 
 def get_caption_model_processor(model_name, model_name_or_path="Salesforce/blip2-opt-2.7b", device=None):
     if not device:
         device = "cuda" if torch.cuda.is_available() else "cpu"
     if model_name == "blip2":
-        from transformers import Blip2Processor, Blip2ForConditionalGeneration
+        from transformers import Blip2ForConditionalGeneration, Blip2Processor
         processor = Blip2Processor.from_pretrained("Salesforce/blip2-opt-2.7b")
         if device == 'cpu':
             model = Blip2ForConditionalGeneration.from_pretrained(
@@ -58,7 +50,7 @@ def get_caption_model_processor(model_name, model_name_or_path="Salesforce/blip2
             model_name_or_path, device_map=None, torch_dtype=torch.float16
         ).to(device)
     elif model_name == "florence2":
-        from transformers import AutoProcessor, AutoModelForCausalLM 
+        from transformers import AutoModelForCausalLM, AutoProcessor
         processor = AutoProcessor.from_pretrained("microsoft/Florence-2-base", trust_remote_code=True)
         if device == 'cpu':
             model = AutoModelForCausalLM.from_pretrained(model_name_or_path, torch_dtype=torch.float32, trust_remote_code=True)

data/__init__.py

@@ -1,15 +1,13 @@
 # datasets
-from .epic import epic
+# data collators
+from .data_collator import DataCollatorForHFDataset, DataCollatorForSupervisedDataset
+
+# (joint) datasets
+from .dataset import build_joint_dataset
 from .ego4d import ego4d
+from .epic import epic
+from .llava import llava
+from .magma import magma
 from .openx import openx
 from .openx_magma import openx_magma
-from .magma import magma
-from .llava import llava
 from .seeclick import seeclick
-
-# (joint) datasets
-from .dataset import build_joint_dataset
-
-# data collators
-from .data_collator import DataCollatorForSupervisedDataset
-from .data_collator import DataCollatorForHFDataset