# Tag Info

## Hot answers tagged hardware-evaluation

7

The development of CPUs didn't quite keep up with Kurzweil's predictions. But if you also allow for GPUs, his prediction for 2009 is pretty accurate. I think Moore's law slowed down recently and has now been pretty much abandoned by the industry. How much that will affect the 2019 prediction remains to be seen. Maybe the industry will hit its stride again ...

7

Regarding specific choices I can't recommend, but if you are completely new, you should probably learn/code some more until you get a GPU. There is a lot to learn in machine learning before GPU speedups make a significant difference, and until then doing the computations on any old CPU would be just fine, especially if you are just starting since you won't ...

6

Usually the problem is to fit the model into video RAM. If it does not, you cannot train your model at all without big efforts (like training parts of the model separately). If it does, time is your only problem. But the difference in training time between consumer GPUs like the Nvidia 1080 and much more expensive GPU accelerators like the Nvidia K80 are not ...

5

The main reason that AMD Radeon graphics card is not used for deep learning is not the hardware and raw speed. Instead it is because the software and drivers for deep learning on Radeon GPU is not actively developed. NVIDIA have good drivers and software stack for deep learning such as CUDA, CUDNN and more. Many deep learning library also have CUDA support. ...

5

Given that you're a student doing this out of personal interest and wanting to do some gaming on the side, I'd suggest the GTX 1060 6GB since at present the GTX 1070Ti is overpriced due to crypto miners (this will date the answer, but for reference the 1060 is going for ~GBP340, the 1070Ti for ~GBP600; two other options are the 1050Ti 4GB for ~GBP160 or the ...

4

I think you should redirect your focus. Learn and play with ML, and only when compute becomes the main bottleneck for your learning, invest in hardware. I've recently engaged into a ML project for which I 've assembled a machine with gtx1080, installed gui-less ubuntu, configured ssh, drivers etc, and than spend 3 months on data collection. And the dataset ...

3

One answer is infinite amount of time because it can always be better. Another answer is: 10k for training set A PC with a GPU (3~4k USD), google colab (10 USD per month), or other cloud service (probably more expensive than colab) One developer, 1 day lol Two kinds is easier than multiple kinds There is no paper that seeks to answer your question the way ...

3

Lets answer the question assuming ideal case. Say you are training a deep learning algorithm in which p proportion of the algorithm is parllelizable and (1-p) proportion as sequential part . Lets assume you can perfectly divide the program into their parallelizable and sequential parts. I don't exactly know the specs of aforementioned CPU's, but lets assume ...

3

As a caveat, I’d suggest that unless you’re pushing up against fundamental technological limits, computation speed and resources should be secondary to design rationale when developing a neural network architecture. That said, earlier this year I finished my MS thesis that involved bioinformatics analytics pipelines with whole genome sequencing data - that ...

3

I don't think you need to invest in any kind of GPU unless you're familiar with the computations required for the task you want to achieve using deep learning. Also, by the time you've sufficiently mastered Deep Learning to a point where you can actually make the most of your GPU, there will be new products in the market. So until then I suggest you use ...

2

Yes, we do have computing systems that do fall in the teraFLOPS range (where 1 teraflop = 1 trillion FLOPS = $10^{12}$ FLOPS) The human brain is a biological system and saying it has some sort of FLOPS ability is just plain dumb because there is no way to take a human brain and measure its FLOPS. You could say "hey, by looking at the neurons activity ...

2

It depends on what you need. You can train any size of network on any resource. The problem is the time of training. If you want to train Inception on an average CPU it will take months to converge. So, it all depends on how long you can wait to see your results based on your network. As in neural nets we do not have only one operation but many (like ...

2

Specialized AI hardware takes advantage of highly parallelizable nature of many neural network designs. GPUs are designed for pixel crunching - coincidentally very paralelisable too. This is why they often offer orders of magnitude better performance (in NN tasks) than CPUs. That being said, GPUs have shortcomings to. First, many of GPU's built in features ...

2

Parallel processing is very important when computing ANNs. Though they are not designed for AIs, GPUs are widely used in machine learning because of their higher capability of paralel processing and it is due to their higher number of cores comparing CPUs. All in all, the dedicated hardware should have GPU-like architecture.

2

I skimmed through your question and understood that the state/action space is finite, so in this case, RL would be a good option for storage. The most basic RL technique will keep track of a matrix Q ∈ ℝs×a, where s is number of possible states, and a is number of possible actions. In addition to a small overhead of agent's parameters: &...

2

First off I would like to post this comprehensive blog which makes comparison between all kinds of NVIDIA GPU's. The most popular deep learning library TensorFlow by default uses 32 bit floating point precision. The choice is made as it helps in 2 causes: Lesser memory requirements Faster calculations 64 bit is only marginally better than 32 bit as very ...

2

The answer is it depends. Are you referring to GPU vs CPU when training, or running inference? When you say CPU is it an x86 CPU or ARM CPU? What algorithms are running? Let's look at image classification. Suppose you select a Convolutional Neural Network(CNN) as your algorithm. CNN's were created in the 1980s, but only really saw use when GPU's were used. ...

2

The Mac Pro uses AMD GPUs. These don't support CUDA, but instead support the OpenCL framework. TensorFlow, the most popular deep learning library, uses CUDA to run on GPUs, although OpenCL support is in the works. That said, one of the main approaches to OpenCL support, SYCL, isn't planning to support OSX: We have no plans to support OSX in near future. ...

2

For a simulation of SNES or N64, the resolution is probably not that high. You can use either online credits or buy new hardware. For online credits, it is recommended if you do teh simulation/training for only several dozens hours, as eache hour costs around 5-10 dollars. AWS is a good choice as it have a large variety of choice for hardware. For higher ...

2

If you are just starting out with Deep Learning, then a laptop with GTX 1060 is enough. I am using a GTX 1060 myself and I find it adequate for many of my personal projects, training large datasets and participating in most (not all) Kaggle competitions as well. But you say that you want to do research work as well. In that case, you may want to contact ...

1

Deep models are very tolerant to arithmetic underflow. You can hope for neglectable differences in prediction accuracy between FP32 and FP16 models. Check this paper for concrete results.

1

Your hardware choice depends majorly on the sensors and camera specifications you need for your solution. For example, we are running a 2536x1920 resolution 20fps camera to detect cars and then read their number plates. Nvidia's RTX 2080Ti 11 GB can handle 2 such cameras at 20fps (real-time). However, this involves 2 models: 1 for detection of plates and a ...

1

Preface "Before answering this question, let me preface by stating that the following is simply MY answer as a Machine Learning Researcher and "Hobbyist" Theoretical Physicist, although I have strong feelings that my answer will most certainly be proven as true, I am more than sure that others will have differing opinions as with everything else ...

1

It might be hard to implement deep reinforcement learning algorithms, especially considering your previous experience and the computing resources you have. They require almost the same (even more) GPU power. Deep reinforcement learning algorithms use deep neural networks for learning the optimal policy. Even if you are given appropriate resources, it would ...

1

The impact of the number of cores, cache size, and clock speed on GPU performance when comparing the ones you mentioned is not project critical when doing either lab research or deploying to the field. More important is the optimization of algorithms to the hardware available, video bus architecture (since data passing between CPUs and GPUs is dependent ...

1

That is a too broad question. It depends on what kind of operations will you do with CPU during DL training. Generally, data reading (from HDD), pre/post processing are being done on CPU while DL training is done at GPU. There are only two things that you should care: Queue system while reading/pre-processing your input. So that when GPU finishes one ...

Only top voted, non community-wiki answers of a minimum length are eligible