Readme new known issue

Author: alexfrostdesu

Readme.md

@@ -112,6 +112,18 @@ BrokenPipeError: [Errno 32] Broken pipe
 This is an `multiprocessing.Queue` closing error, when some items are still in the Queue.
 This should not affect the app in any meaningful way.
 
+#### Tensorflow not releasing GPU memory
+Sometimes Tensorflow session will not release all allocated GPU memory when you stop the analysis. This will result in OOM (out of memory) error.
+See more on that issue on [tensorflow github](https://github.com/tensorflow/tensorflow/issues/19731)
 
+The error itself looks like this:
+```
+2020-02-10 13:44:49.972600: E tensorflow/stream_executor/cuda/cuda_driver.cc:806] failed to allocate 635,29M (666146048 bytes) from device: CUDA_ERROR_OUT_OF_MEMORY: out of memory
+```
+
+The issue can be resolved by closing and opening the app. If not, manually kill all python processes
+```
+killall -9 python
+```
 
 ## License

settings.ini

@@ -2,18 +2,17 @@
 RESOLUTION = 848, 480
 FRAMERATE = 30
 STREAMS = color, depth, infrared
-OUTPUT_DIRECTORY = E:\Output
-MULTIPLE_DEVICES = False
-VIDEO_SOURCE = 2
-ANIMALS_NUMBER = 1
+OUTPUT_DIRECTORY = /home/agschwarz/Output
+MULTIPLE_DEVICES = True
+CAMERA_SOURCE = 2
 
 [DeepLabCut]
-DLC_PATH = D:\DeepLabCut\DeepLabCut-master\deeplabcut
-MODEL = 13PointsFeb3-trainset95shuffle1
+DLC_PATH = /home/agschwarz/DeepLabCut/deeplabcut
+MODEL = openfield_testset2Nov07-trainset95shuffle1
 
 [Experiment]
 EXP_NUMBER = 1
 RECORD_EXP = True
 
 [Video]
-VIDEO_SOURCE = E:\Output\DeepLabCutVideos\ExperimentD1_G2_37-16092019-141541.avi
+VIDEO_SOURCE = /home/agschwarz/Output/ExperimentD1_G2_37-16092019-141541.avi