prgumd / evdodgenet Goto Github PK

View Code? Open in Web Editor NEW

101.0 12.0 29.0 7.72 MB

EVDodgeNet: Deep Dynamic Obstacle Dodging with event cameras

Home Page: http://prg.cs.umd.edu/EVDodgeNet

License: BSD 3-Clause "New" or "Revised" License

CMake 1.05% Python 70.46% Shell 0.12% C++ 28.37%

quadrotor autonomous unmanned-aerial-vehicle robotics robot drone evdodge gapflyt detection tracking

evdodgenet's People

Contributors

Stargazers

Watchers

evdodgenet's Issues

Dataset structure for EVSegNet

Hello. While trying to understand the EVSegNet code you provided, I have seen that the RunEVSegNet.py and TrainEVSegNet.py scripts have a function named "SetupAll" where certain files named "DirNames.txt" and "Train.txt" are trying to be read, like the following:

EVDodgeNet/code/EVSegNet/RunEVSegNet.py

Lines 60 to 62 in 0b6f6a0

    
           DirNamesPath = ReadPath + os.sep + 'DirNames.txt' 
        
           TrainPath = ReadPath + os.sep + 'Train.txt' 
        
           DirNames, TrainNames = ReadDirNames(DirNamesPath, TrainPath)

Checking the dataset from https://drive.google.com/open?id=1pJSVXtRlTU4_b-I8oakoZtl1ChcG0OiQ and also checking the dataset structure desciption in https://github.com/prgumd/EVDodgeNet/wiki/Access-Data-Set-Using-ROS-Python, I could not find any file named "DirNames.txt" or "Train.txt". What am I missing here, can you guide me on this?

Segmentation network name error

Hello. The following code line:

EVDodgeNet/code/EVSegNet/RunEVSegNet.py

Line 291 in 0b6f6a0

tu.SetGPU(GPUDevice)

produces a "NameError" since "global name 'tu' is not defined".

When will you upload the code and instruction guidelines?

question about dataset

I have download your dataset on https://github.com/prgumd/EVDodgeNet/wiki/Access-Data-Set-Using-ROS-Python.
But when I wanted to run your code, I encountered the following problems.
Follow the guide, I run the code about EVDeblurNet, but IOError(can't find DirNames.txt)
Then I found that this file was generate in DataParseUnsrack.py, but print(CurrReadPath + 'events' + os.sep + '*.png') is not in your provided dataset.
I'm not sure whether my idea is correct. I would appreciate your reply.

WIKI

There is something wrong in the dataset you supplied , because there is no downfacing rosbag .what's more, could you tell me the input and output of deblurnet and homography net.I can not understand the wiki instruction.

Cannot read model files

I am trying to read pretrained model files but I got error and could not load any of them.

This is just an example of one of pretrained models. I tried every one of them.

raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.DataLossError: Unable to open table file EVDodgeNet/code/HomographyNetUnsup/Checkpoints/HomographySUSL1LR1e-4/199model.ckpt.data-00000-of-00001: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator?
         [[node save/RestoreV2 (defined at /drone/EVDodgeNet/code/HomographyNetUnsup/RunEVHomographyNet.py:277) ]]

Original stack trace for u'save/RestoreV2':
  File "/drone/EVDodgeNet/code/HomographyNetUnsup/RunEVHomographyNet.py", line 378, in <module>
    main()
  File "/drone/EVDodgeNet/code/HomographyNetUnsup/RunEVHomographyNet.py", line 375, in main
    TestOperation(PatchPH, I1PH, I2PH, PatchSize, ModelPath, ReadPath, WritePath, TrainNames, NumTrainSamples, CropType)
  File "/drone/EVDodgeNet/code/HomographyNetUnsup/RunEVHomographyNet.py", line 277, in TestOperation
    Saver = tf.train.Saver()
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 825, in __init__
    self.build()
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 837, in build
    self._build(self._filename, build_save=True, build_restore=True)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 875, in _build
    build_restore=build_restore)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 508, in _build_internal
    restore_sequentially, reshape)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 328, in _AddRestoreOps
    restore_sequentially)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 575, in bulk_restore
    return io_ops.restore_v2(filename_tensor, names, slices, dtypes)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/ops/gen_io_ops.py", line 1696, in restore_v2
    name=name)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
    op_def=op_def)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
    return func(*args, **kwargs)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 3616, in create_op
    op_def=op_def)
  File "/miniconda3/envs/tf14/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 2005, in __init__
    self._traceback = tf_stack.extract_stack()

EVDodgeNet/code/EVSegNet/RunEVSegNet.py

Lines 280 to 283 in 0b6f6a0

    
           Parser.add_argument('--WritePath', dest='WritePath', default='/media/nitin/Research/EVDodge/DatasetChethanEvents/DeblurredHomographyDodgeNet',\ 
        
           Parser.add_argument('--GPUDevice', type=int, default=0, help='What GPU do you want to use? -1 for CPU, Default:0') 
        
                                                                                    help='Path to write images to, Default:WritePath')

Specifically, lines 281 and 282 should be switched.

Missing dependency

Hello. I found out that your code also depends on scikit-image package, which is missing in dependencies list at https://github.com/prgumd/EVDodgeNet/wiki/RunningCode#dependencies , so you may consider adding it there.

Thanks again for your work.

prgumd / evdodgenet Goto Github PK

evdodgenet's People

Contributors

Stargazers

Watchers

Forkers

evdodgenet's Issues

Dataset structure for EVSegNet

Segmentation network name error

When will you upload the code and instruction guidelines?

question about dataset

WIKI

Cannot read model files

Require your dataset and code

Dataset and Models are in incorrect structure

Segmentation network syntax error

Missing dependency

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	DirNamesPath = ReadPath + os.sep + 'DirNames.txt'
	TrainPath = ReadPath + os.sep + 'Train.txt'
	DirNames, TrainNames = ReadDirNames(DirNamesPath, TrainPath)

	Parser.add_argument('--WritePath', dest='WritePath', default='/media/nitin/Research/EVDodge/DatasetChethanEvents/DeblurredHomographyDodgeNet',\
	Parser.add_argument('--GPUDevice', type=int, default=0, help='What GPU do you want to use? -1 for CPU, Default:0')
	help='Path to write images to, Default:WritePath')