search

andabi / deep-voice-conversion

Deep neural networks for voice conversion (voice style transfer) in Tensorflow
MIT License
3.92k stars 843 forks source link

Stuck when running train1.py #16

Open toannhu opened 6 years ago

toannhu commented 6 years ago

First of all i'm really appreciate the great repo you made here. I'm trying to run train1.py but it's stuck at 0% for hours and I can't figure out what happening with me.

image

Here is the TIMID Dataset folder structure that i put into /datasets.

image

I also changed .wav to .WAV in hparams.py

image

Did I miss something? Plz help me! Thank you!

begeekmyfriend commented 6 years ago

I met the same issue that I still have no idea about. One of the approach is to invoke

train(logdir=logdir, queue=False)

in main function in train1.py.

toannhu commented 6 years ago

@begeekmyfriend Thanks. The problem is solved. I think that the model waits to queue all of TIMID Dataset before running? Maybe something wrong with the queue implementation here!? Hope someone can explain about this problem.

201528014227051 commented 6 years ago

Thanks for this problem, My situation is similar to yours, but i stuck before start running, it just stop there. By debug by pycharm, it stops in line: with tf.Session(config=session_conf) as sess: in train1.py, The timit dataset i used is not complete, is this the reason? Did I miss something? Plz help me! Thank you!

VictoriaBentell commented 6 years ago

I believe I had the same issue, but I solved it here. Hopefully something there helps.

lightwithshadow commented 5 years ago

2018-10-11 15:05:54.311985: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA 2018-10-11 15:05:54.349099: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:892] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2018-10-11 15:05:54.349548: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1030] Found device 0 with properties: name: GeForce GT 755M major: 3 minor: 0 memoryClockRate(GHz): 1.0195 pciBusID: 0000:01:00.0 totalMemory: 1.96GiB freeMemory: 1.74GiB 2018-10-11 15:05:54.349581: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GT 755M, pci bus id: 0000:01:00.0, compute capability: 3.0) [1011 15:05:55 @base.py:211] Initializing the session ... [1011 15:05:55 @base.py:218] Graph Finalized. [1011 15:05:55 @concurrency.py:37] Starting EnqueueThread QueueInput/input_queue ... [1011 15:05:55 @graph.py:73] Running Op sync_variables/sync_variables_from_main_tower ... [1011 15:05:56 @base.py:250] Start Epoch 1 ... 0%| |0/100[00:00<?,?it/s][1011 15:05:56 @input_source.py:550] Pre-filling StagingArea ...

how to solve??? thank u !

iamxiaoyubei commented 5 years ago

@lightwithshadow Have you solved your problem? I met the same problem with you. If you solved, plz help me! Thank you!

pravn commented 5 years ago

@VictoriaBentell's solution works nicely (thank you). Can't say what your problem is, but in my case, renaming the files from .WAV to .wav works. Rename with this tip.

https://stackoverflow.com/questions/21985492/recursively-change-file-extensions-in-bash using the 'find' command. Praveen

pravn.wordpress.com

On Thu, Feb 14, 2019 at 7:58 AM Yubei notifications@github.com wrote:

@lightwithshadow https://github.com/lightwithshadow Have you solved your problem? I met the same problem with you. If you solved, plz help me! Thank you!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/andabi/deep-voice-conversion/issues/16#issuecomment-463681881, or mute the thread https://github.com/notifications/unsubscribe-auth/AAhnVVs5DQYr-lf09pAAlFqSknuRN8HDks5vNYe8gaJpZM4Rd7aD .

iamxiaoyubei commented 5 years ago

@pravn I had solved it. Thank you for your reply!

xminte commented 3 years ago

@VictoriaBentell's solution works nicely (thank you). Can't say what your problem is, but in my case, renaming the files from .WAV to .wav works. Rename with this tip. https://stackoverflow.com/questions/21985492/recursively-change-file-extensions-in-bash using the 'find' command. Praveen pravn.wordpress.com … On Thu, Feb 14, 2019 at 7:58 AM Yubei @.***> wrote: @lightwithshadow https://github.com/lightwithshadow Have you solved your problem? I met the same problem with you. If you solved, plz help me! Thank you! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#16 (comment)>, or mute the thread https://github.com/notifications/unsubscribe-auth/AAhnVVs5DQYr-lf09pAAlFqSknuRN8HDks5vNYe8gaJpZM4Rd7aD .

It works,thx