Comments (5)
I noticed that when generating the dataset, the alt image was lost in accuracy because it was saved as png, so I changed the relevant code
from sensaturban-bev-seg3d.
I noticed that when generating the dataset, the alt image was lost in accuracy because it was saved as png, so I changed the relevant code
Hello, I meet the same results, too. Have you reproduced the results? What's more, I have not found OCRNet-HRNet in code
from sensaturban-bev-seg3d.
I noticed that when generating the dataset, the alt image was lost in accuracy because it was saved as png, so I changed the relevant code
Excuse me, I met this situation what should I do?python train.py --use-balanced-weights --batch-size 8 --base-size 500 --crop-size 500 --loss-type focal --epochs 200 --eval-interval 1
Namespace(base_size=500, batch_size=8, checkname='Res-Unet', crop_size=500, cuda=True, dataset='cityscapes', epochs=200, eval_interval=1, ft=False, gpu_ids=[0, 1], loss_type='focal', lr=0.02, lr_scheduler='poly', momentum=0.9, nesterov=False, no_cuda=False, no_val=False, resume=None, seed=1, start_epoch=0, test_batch_size=8, use_balanced_weights=True, weight_decay=0.0001, workers=4)
0%| | 0/13 [00:00<?, ?it/s]Calculating classes weights
0%| | 0/13 [00:00<?, ?it/s]
Traceback (most recent call last):
File "train.py", line 292, in
main()
File "train.py", line 280, in main
trainer = Trainer(args)
File "train.py", line 49, in init
weight = calculate_weigths_labels(args.dataset, self.train_loader, self.nclass)
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/utils/calculate_weights.py", line 38, in calculate_weigths_labels
y = sample['label']
KeyError: 'label'
from sensaturban-bev-seg3d.
I noticed that when generating the dataset, the alt image was lost in accuracy because it was saved as png, so I changed the relevant code
Hello, I meet the same results, too. Have you reproduced the results? What's more, I have not found OCRNet-HRNet in code
Excuse me, I met this situation what should I do? python /media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py
0it [00:00, ?it/s]loading 6 files
loading file /media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/dataloaders/datasets/data_release/test/birmingham_block_2.ply[00:00<?, ?it/s]
0%| | 0/6 [00:00<?, ?it/s]
0it [00:00, ?it/s]
Traceback (most recent call last):
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 641, in
Sensat.evaluate_batch(Sensat.evaluate_batch_nn(Sensat.eval_offline_img2pts))
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 624, in evaluate_batch
miou, iou_list, gt_list, oacc, macc = batch_mIoU(batch_iterator)
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/helper_metrics.py", line 83, in batch_mIoU
for preds, gt in tqdm(iter_data):
File "/home/hubu/anaconda3/envs/Segment/lib/python3.7/site-packages/tqdm/std.py", line 1182, in iter
for obj in iterable:
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 606, in evaluate_batch_nn
ply_data = self.load_points(ply_path, reformat=True)
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 155, in load_points
c = _ply_data["class"]
ValueError: no field of name class
from sensaturban-bev-seg3d.
Thank you so much for your great work When I train according to the readme process, the model always fails to achieve the same curve as img/single_scale_training.png My experimental configuration is: 2 x 2080ti and batchsize=4 miou can only be up to 41% in the validation set
In addition, can you also provide pre-trained weights? thank you very much
Excuse me, I met this situation what should I do? python /media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py
0it [00:00, ?it/s]loading 6 files
loading file /media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/dataloaders/datasets/data_release/test/birmingham_block_2.ply[00:00<?, ?it/s]
0%| | 0/6 [00:00<?, ?it/s]
0it [00:00, ?it/s]
Traceback (most recent call last):
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 641, in
Sensat.evaluate_batch(Sensat.evaluate_batch_nn(Sensat.eval_offline_img2pts))
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 624, in evaluate_batch
miou, iou_list, gt_list, oacc, macc = batch_mIoU(batch_iterator)
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/helper_metrics.py", line 83, in batch_mIoU
for preds, gt in tqdm(iter_data):
File "/home/hubu/anaconda3/envs/Segment/lib/python3.7/site-packages/tqdm/std.py", line 1182, in iter
for obj in iterable:
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 606, in evaluate_batch_nn
ply_data = self.load_points(ply_path, reformat=True)
File "/media/hubu/Data1/202131116023006_bky/SensatUrban-BEV-Seg3D-main/preprocess/point_EDA_31.py", line 155, in load_points
c = _ply_data["class"]
ValueError: no field of name class
from sensaturban-bev-seg3d.
Related Issues (6)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sensaturban-bev-seg3d.