search

dougsm / ggcnn

Generative Grasping CNN from "Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach" (RSS 2018)
BSD 3-Clause "New" or "Revised" License
499 stars 140 forks source link

deal with the data #12

Closed wjydou closed 5 years ago

wjydou commented 5 years ago

when I download the and extract Cornell Grasping Dataset into a single directorya,and I convert the PCD files to depth images by running python -m utils.dataset_processing.generate_cornell_depth data/cornell Nect,by running train_ggcnn.py ,and have the error INFO:root:Loading Cornell Dataset... Traceback (most recent call last): File /ggcnn/train_ggcnn.py", line 269, in run() File "/ggcnn/train_ggcnn.py", line 202, in run include_depth=args.use_depth, include_rgb=args.use_rgb) File "/ggcnn/utils/data/cornell_data.py", line 26, in init raise FileNotFoundError('No dataset files found. Check path: {}'.format(file_path)) FileNotFoundError: No dataset files found. Check path: /data/cornell

wjydou commented 5 years ago

And run the train_ggcnn.py of dataset jacquard.Then have the error Traceback (most recent call last): File "/ggcnn/train_ggcnn.py", line 269, in run() File "/ggcnn/train_ggcnn.py", line 202, in run include_depth=args.use_depth, include_rgb=args.use_rgb) File "/ggcnn/utils/data/jacquard_data.py", line 27, in init raise FileNotFoundError('No dataset files found. Check path: {}'.format(file_path)) FileNotFoundError: No dataset files found. Check path: /data/jacquard

dougsm commented 5 years ago

Hi, Can you post the whole command that you are running? It looks like you are not passing in or passing in incorrectly the dataset type and path arguments.

wjydou commented 5 years ago

@dougsm l run in the pycharm,and gave the path of dataset .Then,the result is as shown above

dougsm commented 5 years ago

Is the path definitely correct? Is it definitely /data/jacquard and not data/jacquard?

WuXingrui commented 5 years ago

@wjydou You can try to delete '*' in the 22nd line of jacquard_data.py which is in utils/data/

wjydou commented 5 years ago

@WuXingrui Thanks for you,I had delete '*' in the 22nd line of jacquard_data.py which is in utils/data/,but it is also useless.

dougsm commented 5 years ago

Hello, did you resolve this issue? I did notice that the instructions may have been ambiguous. The expected path format for the cornell dataset is that the 01, 02, 03, etc folders are still in the structure. I have updated the readme.

wjydou commented 5 years ago

@dougsm Thanks!

seivazi commented 4 years ago

Hello, did you resolve this issue? I did notice that the instructions may have been ambiguous. The expected path format for the cornell dataset is that the 01, 02, 03, etc folders are still in the structure. I have updated the readme.

@dougsm I don't understand above very well. Could you explain it one more time? Let's say I have cornell_dataset folder then as sub-folders I should have 01,02,03,... am I right? For train then I should provide the path to ./cornell_dataset/ Do you load each sub-folder during train? or should I have only one folder e.g.01 with all images inside

dougsm commented 4 years ago

Yes, that's right, the directory should contain the 01, 02, etc subdirectories. Judging by the line that loads the files, every subfolder is loaded: https://github.com/dougsm/ggcnn/blob/ad48bc5f768fe0a9ba9fd47729638e0aed46e47b/utils/data/cornell_data.py#L22