HandGestureClassify icon indicating copy to clipboard operation
HandGestureClassify copied to clipboard

Published 20 hours ago • verified publisher icon yyyerica

run_demo_hand_with_tracker.py Errors

Open LeonardChin2017 opened this issue 6 years ago • 15 comments

Hi, anyone know how to fix this error:

tensorflow.python.framework.errors_impl.NotFoundError: FindFirstFile failed for: /models/weights : The system cannot find the path specified. ; No such process

Thanks

LeonardChin2017 avatar Nov 04 '18 16:11 LeonardChin2017

  1. goto: https://drive.google.com/open?id=1gOwBY5puCusYPCQaPcEUMmQtPnGHCPyl

  2. download the rar file: cpm_hand_tf.rar

  3. mkdir weights in folder models

  4. mv rar file to folder weights

  5. unrar e cpm_hand_tf.rar

MyJoiT avatar Nov 07 '18 17:11 MyJoiT

Error fixed few days ago, sorry for late update and thanks @MyJoiT

LeonardChin2017 avatar Nov 08 '18 06:11 LeonardChin2017

You're welcome :) @LeonardChin2017

MyJoiT avatar Nov 08 '18 06:11 MyJoiT

@liang0724s

Well, I received 4 emails from this issue notification.

My environment: archlinux pyenv + virtualenv python 3.6.4 pip 18.1 tensorflow 1.12.0

I advise you to change you OS. Use Unix / Linux, not windows. Then that will be ok.

MyJoiT avatar Jan 09 '19 14:01 MyJoiT

@liang0724s

Well, I received 4 emails from this issue notification.

My environment: archlinux pyenv + virtualenv python 3.6.4 pip 18.1 tensorflow 1.12.0

I advise you to change you OS. Use Unix / Linux, not windows. Then that will be ok.

@MyJoiT Thank you. I will try to compile it on Ubuntu 14.4 in VMware environment.

liang0724s avatar Jan 10 '19 07:01 liang0724s

@MyJoiT， I'm sorry to trouble you. I forget to ask you which version I should install for tensorflow: CPU or GPU? From the compile error info I done before，it reminds me have no GPU to use. Because I compile it on CPU version of tensorflow. Did you try to compile it on CPU version of tensorflow? I want to make it possible, although my computer has NVIDIA GPU. But I don't know whether it is possible and where I should modify the code. Do you think it is a good way to set up the compile environment using Anaconda? After all, Anaconda is very convenience.

liang0724s avatar Jan 11 '19 03:01 liang0724s

@liang0724s Do not use GPU for your first try. Tensorflow-GPU require many libs. It will be more complex if you use Virtual Machine.

MyJoiT avatar Jan 11 '19 03:01 MyJoiT

Hi @liang0724s, I am using Tensorflow - GPU for the first try.

Here are my gears:

  1. Window 10
  2. Cuda 9.0 and Cudnn 7
  3. Anaconda 5.2.0
  4. Pycharm (use any IDE you like)

If you are using Anaconda, you can setup the environment with the command "conda install -c aaronzs tensorflow-gpu" after you done creating and activate your virtual environment in Anaconda prompt.

LeonardChin2017 avatar Jan 15 '19 09:01 LeonardChin2017

Hi @liang0724s, I am using Tensorflow - GPU for the first try.

Here are my gears:

  1. Window 10
  2. Cuda 9.0 and Cudnn 7
  3. Anaconda 5.2.0
  4. Pycharm (use any IDE you like)

If you are using Anaconda, you can setup the environment with the command "conda install -c aaronzs tensorflow-gpu" after you done creating and activate your virtual environment in Anaconda prompt.

@LeonardChin2017 , Thanks.My computer GPU devices is GeForce GT 720M, although this version support CUDA, but may the hashrate is lower than the mainstream. So I can only choose to try Tensorflow - CPU first.

liang0724s avatar Jan 15 '19 12:01 liang0724s

you are welcome :) @liang0724s

LeonardChin2017 avatar Jan 17 '19 02:01 LeonardChin2017

@liang0724s Do not use GPU for your first try. Tensorflow-GPU require many libs. It will be more complex if you use Virtual Machine.

Hi @MyJoiT , How to modify the code if I use CPU first?

liang0724s avatar Jan 23 '19 12:01 liang0724s

you are welcome :) @liang0724s Hi @LeonardChin2017 , sorry,I need your help if you can. the following is the compiling info, I don‘t know where is wrong? I think it has successfully loaded the model, because the compiling info has printed this message: "Load model from [models\weights\cpm_hand\input_256_output_32\joints_21\stages_3
init_0.001_rate_0.5_step_10000\cpm_hand]"

`(D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv) C:\Users\asus>python D:\Progra mData\HandGestureClassify-master\run_demo_hand_with_tracker.py 2019-01-23 21:49:52.215849: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX 2019-01-23 21:49:53.282853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1 405] Found device 0 with properties: name: GeForce GT 720M major: 2 minor: 1 memoryClockRate(GHz): 1.25 pciBusID: 0000:01:00.0 totalMemory: 2.00GiB freeMemory: 1.94GiB 2019-01-23 21:49:53.292853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1 455] Ignoring visible gpu device (device: 0, name: GeForce GT 720M, pci bus id: 0000:01:00.0, compute capability: 2.1) with Cuda compute capability 2.1. The min imum required Cuda capability is 3.0. 2019-01-23 21:49:53.292853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:9 65] Device interconnect StreamExecutor with strength 1 edge matrix: 2019-01-23 21:49:53.302853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:9 71] 0 2019-01-23 21:49:53.302853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:9 84] 0: N Load model from [models\weights\cpm_hand\input_256_output_32\joints_21\stages_3
init_0.001_rate_0.5_step_10000\cpm_hand] Traceback (most recent call last): File "D:\ProgramData\HandGestureClassify-master\run_demo_hand_with_tracker.py" , line 391, in tf.app.run() File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\platform\app.py", line 125, in run _sys.exit(main(argv)) File "D:\ProgramData\HandGestureClassify-master\run_demo_hand_with_tracker.py" , line 60, in main saver.restore(sess, 'models/weights/cpm_hand') File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\training\saver.py", line 1715, in restore if not checkpoint_exists(compat.as_text(save_path)): File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\training\saver.py", line 2056, in checkpoint_exists if file_io.get_matching_files(pathname): File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\lib\io\file_io.py", line 342, in get_matching_files for single_filename in filename File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\framework\errors_impl.py", line 519, in exit c_api.TF_GetCode(self.status.status)) tensorflow.python.framework.errors_impl.NotFoundError: FindFirstFile failed for: models/weights : ϵͳ\udcd5Ҳ\udcbb\udcb5\udcbdָ\udcb6\udca8\udcb5\udcc4·\udcbe\u dcb6\udca1\udca3 ; No such process

(D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv) C:\Users\asus> `

liang0724s avatar Jan 23 '19 14:01 liang0724s

you are welcome :) @liang0724s Hi @LeonardChin2017 , sorry,I need your help if you can. the following is the compiling info, I don‘t know where is wrong? I think it has successfully loaded the model, because the compiling info has printed this message: "Load model from [models\weights\cpm_hand\input_256_output_32\joints_21\stages_3 init_0.001_rate_0.5_step_10000\cpm_hand]"

`(D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv) C:\Users\asus>python D:\Progra mData\HandGestureClassify-master\run_demo_hand_with_tracker.py 2019-01-23 21:49:52.215849: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX 2019-01-23 21:49:53.282853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1 405] Found device 0 with properties: name: GeForce GT 720M major: 2 minor: 1 memoryClockRate(GHz): 1.25 pciBusID: 0000:01:00.0 totalMemory: 2.00GiB freeMemory: 1.94GiB 2019-01-23 21:49:53.292853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1 455] Ignoring visible gpu device (device: 0, name: GeForce GT 720M, pci bus id: 0000:01:00.0, compute capability: 2.1) with Cuda compute capability 2.1. The min imum required Cuda capability is 3.0. 2019-01-23 21:49:53.292853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:9 65] Device interconnect StreamExecutor with strength 1 edge matrix: 2019-01-23 21:49:53.302853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:9 71] 0 2019-01-23 21:49:53.302853: I tensorflow/core/common_runtime/gpu/gpu_device.cc:9 84] 0: N Load model from [models\weights\cpm_hand\input_256_output_32\joints_21\stages_3 init_0.001_rate_0.5_step_10000\cpm_hand] Traceback (most recent call last): File "D:\ProgramData\HandGestureClassify-master\run_demo_hand_with_tracker.py" , line 391, in tf.app.run() File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\platform\app.py", line 125, in run _sys.exit(main(argv)) File "D:\ProgramData\HandGestureClassify-master\run_demo_hand_with_tracker.py" , line 60, in main saver.restore(sess, 'models/weights/cpm_hand') File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\training\saver.py", line 1715, in restore if not checkpoint_exists(compat.as_text(save_path)): File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\training\saver.py", line 2056, in checkpoint_exists if file_io.get_matching_files(pathname): File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\lib\io\file_io.py", line 342, in get_matching_files for single_filename in filename File "D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv\lib\site-packages\tensor flow\python\framework\errors_impl.py", line 519, in exit c_api.TF_GetCode(self.status.status)) tensorflow.python.framework.errors_impl.NotFoundError: FindFirstFile failed for: models/weights : ϵͳ\udcd5Ҳ\udcbb\udcb5\udcbdָ\udcb6\udca8\udcb5\udcc4·\udcbe\u dcb6\udca1\udca3 ; No such process

(D:\ProgramData\Anaconda3\envs\tensorflow_gpuenv) C:\Users\asus> `

@LeonardChin2017 Thanks,Leonard. I have compiled it successfully. But the average FPS=0.43. Also, the GPU compute capability is 2.1. It's too low. I found an excuse to change a new notebook.

liang0724s avatar Jan 26 '19 09:01 liang0724s

Does anyone has the error below.I have already unrar , and mkdir weights ValueError: The passed save_path is not a valid checkpoint: models/weights/cpm_hand

cwsdrogo avatar Sep 07 '19 07:09 cwsdrogo

Does anyone has the error below.I have already unrar , and mkdir weights ValueError: The passed save_path is not a valid checkpoint: models/weights/cpm_hand ~%EART (_$13$RU~P5K H5D the right path showld be this maybe,i run successfully,and fps is 10

BothSavage avatar Mar 22 '20 10:03 BothSavage