Skip to content

Updated readme.md #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
tejaswinijayashanker943 wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Updated readme.md #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
254707b
Updated readme.md
tejaswinijayashanker943 Dec 9, 2025
3e6afe3
Auto-format by https://ultralytics.com/actions
UltralyticsAssistant Dec 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
549 changes: 261 additions & 288 deletions data_engine.py
View file
Open in desktop

Large diffs are not rendered by default.

247 changes: 128 additions & 119 deletions data_engine_agent.py
View file
Open in desktop

Large diffs are not rendered by default.

42 changes: 17 additions & 25 deletions data_visual.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,51 +1,43 @@
import ultralytics,os
import os

import ultralytics

workspace = os.path.dirname(os.path.dirname(os.path.abspath(ultralytics.__file__)))
os.chdir(workspace)
print("set workspace:", workspace)


from data_engine import DataEngine





if __name__ == "__main__":


# device="cuda:1"
# de=DataEngine(device=device)
# cache_path="/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.merged.cache"
# text_embed_pt="/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# text_embed_pt=text_embed_pt)

im_index=0

im_index = 0

de=DataEngine()
cache_path="/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.merged.cache"
text_embed_pt="/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path,
data_style="grounding",
text_embed_pt=text_embed_pt)
de = DataEngine()
cache_path = "/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.merged.cache"
text_embed_pt = "/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path, data_style="grounding", text_embed_pt=text_embed_pt)
de.load_yoloe()
print("length of labels:", len(de.labels))
print(de.labels[im_index]["im_file"])

# de.visual_and_save2(im_index, save_path="./visualized_grounding_example.jpg")


de=DataEngine()
cache_path="/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.merged.updated.cache"
text_embed_pt="/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path,
data_style="grounding",
text_embed_pt=text_embed_pt)
de = DataEngine()
cache_path = (
"/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.merged.updated.cache"
)
text_embed_pt = "/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path, data_style="grounding", text_embed_pt=text_embed_pt)
print("length of labels:", len(de.labels))
# de.load_yoloe()
print(de.labels[im_index]["im_file"])

# de.visual_and_save2(im_index, save_path="./visualized_grounding_example1.jpg")

39 changes: 15 additions & 24 deletions data_visual_flickr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,46 +1,37 @@
import ultralytics,os
import os

import ultralytics

workspace = os.path.dirname(os.path.dirname(os.path.abspath(ultralytics.__file__)))
os.chdir(workspace)
print("set workspace:", workspace)


from data_engine import DataEngine





if __name__ == "__main__":


# device="cuda:1"
# de=DataEngine(device=device)
# cache_path="/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.merged.cache"
# text_embed_pt="/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# text_embed_pt=text_embed_pt)

im_index=0
im_index = 0


de=DataEngine()
cache_path="/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.merged.cache"
text_embed_pt="/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path,
data_style="grounding",
text_embed_pt=text_embed_pt)
de = DataEngine()
cache_path = "/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.merged.cache"
text_embed_pt = "/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path, data_style="grounding", text_embed_pt=text_embed_pt)
de.print_data_info()

# de.visual_and_save2(im_index, save_path="./visualized_grounding_example.jpg")


de=DataEngine()
cache_path="/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.updated.cache"
text_embed_pt="/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path,
data_style="grounding",
text_embed_pt=text_embed_pt)
de = DataEngine()
cache_path = "/root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.updated.cache"
text_embed_pt = "/root/ultra_louis_work/datasets/flickr/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path, data_style="grounding", text_embed_pt=text_embed_pt)
de.print_data_info()

# de.visual_and_save2(im_index, save_path="./visualized_grounding_example1.jpg")
Expand Down
37 changes: 15 additions & 22 deletions data_visual_mixed.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,48 +1,41 @@
import ultralytics,os
import os

import ultralytics

workspace = os.path.dirname(os.path.dirname(os.path.abspath(ultralytics.__file__)))
os.chdir(workspace)
print("set workspace:", workspace)


from data_engine import DataEngine
import numpy as np




if __name__ == "__main__":


# device="cuda:1"
# de=DataEngine(device=device)
# cache_path="/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.merged.cache"
# text_embed_pt="/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# text_embed_pt=text_embed_pt)

im_index=0

im_index = 0

# de=DataEngine()
# cache_path="/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.merged.cache"
# text_embed_pt="/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# text_embed_pt=text_embed_pt)
# de.print_data_info()

# de.visual_and_save2(im_index, save_path="./visualized_grounding_example.jpg")


de=DataEngine()
cache_path="/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.updated.cache"
text_embed_pt="/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path,
data_style="grounding",
text_embed_pt=text_embed_pt)
de = DataEngine()
cache_path = (
"/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.updated.cache"
)
text_embed_pt = "/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
de.load_cached_label(cache_path=cache_path, data_style="grounding", text_embed_pt=text_embed_pt)
de.print_data_info()

de.visual_and_save2(filename="353913.jpg", save_path="./visualized_grounding_example_v2.jpg")


34 changes: 15 additions & 19 deletions data_visual_object365.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,40 +1,36 @@
import ultralytics,os
import os

import ultralytics

workspace = os.path.dirname(os.path.dirname(os.path.abspath(ultralytics.__file__)))
os.chdir(workspace)
print("set workspace:", workspace)


from data_engine import DataEngine





if __name__ == "__main__":


# device="cuda:1"
# de=DataEngine(device=device)
# cache_path="/root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.merged.cache"
# text_embed_pt="/root/ultra_louis_work/datasets/mixed_grounding/gqa/text_embeddings_mobileclip_blt.pt"
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# de.load_cached_label(cache_path=cache_path,
# data_style="grounding",
# text_embed_pt=text_embed_pt)

im_index=0
im_index = 0


de=DataEngine(device="cuda")
de=DataEngine(device="cuda")
yaml_config="/root/ultra_louis_work/datasets/Objects365v1.yaml"
cache_path="/root/ultra_louis_work/datasets/Objects365v1/labels/train.cache"
de = DataEngine(device="cuda")
de = DataEngine(device="cuda")
yaml_config = "/root/ultra_louis_work/datasets/Objects365v1.yaml"
cache_path = "/root/ultra_louis_work/datasets/Objects365v1/labels/train.cache"
de.load_cached_label(cache_path=cache_path, data_style="detection", yaml_config=yaml_config)
de.print_data_info()

# de.visual_and_save2(im_index, save_path="./visualized_grounding_example.jpg")
de=DataEngine(device="cuda")
yaml_config="/root/ultra_louis_work/datasets/Objects365v1.yaml"
cache_path="/root/ultra_louis_work/datasets/Objects365v1/labels/train.updated.cache"
de = DataEngine(device="cuda")
yaml_config = "/root/ultra_louis_work/datasets/Objects365v1.yaml"
cache_path = "/root/ultra_louis_work/datasets/Objects365v1/labels/train.updated.cache"
de.load_cached_label(cache_path=cache_path, data_style="detection", yaml_config=yaml_config)

de.print_data_info()
de.print_data_info()
11 changes: 4 additions & 7 deletions do_flickr.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,17 +1,14 @@
# activate clipenv conda env
# activate clipenv conda env
source ~/miniconda3/etc/profile.d/conda.sh
conda activate clipenv


# remove /root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.cache.A if it exists
if [ -f /root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.cache.A ]; then
rm /root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.cache.A
rm /root/ultra_louis_work/datasets/flickr/annotations/final_flickr_separateGT_train_segm.cache.A
fi


# run the refine_text.py script to generate refined labels and cache for Flickr dataset
python3 yoloe_data_engine/refine_text.py
python3 yoloe_data_engine/refine_text.py

# run data visualization script
python3 yoloe_data_engine/data_visual_flickr.py

python3 yoloe_data_engine/data_visual_flickr.py
14 changes: 4 additions & 10 deletions do_mixed.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,16 +1,10 @@
# activate clipenv conda env
# activate clipenv conda env
source ~/miniconda3/etc/profile.d/conda.sh
conda activate clipenv


# set gpu id to 2,3
# set gpu id to 2,3
export CUDA_VISIBLE_DEVICES=2,3




# run the refine_text.py script to generate refined labels and cache for Flickr dataset
python3 yoloe_data_engine/refine_text.py --img_path /root/ultra_louis_work/datasets/mixed_grounding/gqa/images \
--json_file /root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.json


python3 yoloe_data_engine/refine_text.py --img_path /root/ultra_louis_work/datasets/mixed_grounding/gqa/images \
--json_file /root/ultra_louis_work/datasets/mixed_grounding/annotations/final_mixed_train_no_coco_segm.json
Loading