first pass at ruff linter fixes

.pre-commit-config.yaml

@@ -10,6 +10,10 @@ repos:
  - repo: https://github.com/astral-sh/ruff-pre-commit
  rev: v0.5.2
  hooks:
+ - id: ruff
+ types_or: [ python, pyi, jupyter ]
+ args: [ --fix ]
+
  - id: ruff-format
  args: [ --exclude, kitti.ipynb ]
  types_or: [ python, pyi, jupyter ]

b3d/__init__.py

@@ -7,7 +7,7 @@
 from .pose import Pose, Rot
 from . import camera, colors, pose, types, utils
 from . import renderer, io, bayes3d, chisight
-from .renderer import Renderer, RendererOriginal
+from .renderer import RendererOriginal
 
 __version__ = metadata.version("genjax")
 __all__ = [

b3d/bayes3d/enumerative_proposals.py

@@ -2,7 +2,6 @@
 import jax.numpy as jnp
 import b3d
 from b3d.pose import Pose
-import genjax
 from genjax import Pytree
 
 

b3d/bayes3d/model.py

@@ -1,5 +1,4 @@
 import genjax
-from genjax.generative_functions.distributions import ExactDensity
 import jax.numpy as jnp
 import b3d
 from b3d.pose import Pose

b3d/bucket_utils/b3d_pull.py

@@ -1,10 +1,8 @@
 # gcloud storage cp --recursive gs://hgps_data_bucket/shared .
 import argparse
-import json
 import os
 import subprocess
 import b3d
-from pathlib import Path
 
 
 ## Paths.

b3d/bucket_utils/b3d_push.py

@@ -1,9 +1,6 @@
 import argparse
-import json
-import os
 import subprocess
 import b3d
-from pathlib import Path
 
 
 ## Paths.

b3d/chisight/__init__.py

@@ -1,2 +0,0 @@
-from importlib import metadata
-from . import particle_system, dense, sparse

b3d/chisight/dense/dense_model.py

@@ -1,8 +1,7 @@
-from b3d.modeling_utils import uniform_discrete, uniform_pose, gaussian_vmf
+from b3d.modeling_utils import uniform_pose
 import genjax
 import b3d
 from b3d import Pose, Mesh
-import jax
 import jax.numpy as jnp
 import b3d.chisight.dense.likelihoods.image_likelihood
 

b3d/chisight/dense/likelihoods/image_likelihood.py

@@ -1,12 +1,4 @@
 import genjax
-from genjax.generative_functions.distributions import ExactDensity
-import jax.numpy as jnp
-import b3d
-from b3d import Mesh, Pose
-from collections import namedtuple
-from b3d.modeling_utils import uniform_discrete, uniform_pose
-import jax
-import os
 
 from genjax import Pytree
 

b3d/chisight/dense/likelihoods/kray_outlier_volume.py

@@ -1,14 +1,5 @@
-import genjax
-from genjax.generative_functions.distributions import ExactDensity
 import jax.numpy as jnp
 import b3d
-from b3d import Mesh, Pose
-from collections import namedtuple
-from b3d.modeling_utils import uniform_discrete, uniform_pose
-import jax
-import os
-
-from genjax import Pytree
 
 
 def get_rgb_depth_inliers_from_observed_rendered_args(

b3d/chisight/dense/likelihoods/krays.py

@@ -1,12 +1,7 @@
 import genjax
-from genjax.generative_functions.distributions import ExactDensity
 import jax.numpy as jnp
 import b3d
-from b3d import Mesh, Pose
 from collections import namedtuple
-from b3d.modeling_utils import uniform_discrete, uniform_pose
-import jax
-import os
 
 from genjax import Pytree
 

b3d/chisight/dense/likelihoods/simple_likelihood.py

@@ -7,7 +7,6 @@
 def simple_likelihood(observed_rgbd, rendered_rgbd, likelihood_args):
  fx = likelihood_args["fx"]
  fy = likelihood_args["fy"]
- far = likelihood_args["far"]
 
  rendered_rgb = rendered_rgbd[..., :3]
  observed_rgb = observed_rgbd[..., :3]

b3d/chisight/particle_system.py

@@ -2,10 +2,9 @@
 import b3d
 from b3d import Pose
 import jax
-import jax.numpy as jnp
 import genjax
 from genjax import gen
-from b3d import Pose, Mesh
+from b3d import Mesh
 from b3d.chisight.sparse.gps_utils import add_dummy_var
 from b3d.pose import uniform_pose_in_ball
 
@@ -244,7 +243,6 @@ def visualize_particle_system(
  camera_pose_prior_params,
  ) = latent_particle_model_args
 
- colors = b3d.distinct_colors(num_clusters.const)
  absolute_particle_poses = particle_dynamics_summary["absolute_particle_poses"]
  object_poses = particle_dynamics_summary["object_poses"]
  camera_pose = particle_dynamics_summary["camera_pose"]

b3d/chisight/sparse/gps_utils.py

@@ -13,7 +13,6 @@
 from b3d.pose.pose_utils import (
  uniform_samples_from_disc,
 )
-from .dynamic_gps import DynamicGPS
 from typing import TypeAlias
 
 

b3d/chisight/sparse/sparse_gps_model.py

@@ -5,6 +5,7 @@
 from .dynamic_gps import DynamicGPS
 from typing import Any, TypeAlias
 from b3d.camera import screen_from_world
+from genjax import ChoiceMapBuilder as C
 
 
 @genjax.gen
@@ -108,7 +109,6 @@ def kernel(carried_state, _):
  )
 
  new_state = (t + 1, new_object_poses, new_camera_pose, static_carries)
- vector_output = None
  return (new_state, new_state)
 
  # TODO: What arguments should be passed to the model?
@@ -279,8 +279,6 @@ def get_dynamic_gps(tr: SparseGPSModelTrace):
 # -----------
 # Setters
 # -----------
-from genjax import ChoiceMapBuilder as C
-from genjax._src.core.generative.choice_map import EmptyChm
 
 
 def set_camera_choice(t, cam: Pose, ch=None):

b3d/chisight/sparse/sparse_model_reality_check.py

@@ -1,10 +1,7 @@
-import matplotlib.pyplot as plt
-import genjax
 import jax
-import jax.numpy as jnp
 from b3d.utils import keysplit
-from b3d.camera import Intrinsics, screen_from_camera
-from b3d.pose import Pose, camera_from_position_and_target
+from b3d.camera import Intrinsics
+from b3d.pose import Pose
 from b3d.pose.pose_utils import uniform_pose_in_ball
 from b3d.chisight.sparse.gps_utils import add_dummy_var
 from b3d.chisight.sparse.sparse_gps_model import minimal_observation_model

b3d/io/data_loader.py

@@ -5,11 +5,10 @@
 import cv2
 import imageio
 
-import jax.numpy as jnp
 import numpy as np
 from PIL import Image
 from tqdm import tqdm
-from b3d import Mesh, Pose
+from b3d import Pose
 import glob
 
 YCB_MODEL_NAMES = [
@@ -70,7 +69,6 @@ def get_ycbv_test_images(ycb_dir, scene_id, images_indices, fields=[]):
 
  scene_rgb_images_dir = os.path.join(scene_data_dir, "rgb")
  scene_depth_images_dir = os.path.join(scene_data_dir, "depth")
- mask_visib_dir = os.path.join(scene_data_dir, "mask_visib")
 
  with open(os.path.join(scene_data_dir, "scene_camera.json")) as scene_cam_data_json:
  scene_cam_data = json.load(scene_cam_data_json)
@@ -156,8 +154,6 @@ def get_ycb_mesh(ycb_dir, id):
 
 
 def get_ycbineoat_images(ycbineaot_dir, video_name, images_indices):
- id_strs = []
-
  video_dir = os.path.join(ycbineaot_dir, f"{video_name}")
 
  color_files = sorted(glob.glob(f"{video_dir}/rgb/*.png"))

b3d/io/utils.py

@@ -1,19 +1,12 @@
-import os
 from .video_input import VideoInput
-from b3d.utils import get_shared
 from b3d.types import Array
-from b3d.io import VideoInput, FeatureTrackData
 import numpy as np
-import jax.numpy as jnp
 import matplotlib.pyplot as plt
-import jax
 import cv2
-import numpy as np
 from sklearn.utils import Bunch
 from pathlib import Path
 import argparse
 import inspect
-import sys
 
 
 def add_argparse(f):
@@ -129,7 +122,6 @@ def load_video_to_numpy(
  T = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
  w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
  h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
- fps = int(cap.get(cv2.CAP_PROP_FPS))
 
  if times is None:
  times = np.arange(T, step=step)
@@ -222,9 +214,6 @@ def plot_video_summary(
  reverse_color_channel=reverse_color_channel,
  )
 
- w = vid.shape[2]
- h = vid.shape[1]
-
  # Create a plot with the summary
  # TODO: Should we hand in an axis?
  fig, ax = plt.subplots(1, 1, figsize=(15, 4))

b3d/mesh.py

@@ -1,6 +1,5 @@
 import b3d
 import jax.numpy as jnp
-from b3d import Pose
 import jax
 import trimesh
 from jax.tree_util import register_pytree_node_class

b3d/renderer/renderer.py

@@ -136,7 +136,7 @@ def _rasterize_bwd(self, saved_tensors, diffs):
  )
  dy, ddb = diffs
 
- grads = _rasterize_bwd_custom_call(
+ _rasterize_bwd_custom_call(
  self,
  pose,
  pos,
@@ -458,7 +458,6 @@ def _render_batch(args, axes):
  if pose.ndim != 5:
  raise NotImplementedError("Underlying primitive must operate on 4D poses.")
 
- original_shape = pose.shape
  poses = jnp.moveaxis(pose, axes[0], 0)
  size_1 = poses.shape[0]
  size_2 = poses.shape[1]
@@ -749,9 +748,6 @@ def _interpolate_fwd_lowering(
  def _render_batch_interp(args, axes):
  attributes, uvs, triangle_ids, faces = args
 
- original_shape_uvs = uvs.shape
- original_shape_triangle_ids = triangle_ids.shape
-
  uvs = jnp.moveaxis(uvs, axes[1], 0)
  size_1 = uvs.shape[0]
  size_2 = uvs.shape[1]

b3d/renderer/renderer_original.py

@@ -8,11 +8,8 @@
 from jax.interpreters import batching, mlir, xla
 from jax.lib import xla_client
 from jaxlib.hlo_helpers import custom_call
-import functools
-import os
 import b3d
 import b3d.renderer.nvdiffrast_jax.nvdiffrast.jax as dr
-import rerun as rr
 
 
 def projection_matrix_from_intrinsics(w, h, fx, fy, cx, cy, near, far):
@@ -192,12 +189,12 @@ def _register_custom_calls():
 
 
 # @functools.partial(jax.jit, static_argnums=(0,))
-def _rasterize_fwd_custom_call(r: "Renderer", pos, tri, resolution):
+def _rasterize_fwd_custom_call(r: "RendererOriginal", pos, tri, resolution):
  return _build_rasterize_fwd_primitive(r).bind(pos, tri, resolution)
 
 
 @functools.lru_cache(maxsize=None)
-def _build_rasterize_fwd_primitive(r: "Renderer"):
+def _build_rasterize_fwd_primitive(r: "RendererOriginal"):
  _register_custom_calls()
  # For JIT compilation we need a function to evaluate the shape and dtype of the
  # outputs of our op for some given inputs
@@ -277,7 +274,6 @@ def _rasterize_fwd_lowering(ctx, pos, tri, resolution):
  def _render_batch(args, axes):
  pos, tri, resolution = args
 
- original_shape = pos.shape
  pos_reshaped = pos.reshape(-1, *pos.shape[-2:])
 
  (rendered,) = _rasterize_fwd_custom_call(r, pos_reshaped, tri, resolution)
@@ -304,7 +300,7 @@ def _render_batch(args, axes):
 
 # @functools.partial(jax.jit, static_argnums=(0,))
 def _interpolate_fwd_custom_call(
- r: "Renderer",
+ r: "RendererOriginal",
  attributes,
  rast,
  faces,
@@ -317,7 +313,7 @@ def _interpolate_fwd_custom_call(
 
 
 # @functools.lru_cache(maxsize=None)
-def _build_interpolate_fwd_primitive(r: "Renderer"):
+def _build_interpolate_fwd_primitive(r: "RendererOriginal"):
  _register_custom_calls()
  # For JIT compilation we need a function to evaluate the shape and dtype of the
  # outputs of our op for some given inputs
@@ -390,7 +386,6 @@ def _interpolate_fwd_lowering(
  def _interpolate_batch(args, axes):
  attributes, rast, faces = args
 
- original_shape = attributes.shape
  attributes_reshaped = attributes.reshape(-1, *attributes.shape[-2:])
  rast_reshaped = rast.reshape(-1, *rast.shape[-3:])
 

b3d/renderer/torch/pose.py

@@ -1,3 +0,0 @@
-import torch
-import pytorch3d.transforms
-import pytorch3d.renderer

b3d/renderer/torch/renderutils/ops.py

@@ -8,9 +8,7 @@
 # its affiliates is strictly prohibited.
 
 import os
-import sys
 
-import numpy as np
 import torch
 import torch.utils.cpp_extension
 
@@ -75,7 +73,7 @@ def find_cl_path():
  )
  if os.path.exists(lock_fn):
  print("Warning: Lock file exists in build directory: '%s'" % lock_fn)
- except:
+ except Exception:
  pass
 
  # Compile and load.