From awertz at cmu.edu Fri Jul 1 11:38:48 2016 From: awertz at cmu.edu (Anthony Wertz) Date: Fri, 1 Jul 2016 11:38:48 -0400 Subject: Tensorflow on GPU In-Reply-To: References: <927D29B3-F1C3-4513-A140-28C044902F16@andrew.cmu.edu>

Message-ID: Hanqi, I'm not sure if you solved this yet. The problem I see is the version of the CUDA runtime your package is looking for: 7.0 is not installed on either GPU1 or GPU2, it's 7.5. This means you either need to use a newer version of the tensorflow package or (my recommendation) build your own package in a python virtualenv and use that. Predrag It might be a good idea for us to build the tensorflow package globally on the GPU machines. 2016-06-30 17:17 GMT-04:00 Hanqi Sun : > I just set the environment path on gpu2 and the problem persists. > > Thanks, > Hanqi > > > > On Thu, Jun 30, 2016 at 5:10 PM, Hanqi Sun wrote: > >> Hi all, >> >> Thanks for all the replies! Actually, the error message was generated on >> the server gpu2.int.autonlab.org. And when I tried >> >> printenv LD_LIBRARY_PATH >> >> >> I got >> >> /usr/local/cuda/lib64:/usr/lib64/mpich/lib >> >> >> I do not know whether I should change the environment variable on the >> server. Could someone have a look at it? >> >> Thanks, >> Hanqi >> >> On Thu, Jun 30, 2016 at 4:44 PM, Benedikt Boecking < >> boecking at andrew.cmu.edu> wrote: >> >>> Have you tried setting the cuda environment path? >>> >>> export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64 >>> >>> On Jun 30, 2016, at 4:31 PM, Hanqi Sun wrote: >>> >>> Hi all, >>> >>> I am a research assistant for Prof. Poczos and would like to use >>> Tensorflow for my project. >>> >>> When I try to execute >>> >>> import tensorflow >>> >>> >>> on the GPU node, I get the following error message >>> >>> Traceback (most recent call last): >>> >>> File "", line 1, in >>> >>> File "/usr/lib/python2.7/site-packages/tensorflow/__init__.py", line >>>> 23, in >>> >>> from tensorflow.python import * >>> >>> File "/usr/lib/python2.7/site-packages/tensorflow/python/__init__.py", >>>> line 50, in >>> >>> from tensorflow.python.framework.framework_lib import * >>> >>> File >>>> "/usr/lib/python2.7/site-packages/tensorflow/python/framework/framework_lib.py", >>>> line 62, in >>> >>> from tensorflow.python.framework.ops import Graph >>> >>> File >>>> "/usr/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line >>>> 40, in >>> >>> from tensorflow.python.framework import versions >>> >>> File >>>> "/usr/lib/python2.7/site-packages/tensorflow/python/framework/versions.py", >>>> line 24, in >>> >>> from tensorflow.python import pywrap_tensorflow >>> >>> File >>>> "/usr/lib/python2.7/site-packages/tensorflow/python/pywrap_tensorflow.py", >>>> line 28, in >>> >>> _pywrap_tensorflow = swig_import_helper() >>> >>> File >>>> "/usr/lib/python2.7/site-packages/tensorflow/python/pywrap_tensorflow.py", >>>> line 24, in swig_import_helper >>> >>> _mod = imp.load_module('_pywrap_tensorflow', fp, pathname, >>>> description) >>> >>> ImportError: libcudart.so.7.0: cannot open shared object file: No such >>>> file or directory >>> >>> >>> >>> >>> Could anyone check this out and try to fix it? Thanks! >>> >>> Also, I will greatly appreciate it if any of you could share your >>> experience with Tensorflow (especially about image processing) with me as I >>> am new to this framework. If you are interested, please send a private >>> email to me and I am happy to stop by at your office at any time. >>> >>> Best, >>> Hanqi >>> >>> >>> >> > -- *Anthony Wertz* Research Programmer and Analyst Robotics Institute - Auton Lab Carnegie Mellon University awertz at cmu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From awd at cs.cmu.edu Mon Jul 4 19:01:16 2016 From: awd at cs.cmu.edu (Artur Dubrawski) Date: Mon, 4 Jul 2016 19:01:16 -0400 Subject: Artur out of office Tue-Wed and Thu AM this week, on Fri only if needed Message-ID: <3fb7a351-9ff7-730c-0f00-52738aecd37b@cs.cmu.edu> Team, We are going to have to do a lot of communication on line this and the next week. On top of what is stated in the subject line, I will be out of town on business all of the next week, and on Monday 7/18. Please let me know if a meeting in person is needed and please coordinate with Karen Widmaier who/when this Friday and Thursday afternoon. Cheers Artur From mille856 at andrew.cmu.edu Mon Jul 11 14:41:35 2016 From: mille856 at andrew.cmu.edu (Kyle Miller) Date: Mon, 11 Jul 2016 14:41:35 -0400 Subject: Going away lunch for Mathieu: Thu. Noon Message-ID: <5783E85F.7000309@andrew.cmu.edu> All, Mathieu will be leaving us this week. We will sorely miss his presence. Therefore, we will be going to lunch on Thursday (noon) to say farewell. Please let me know if you'd like to attend. Location is still TBD. -Kyle From awd at cs.cmu.edu Mon Jul 11 14:43:22 2016 From: awd at cs.cmu.edu (Artur Dubrawski) Date: Mon, 11 Jul 2016 14:43:22 -0400 Subject: Going away lunch for Mathieu: Thu. Noon In-Reply-To: <5783E85F.7000309@andrew.cmu.edu> References: <5783E85F.7000309@andrew.cmu.edu> Message-ID: <155db455510.27e0.6e9e125e0ec5410dbacdcf726a0bfb4f@cs.cmu.edu> Team There is also a pub/tavern outing being planned for Friday evening so that Mathieu can be sent back to Europe in a properly comprehensive fashion. Everyone invited, especially those who cannot make Thursday lunch. Exact place and time to be determined. Thanks Artur Sent with AquaMail for Android http://www.aqua-mail.com On July 11, 2016 2:37:34 PM Kyle Miller wrote: > All, > Mathieu will be leaving us this week. We will sorely miss his > presence. Therefore, we will be going to lunch on Thursday (noon) to say > farewell. Please let me know if you'd like to attend. Location is still TBD. > > -Kyle From jkoushik at cs.cmu.edu Tue Jul 12 14:56:46 2016 From: jkoushik at cs.cmu.edu (Jayanth Koushik) Date: Tue, 12 Jul 2016 14:56:46 -0400 Subject: Unexplained GPU memory usage on GPU1 Message-ID: Hi Does anyone know what?s happening with the second GPU on GPU1? nvidia-smi says 10212MiB is being used but there are no processes running on it. Earlier I thought this was a fragmentation issue, but that doesn?t seem to be the case. Thanks Jayanth PS: If you have idle notebooks running, please close them - the GPU memory doesn?t get freed. From mille856 at andrew.cmu.edu Thu Jul 14 09:52:28 2016 From: mille856 at andrew.cmu.edu (Kyle Miller) Date: Thu, 14 Jul 2016 09:52:28 -0400 Subject: Going away lunch for Mathieu: Thu. Noon In-Reply-To: <1468331689848.71957@andrew.cmu.edu> References: <5783E85F.7000309@andrew.cmu.edu> <1468331689848.71957@andrew.cmu.edu> Message-ID: <5787991C.1000401@andrew.cmu.edu> All, A reminder that we will be going to lunch with Mathieu today. We will gather in the lab's hall at noon to head out. Still no decision on where, so please let me know if you have any idea. Also, we will each be responsible for our own check. I'll pick up Mathieu's bill. -Kyle > ________________________________________ > From: Autonlab-users on behalf of Kyle Miller > Sent: Monday, July 11, 2016 2:41 PM > To: users at autonlab.org > Subject: Going away lunch for Mathieu: Thu. Noon > > All, > Mathieu will be leaving us this week. We will sorely miss his > presence. Therefore, we will be going to lunch on Thursday (noon) to say > farewell. Please let me know if you'd like to attend. Location is still TBD. > > -Kyle From mille856 at andrew.cmu.edu Thu Jul 14 10:41:52 2016 From: mille856 at andrew.cmu.edu (Kyle Miller) Date: Thu, 14 Jul 2016 10:41:52 -0400 Subject: Going away lunch for Mathieu: Thu. Noon In-Reply-To: <5787991C.1000401@andrew.cmu.edu> References: <5783E85F.7000309@andrew.cmu.edu> <1468331689848.71957@andrew.cmu.edu> <5787991C.1000401@andrew.cmu.edu> Message-ID: <5787A4B0.3060901@andrew.cmu.edu> All, The plan is to head to Union Grill (plan A). If there is no room there, we will go over to the museum (plan B). Currently we have ~10 ppl planning on coming. -Kyle On 07/14/2016 09:52 AM, Kyle Miller wrote: > All, > A reminder that we will be going to lunch with Mathieu today. We > will gather in the lab's hall at noon to head out. > Still no decision on where, so please let me know if you have any idea. > Also, we will each be responsible for our own check. I'll pick up > Mathieu's bill. > > -Kyle > > >> ________________________________________ >> From: Autonlab-users on behalf >> of Kyle Miller >> Sent: Monday, July 11, 2016 2:41 PM >> To: users at autonlab.org >> Subject: Going away lunch for Mathieu: Thu. Noon >> >> All, >> Mathieu will be leaving us this week. We will sorely miss his >> presence. Therefore, we will be going to lunch on Thursday (noon) to say >> farewell. Please let me know if you'd like to attend. Location is >> still TBD. >> >> -Kyle > From awd at cs.cmu.edu Fri Jul 15 09:16:55 2016 From: awd at cs.cmu.edu (Artur Dubrawski) Date: Fri, 15 Jul 2016 09:16:55 -0400 Subject: Mathieu's farewell pub party tonight: 8:30pm at Mario's on Walnut Street (in Shadyside) Message-ID: <11b17c05-365d-71f9-903e-f75e7898524f@cs.cmu.edu> http://www.mariospgh.com/eastside/ We will try to get a table on the deck upstairs so please let me know if you are coming and provide the headcount (significant others are welcome). Thanks Artur From awd at cs.cmu.edu Fri Jul 15 11:44:18 2016 From: awd at cs.cmu.edu (Artur Dubrawski) Date: Fri, 15 Jul 2016 11:44:18 -0400 Subject: CMU Auton Lab publicly recognized by a Department of Homeland Security official Message-ID: Team, I was surprised but very pleased to hear one of our projects showcased as the sole template of a success story in a plenary talk by Dr. Joel Rynes, Assistant Director for R&D at the U.S. Department of Homeland Security, at the DNDO ARI conference earlier this week. This was completely unexpected, because the government officials at this level of responsibility typically refrain from mentioning any vendor or partner names explicitly in their public appearances. In the past, we had prepared numerous batches of presentation and press-releasable materials for the use by our government sponsors, from which we had to remove any and all references to CMU, the Robotics Institute, or the Auton Lab. Dr. Rynes basically said that everyone in the R&D community working in the space of radiation safety should look at how our CMU team has developed the foundations of ERNIE system and how the results of our research are being successfully transitioned to the U.S. Customs practice, and how significant difference it will make when the system is fully deployed throughout the organization. Hats go down above all to Saswati who continues to deliver exceptional implementations of our algorithms and who makes our partners from Lawrence Livermore National Laboratory and various groups of our sponsors and end-users very pleased with them. Cheers! Artur PS ERNIE stands for " Enhanced Radiological Nuclear Inspection and Evaluation". This is the project in which we look at the data collected using Radiation Portal Monitors at the U.S. ports of entry, and the primary objective is to combine physics of radiation models and machine learning to allow dismissing a large number of alerts that are generated by non-threatening sources of radiation, while never missing a real threat. -------------- next part -------------- An HTML attachment was scrubbed... URL: From awm at cs.cmu.edu Fri Jul 15 12:32:48 2016 From: awm at cs.cmu.edu (Andrew Moore) Date: Fri, 15 Jul 2016 12:32:48 -0400 Subject: CMU Auton Lab publicly recognized by a Department of Homeland Security official In-Reply-To: References: Message-ID: Can we talk about this one publicly? I think we could get good venues for this one. On Fri, Jul 15, 2016 at 11:44 AM, Artur Dubrawski wrote: > Team, > > I was surprised but very pleased to hear one of our projects showcased as > the sole template of > a success story in a plenary talk by Dr. Joel Rynes, Assistant Director > for R&D at the U.S. Department > of Homeland Security, at the DNDO ARI conference earlier this week. > > This was completely unexpected, because the government officials at this > level of responsibility > typically refrain from mentioning any vendor or partner names explicitly > in their public appearances. > In the past, we had prepared numerous batches of presentation and > press-releasable materials for > the use by our government sponsors, from which we had to remove any and > all references to CMU, > the Robotics Institute, or the Auton Lab. > > Dr. Rynes basically said that everyone in the R&D community working in the > space of radiation safety > should look at how our CMU team has developed the foundations of ERNIE > system and how the results of > our research are being successfully transitioned to the U.S. Customs > practice, and how significant difference > it will make when the system is fully deployed throughout the organization. > > Hats go down above all to Saswati who continues to deliver exceptional > implementations of our > algorithms and who makes our partners from Lawrence Livermore National > Laboratory and various > groups of our sponsors and end-users very pleased with them. > > Cheers! > Artur > > PS ERNIE stands for " Enhanced Radiological Nuclear Inspection and > Evaluation". > This is the project in which we look at the data collected using Radiation > Portal Monitors > at the U.S. ports of entry, and the primary objective is to combine > physics of radiation models > and machine learning to allow dismissing a large number of alerts that are > generated > by non-threatening sources of radiation, while never missing a real > threat. > -- Andrew Moore , Dean, School of Computer Science , Carnegie Mellon. Twitter feed -------------- next part -------------- An HTML attachment was scrubbed... URL: From bspice at cs.cmu.edu Fri Jul 15 13:35:38 2016 From: bspice at cs.cmu.edu (Byron Spice) Date: Fri, 15 Jul 2016 13:35:38 -0400 Subject: CMU Auton Lab publicly recognized by a Department of Homeland Security official In-Reply-To: References: Message-ID: <016501d1debf$4dc5e630$e951b290$@cs.cmu.edu> Artur says he will get me in touch with the right people at the DNDO. Byron Spice 412 268 9068 From: awmcmu at gmail.com [mailto:awmcmu at gmail.com] On Behalf Of Andrew Moore Sent: Friday, July 15, 2016 12:33 PM To: Artur Dubrawski Cc: users at autonlab.org; hebert; Byron Spice; Manuela Veloso Subject: Re: CMU Auton Lab publicly recognized by a Department of Homeland Security official Can we talk about this one publicly? I think we could get good venues for this one. On Fri, Jul 15, 2016 at 11:44 AM, Artur Dubrawski wrote: Team, I was surprised but very pleased to hear one of our projects showcased as the sole template of a success story in a plenary talk by Dr. Joel Rynes, Assistant Director for R&D at the U.S. Department of Homeland Security, at the DNDO ARI conference earlier this week. This was completely unexpected, because the government officials at this level of responsibility typically refrain from mentioning any vendor or partner names explicitly in their public appearances. In the past, we had prepared numerous batches of presentation and press-releasable materials for the use by our government sponsors, from which we had to remove any and all references to CMU, the Robotics Institute, or the Auton Lab. Dr. Rynes basically said that everyone in the R&D community working in the space of radiation safety should look at how our CMU team has developed the foundations of ERNIE system and how the results of our research are being successfully transitioned to the U.S. Customs practice, and how significant difference it will make when the system is fully deployed throughout the organization. Hats go down above all to Saswati who continues to deliver exceptional implementations of our algorithms and who makes our partners from Lawrence Livermore National Laboratory and various groups of our sponsors and end-users very pleased with them. Cheers! Artur PS ERNIE stands for " Enhanced Radiological Nuclear Inspection and Evaluation". This is the project in which we look at the data collected using Radiation Portal Monitors at the U.S. ports of entry, and the primary objective is to combine physics of radiation models and machine learning to allow dismissing a large number of alerts that are generated by non-threatening sources of radiation, while never missing a real threat. -- Andrew Moore , Dean, School of Computer Science , Carnegie Mellon. Twitter feed -------------- next part -------------- An HTML attachment was scrubbed... URL: From awd at cs.cmu.edu Fri Jul 29 14:57:47 2016 From: awd at cs.cmu.edu (Artur Dubrawski) Date: Fri, 29 Jul 2016 14:57:47 -0400 Subject: Nick receives Presidential Fellowship Message-ID: <64973a0e-3451-8a1c-1e9c-3055a0d5769c@cs.cmu.edu> Dear Autonians, Please join me in congratulating our distinguished colleague Nick Gisolfi on his receipt of the 2016 CMU Presidential Fellowship Award! This great honor for Nick and our team comes with an added bonus of relieving the Lab of Nick's stipend and tuition costs for the whole year, so the next step for us is to make sure every Auton Lab graduate student receives the same award, in perpetuum, and our budget would look really really stable :) Cheers to Nick! Artur From bapoczos at cs.cmu.edu Fri Jul 29 15:03:13 2016 From: bapoczos at cs.cmu.edu (Barnabas Poczos) Date: Fri, 29 Jul 2016 15:03:13 -0400 Subject: Nick receives Presidential Fellowship In-Reply-To: <64973a0e-3451-8a1c-1e9c-3055a0d5769c@cs.cmu.edu> References: <64973a0e-3451-8a1c-1e9c-3055a0d5769c@cs.cmu.edu> Message-ID: This is awesome! Congrats Nick! Best, Barnabas ====================== Barnabas Poczos, PhD Assistant Professor Machine Learning Department Carnegie Mellon University On Fri, Jul 29, 2016 at 2:57 PM, Artur Dubrawski wrote: > Dear Autonians, > > Please join me in congratulating our distinguished colleague Nick Gisolfi on > his receipt of the 2016 CMU Presidential Fellowship Award! > > This great honor for Nick and our team comes with an added bonus of > relieving the Lab of Nick's stipend and tuition costs for the whole year, so > the next step for us is to make sure every Auton Lab graduate student > receives the same award, in perpetuum, and our budget would look really > really stable :) > > Cheers to Nick! > > Artur