From 9d4ffa43bcdf9da9126e198c46974cd7a3edf2b4 Mon Sep 17 00:00:00 2001
From: Glenn Jocher <glenn.jocher@ultralytics.com>
Date: Sat, 13 Jan 2024 19:32:49 +0100
Subject: [PATCH] Use `pathlib` in DOTA ops (#7552)
Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>
---
docker/Dockerfile-cpu | 2 +-
ultralytics/data/split_dota.py | 27 ++++++++--------------
ultralytics/models/sam/modules/encoders.py | 7 +-----
ultralytics/utils/benchmarks.py | 8 ++-----
ultralytics/utils/callbacks/comet.py | 26 ++++-----------------
ultralytics/utils/loss.py | 4 +---
6 files changed, 19 insertions(+), 55 deletions(-)
diff --git a/docker/Dockerfile-cpu b/docker/Dockerfile-cpu
index 31d1ea8c..15b74110 100644
--- a/docker/Dockerfile-cpu
+++ b/docker/Dockerfile-cpu
@@ -32,7 +32,7 @@ RUN pip install --no-cache -e ".[export]" lancedb --extra-index-url https://down
RUN yolo export model=tmp/yolov8n.pt format=edgetpu imgsz=32
RUN yolo export model=tmp/yolov8n.pt format=ncnn imgsz=32
# Requires <= Python 3.10, bug with paddlepaddle==2.5.0 https://github.com/PaddlePaddle/X2Paddle/issues/991
-# RUN pip install --no-cache paddlepaddle==2.4.2 x2paddle
+# RUN pip install --no-cache paddlepaddle>=2.6.0 x2paddle
# Remove exported models
RUN rm -rf tmp
diff --git a/ultralytics/data/split_dota.py b/ultralytics/data/split_dota.py
index 4d57ff97..8a5469b8 100644
--- a/ultralytics/data/split_dota.py
+++ b/ultralytics/data/split_dota.py
@@ -1,7 +1,6 @@
# Ultralytics YOLO 🚀, AGPL-3.0 license
import itertools
-import os
from glob import glob
from math import ceil
from pathlib import Path
@@ -73,9 +72,9 @@ def load_yolo_dota(data_root, split="train"):
- val
"""
assert split in ["train", "val"]
- im_dir = os.path.join(data_root, f"images/{split}")
- assert Path(im_dir).exists(), f"Can't find {im_dir}, please check your data root."
- im_files = glob(os.path.join(data_root, f"images/{split}/*"))
+ im_dir = Path(data_root) / "images" / split
+ assert im_dir.exists(), f"Can't find {im_dir}, please check your data root."
+ im_files = glob(str(Path(data_root) / "images" / split / "*"))
lb_files = img2label_paths(im_files)
annos = []
for im_file, lb_file in zip(im_files, lb_files):
@@ -94,7 +93,7 @@ def get_windows(im_size, crop_sizes=[1024], gaps=[200], im_rate_thr=0.6, eps=0.0
Args:
im_size (tuple): Original image size, (h, w).
crop_sizes (List(int)): Crop size of windows.
- gaps (List(int)): Gap between each crops.
+ gaps (List(int)): Gap between crops.
im_rate_thr (float): Threshold of windows areas divided by image ares.
"""
h, w = im_size
@@ -173,7 +172,7 @@ def crop_and_save(anno, windows, window_objs, im_dir, lb_dir):
patch_im = im[y_start:y_stop, x_start:x_stop]
ph, pw = patch_im.shape[:2]
- cv2.imwrite(os.path.join(im_dir, f"{new_name}.jpg"), patch_im)
+ cv2.imwrite(str(Path(im_dir) / f"{new_name}.jpg"), patch_im)
label = window_objs[i]
if len(label) == 0:
continue
@@ -182,7 +181,7 @@ def crop_and_save(anno, windows, window_objs, im_dir, lb_dir):
label[:, 1::2] /= pw
label[:, 2::2] /= ph
- with open(os.path.join(lb_dir, f"{new_name}.txt"), "w") as f:
+ with open(Path(lb_dir) / f"{new_name}.txt", "w") as f:
for lb in label:
formatted_coords = ["{:.6g}".format(coord) for coord in lb[1:]]
f.write(f"{int(lb[0])} {' '.join(formatted_coords)}\n")
@@ -269,7 +268,7 @@ def split_test(data_root, save_dir, crop_size=1024, gap=200, rates=[1.0]):
save_dir = Path(save_dir) / "images" / "test"
save_dir.mkdir(parents=True, exist_ok=True)
- im_dir = Path(os.path.join(data_root, "images/test"))
+ im_dir = Path(data_root) / "images" / "test"
assert im_dir.exists(), f"Can't find {im_dir}, please check your data root."
im_files = glob(str(im_dir / "*"))
for im_file in tqdm(im_files, total=len(im_files), desc="test"):
@@ -281,15 +280,9 @@ def split_test(data_root, save_dir, crop_size=1024, gap=200, rates=[1.0]):
x_start, y_start, x_stop, y_stop = window.tolist()
new_name = f"{name}__{x_stop - x_start}__{x_start}___{y_start}"
patch_im = im[y_start:y_stop, x_start:x_stop]
- cv2.imwrite(os.path.join(str(save_dir), f"{new_name}.jpg"), patch_im)
+ cv2.imwrite(str(save_dir / f"{new_name}.jpg"), patch_im)
if __name__ == "__main__":
- split_trainval(
- data_root="DOTAv2",
- save_dir="DOTAv2-split",
- )
- split_test(
- data_root="DOTAv2",
- save_dir="DOTAv2-split",
- )
+ split_trainval(data_root="DOTAv2", save_dir="DOTAv2-split")
+ split_test(data_root="DOTAv2", save_dir="DOTAv2-split")
diff --git a/ultralytics/models/sam/modules/encoders.py b/ultralytics/models/sam/modules/encoders.py
index a43e115c..a51c3472 100644
--- a/ultralytics/models/sam/modules/encoders.py
+++ b/ultralytics/models/sam/modules/encoders.py
@@ -198,12 +198,7 @@ class PromptEncoder(nn.Module):
"""
return self.pe_layer(self.image_embedding_size).unsqueeze(0)
- def _embed_points(
- self,
- points: torch.Tensor,
- labels: torch.Tensor,
- pad: bool,
- ) -> torch.Tensor:
+ def _embed_points(self, points: torch.Tensor, labels: torch.Tensor, pad: bool) -> torch.Tensor:
"""Embeds point prompts."""
points = points + 0.5 # Shift to center of pixel
if pad:
diff --git a/ultralytics/utils/benchmarks.py b/ultralytics/utils/benchmarks.py
index 7dcc23a5..9df3d554 100644
--- a/ultralytics/utils/benchmarks.py
+++ b/ultralytics/utils/benchmarks.py
@@ -84,12 +84,8 @@ def benchmark(
emoji, filename = "❌", None # export defaults
try:
assert i != 9 or LINUX, "Edge TPU export only supported on Linux"
- if i == 5:
- assert MACOS or LINUX, "CoreML export only supported on macOS and Linux"
- elif i == 10:
- assert MACOS or LINUX, "TF.js export only supported on macOS and Linux"
- # elif i == 11:
- # assert sys.version_info < (3, 11), "PaddlePaddle export only supported on Python<=3.10"
+ if i in {5, 10}: # CoreML and TF.js
+ assert MACOS or LINUX, "export only supported on macOS and Linux"
if "cpu" in device.type:
assert cpu, "inference not supported on CPU"
if "cuda" in device.type:
diff --git a/ultralytics/utils/callbacks/comet.py b/ultralytics/utils/callbacks/comet.py
index 2ed60577..1c5f585d 100644
--- a/ultralytics/utils/callbacks/comet.py
+++ b/ultralytics/utils/callbacks/comet.py
@@ -105,12 +105,7 @@ def _fetch_trainer_metadata(trainer):
save_interval = curr_epoch % save_period == 0
save_assets = save and save_period > 0 and save_interval and not final_epoch
- return dict(
- curr_epoch=curr_epoch,
- curr_step=curr_step,
- save_assets=save_assets,
- final_epoch=final_epoch,
- )
+ return dict(curr_epoch=curr_epoch, curr_step=curr_step, save_assets=save_assets, final_epoch=final_epoch)
def _scale_bounding_box_to_original_image_shape(box, resized_image_shape, original_image_shape, ratio_pad):
@@ -218,11 +213,7 @@ def _log_confusion_matrix(experiment, trainer, curr_step, curr_epoch):
conf_mat = trainer.validator.confusion_matrix.matrix
names = list(trainer.data["names"].values()) + ["background"]
experiment.log_confusion_matrix(
- matrix=conf_mat,
- labels=names,
- max_categories=len(names),
- epoch=curr_epoch,
- step=curr_step,
+ matrix=conf_mat, labels=names, max_categories=len(names), epoch=curr_epoch, step=curr_step
)
@@ -294,12 +285,7 @@ def _log_plots(experiment, trainer):
def _log_model(experiment, trainer):
"""Log the best-trained model to Comet.ml."""
model_name = _get_comet_model_name()
- experiment.log_model(
- model_name,
- file_or_folder=str(trainer.best),
- file_name="best.pt",
- overwrite=True,
- )
+ experiment.log_model(model_name, file_or_folder=str(trainer.best), file_name="best.pt", overwrite=True)
def on_pretrain_routine_start(trainer):
@@ -320,11 +306,7 @@ def on_train_epoch_end(trainer):
curr_epoch = metadata["curr_epoch"]
curr_step = metadata["curr_step"]
- experiment.log_metrics(
- trainer.label_loss_items(trainer.tloss, prefix="train"),
- step=curr_step,
- epoch=curr_epoch,
- )
+ experiment.log_metrics(trainer.label_loss_items(trainer.tloss, prefix="train"), step=curr_step, epoch=curr_epoch)
if curr_epoch == 1:
_log_images(experiment, trainer.save_dir.glob("train_batch*.jpg"), curr_step)
diff --git a/ultralytics/utils/loss.py b/ultralytics/utils/loss.py
index 03e112fe..c4fe6a85 100644
--- a/ultralytics/utils/loss.py
+++ b/ultralytics/utils/loss.py
@@ -38,9 +38,7 @@ class VarifocalLoss(nn.Module):
class FocalLoss(nn.Module):
"""Wraps focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5)."""
- def __init__(
- self,
- ):
+ def __init__(self):
"""Initializer for FocalLoss class with no parameters."""
super().__init__()