(eGPU eGFX Dock).1: Please no selling/trading of eGPUs, video cards, power supplies and other hardware.
Bizonbox 2016 Mac Book ProThe 2016 MacBook Pro is the most versatile game development machine. Multiplatform: MacBookPro can run and develop both Mac, Windows, Linux, iOS and Android software. (Linux drivers were not available when this was written but that will change, virtual machine Linux until native tri-boot support drivers are available).Download Citrix Workspace app. Citrix Workspace app is the easy-to-install client software that provides seamless secure access to everything you need to get work done.0: Posts must be related to eGPUs (External Video Cards).The BizonBOX eGPU is available in two versions depending on the age of your Apple computer, BizonBOX 2S for machines made between 20, and BizonBOX 3, compatible with the Mac Book Pro 2016 and, hopefully, the Mac Book Pro 2017 when the overdue big beef-up MBP upgrade is finally here sometime later this year.Presenting BizonBOX external graphics c.I would like to know what the external GPU (eGPU) options are for macOS in 2017 with the late 2016 MacBook Pro.I did my research, however on the internet I find a lot of confusing information. Discussing a sale in comments is not allowed - Take it to PMs or to the linked sale medium. We typically lock approved links to sales to enforce the no-discussion rule.2: No sensationalized, misleading or non-descriptive titles.3: Keep discussions civil and respectful. Comments should be on-topic and contribute to the conversation.4: No short, ref and affiliate links. Automod is pretty aggressive here.People suggested renting server space instead, or using Windows (better graphics card support) or even building a new PC for the same price that allows you to use a eGPU on Mac. My question therefore is, is it possible to use TensorFlow with CUDA and eGPU on the late MacBook Pro 2016 (15")? I want to use the graphics card in macOS (with late MacBook Pro 15") as an eGPU (no dual-boot/Windows/Linux partition).Side note: I have seen users making use of eGPU's on macbook's before (Razor Core, AKiTiO Node), but never in combination with CUDA and Machine Learning (or the 1080 GTX for that matter). However, I do not know all the things that are important to get it to work. My only purpose is to use Keras and TensorFlow with it. Ideally, I would like to run the 1080 GTX of NVIDIA. Others say, it can only work for the older graphics cards as CUDA is not supported for the newer graphics cards (GTX 1080).GPU Driver Version: 10.18.5 (378.05.05.25f01) Apple Thunderbolt3 to Thunderbolt2 Adapter MacBook Pro (Retina, 13-inch, Early 2015) LD_LIBRARY_PATH: /usr/local/cuda/lib:/usr/local/cuda:/usr/local/cuda/extras/CUPTI/libI tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcuda.dylib locallyI tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcurand.8.0.dylib locallyI was able to get a NVIDIA GTX 1080 Ti working on the Akitio Node on my iMac (late 2013). Downloading and installing cuDNN (cudnn-8.0-osx-x64-v5.1) Need to register before downloading itCopy cuDNN files to CUDA cd ~/Downloads/cudaSudo cp include/* /usr/local/cuda/include/Create envirenment and install tensorflow conda create -n egpu python=3Run the following script: import tensorflow as tfA = tf.constant(, shape=, name='a')B = tf.constant(, shape=, name='b')Install Keras in the envirenment and set tensorflow as backend: pip install -upgrade -no-deps keras # Need no-deps flag to prevent from installing tensorflow dependencyKERAS_BACKEND=tensorflow python -c "from keras import backend"I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcublas.8.0.dylib locallyI tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcudnn.5.dylib locallyI tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcufft.8.0.dylib locallyI tensorflow/stream_executor/dso_loader.cc:126] Couldn't open CUDA library libcuda.1.dylib. This is executed everytime you open a terminal window) Download and installing CUDA 8.0 CUDA Toolkit 8.0 GA2 (Feb 2017)Export DYLD_LIBRARY_PATH="$CUDA_HOME/lib:$CUDA_HOME:$CUDA_HOME/extras/CUPTI/lib"Export LD_LIBRARY_PATH=$DYLD_LIBRARY_PATH(If your bash_profile does not exist, create it. Cadillacs and dinosaurs android apk full downloadRun the automate-eGPU.sh script (with sudo) that everybody at eGPU.io seems to rely on. It's needed by the eGPU.sh script and later also by TensorFlow. Disable SIP (Google for a tutorial). Install CUDA according to official documentation. These tutorials are mostly for accelerating a display with an eGPU, though for training NNs you don't obviously need to follow all the steps. This TensorFlow installation didn't work with Jupyter, though, but hopefully there's a workaround for that.I haven't used this set-up much so not sure about the performance increase (or bandwidth limitations), but eGPU + TensorFlow/CUDA certainly is possible now, since NVIDIA started releasing proper drivers for macOS.EGPU support on macOS is a difficult topic, but I will do my best to answer your question.Let's begin with graphics cards! For the sake of time, and because we're talking CUDA, we'll stick with Nvidia cards. I also tried compiling TensorFlow from source, which didn't work before I set the env vars as described in the previous step.From iStat Menus I can verify that my external GPU is indeed used during training. Turns out there are some environment variables that have to be set (a bit differently than the CUDA installer suggests), which were described in a GitHub issue comment. When installing with pip install tensorflow-gpu, I had no installation errors, but got a segfault when requiring TensorFlow in Python. Uninstall CPU-only TensorFlow and install one with GPU support. On any Mac operating system. GTX 10xx cards WILL NOT WORK. However, the Nvidia graphics drivers actually work on almost all of Nvidia's GeForce and Quadro cards, with one big exception. Apple, however, only officially supports a few Nvidia graphics cards, mainly very old ones. I don't believe you can run the eGPU without an external monitor in macOS. From what I understand, to utilize the power of the eGPU, you need to plug an external monitor into the eGPU. I did a bit of eGPU researching towards the end of 2016, and the information I got was relatively confusing, so if anyone has any comments or corrections, please let me know. Well, that's easier said than done. I'm going to assume, because you mentioned specifically eGPUs, that you've budgeted for an actual eGPU enclosure (let's use the AKiTiO Node as an example), instead of a PCIe expansion chassis with an external power supply, as this is not a great idea.So now we have a graphics card (GTX 980Ti) in an eGPU enclosure (AKiTiO Node) and we want to get it to work. If you're looking for power, you'll want to look at the GTX 980Ti or Titan X (many good Quadro cards would also work well).Now that we've got that covered, let's move onto eGPU enclosures.
0 Comments
Leave a Reply. |
AuthorChristine ArchivesCategories |