I'm new to deep learning. I wanted to build an image classifier using CNN to classify clothing images. I decided to train over the fashion MNIST-dataset which is a dataset of 60,000 images. But I'm aware that training is a very heavy task.
I wanted to know how long will my PC take to train over this dataset and should I go for pre-trained models instead with a compromise of accuracy.
My PC configurations are:
- Intel Core i5-6400 CPU # 2.70 GHz
- 8GB RAM.
- NVIDIA GeForce GTX 1050 Ti.
Even though it depends on data-set size & number of EPOCS(i tried with 50 Epocs) ,here it is small that is 32x32.
So for me when i tried on a machine with
Intel Core i7-6400 CPU # 2.70 GHz
8GB RAM.
NVIDIA GeForce GTX 1050 Ti.
with image size(28x28) as provided in MNIST dataset in Tensorflow.org it took less than 5 minutes.
Related
Is anyone aware of how long the training took to achieve the mAP and FPS of YoloV4 on the MS COCO dataset as referenced in https://github.com/AlexeyAB/darknet and the corresponding paper: https://arxiv.org/abs/2004.10934.
Trying to estimate training time and final mAP for RTX 3050 without running the full training, as it is projected to take ~1500 hours.
Have not been able to find any stats for how long the training took for the Tesla V100 and RTX 2070 referenced in the paper.
Ideally could take length of time RTX 2070 took and scale according to the difference in BFLOPS, and assume accuracy is roughly similar.
I have two laptops, both with Windows 10, that I use for work:
MSI GE70: i7 4720, 12 GB Ram, GTX 960m 2GB, 258 GB SSD.
Dell G7: i7 9750, 32 GB Ram, RTX 2070 Max-Q 8Gb, 500 GB SSD.
I made a 'mirror' installation of TensorFlow in both laptops following the official TensorFlow page.
In both laptops I installed Python 3.6.8, TensorFlow 2.2, CUDA 10.1, cuDNN 7.6 and 456.71 Nvidia Driver version. When I run the following line in CMD I can see that both GPUs are visibles to TensorFlow and ready to use.
python -c "import tensorflow as tf;print(tf.reduce_sum(tf.random.normal([1000, 1000])))
MSI with 960m
Dell with 2070 Max-Q
Then, when I train the same Neural Network in both laptops, I can see that the MSI takes 7 minutes per epoch, while the Dell G7 takes almost an one hour per epoch. Why the GPU 2070 Max-Q takes so longer time for train the Neural Network in comparison with the 960m? There is some problem with the Dell G7 that I can't see?
This is the structure of the NN:
modelo=Sequential()
modelo.add(Bidirectional(LSTM(units=na, return_sequences=True),input_shape=dim_entrada))
modelo.add(Dropout(0.25))
modelo.add(Bidirectional(LSTM(units=na)))
modelo.add(Dropout(0.25))
modelo.add(Dense(units=3))
opt = tf.optimizers.Adam(learning_rate=0.0015)
modelo.compile(optimizer=opt, loss='mse', metrics=['accuracy'])
modelo.fit(X_train,Y_train,epochs=20,batch_size=32,validation_data=(X_validacion_imu12,Y_validacion_vi12))
I found the problem. I don't know why but the Dell G7 must be plugged into electricity. I think is a power option that prevents the use of the GPU if it is not plugged in.
I'm running a Mask R-CNN model on an edge device (with an NVIDIA GTX 1080). I am currently using the Detectron2 Mask R-CNN implementation and I archieve an inference speed of around 5 FPS.
To speed this up I looked at other inference engines and model implementations. For example ONNX, but I'm not able to gain a faster inference speed.
TensorRT looks very promising to me but I did not found a ready "out-of-the-box" implementation for it.
Are there any other mature and fast inference engines or other techniques to speed up the inference?
It's almost impossible to get higher inference speed for Mask R-CNN on GTX 1080. You may check detectron2 by Facebook AI Research.
Otherwise, I'd suggest to use YOLACT - (You Only Look At CoefficienTs), it can achieve real-time instance segmentation.
On the other hand, if you don't need instance segmentation, you can use YOLO, SSD, etc for object detection.
OpenCV 4.5.0 with DNN_BACKEND_CUDA and DNN_TARGET_CUDA/DNN_TARGET_CUDA_FP16.
Mask RCNN with 1024 x 1024 input image
Device | FPS
------------------ | -------
GTX 1080 Ti (FP32) | 29
RTX 2080 Ti (FP16) | 60
FPS measured includes NMS but excludes other preprocessing and postprocessing. The network fully runs end-to-end on GPU.
Benchmark code: https://gist.github.com/YashasSamaga/48bdb167303e10f4d07b754888ddbdcf
As #kkHarshit already mentioned it is very hard to speed up a Mask R-CNN any further.
The fastest instance segmentation model that I found is YolactEdge: Real-time Instance Segmentation on the Edge (Jetson AGX Xavier: 30 FPS, RTX 2080 Ti: 170 FPS).
It's perfomance is worse than Mask R-CNN or Yolact even but still very good.
I am running deep learning CNN (4-CNN layers and 3 FNN layers) model (written in Keras with tensorflow as backend) on two different machines.
I have 2 machines (A: with a GTX 960 graphics GPU with 2GB memory & clock speed: 1.17 GHz and B: with a Tesla K40 computation GPU with 12GB memory & clock speed: 745MHz)
But when I run the CNN model on A:
Epoch 1/35
50000/50000 [==============================] - 10s 198us/step - loss: 0.0851 - acc: 0.2323
on B:
Epoch 1/35
50000/50000 [==============================] - 43s 850us/step - loss: 0.0800 - acc: 0.3110
The numbers are not even comparable. I am quite new to deep learning and running code on GPUs. Could someone please help me explain why the numbers are so different?
Dataset: CIFAR-10 (32x32 RGB images)
Model batch size: 128
Model number of parameters: 1.2M
OS: Ubuntu 16.04
Nvidia driver version: 384.111
Cuda version: 7.5, V7.5.17
Please let me know if you need any more data.
Edit 1: (adding CPU info)
Machine A (GTX 960): 8 cores - Intel(R) Core(TM) i7-6700 CPU # 3.40GHz
Machine B (Tesla K40c):8 cores - Intel(R) Xeon(R) CPU E5-2637 v4 # 3.50GHz
TL;DR: Measure again with a larger batch size.
Those results do not surprise me much. It's a common mistake to think that an expensive Tesla card (or a GPU for that matter) will automatically do everything faster. You have to understand how GPUs work in order to harness their power.
If you compare the base clock speeds of your devices, you will find that your Xeon CPU has the fastest one:
Nvidia K40c: 745MHz
Nvidia GTX 960: 1127MHz
Intel i7: 3400MHz
Intel Xeon: 3500MHz
This gives you a hint of the speeds at which these devices operate and gives a very rough estimate of how fast they can crunch numbers if they would only do one thing at a time, that is, with no parallelization.
So as you see, GPUs are not fast at all (for some definition of fast), in fact they're quite slow. Also note how the K40c is in fact slower than the GTX 960.
However, the real power of a GPU comes from its ability to process a lot of data simultaneously! If you now check again at how much parallelization is possible on these devices, you will find that your K40c is not so bad after all:
Nvidia K40c: 2880 cuda cores
Nvidia GTX 960: 1024 cuda cores
Intel i7: 8 threads
Intel Xeon: 8 threads
Again, these numbers give you a very rough estimate of how many things these devices can do simultaneously.
Note: I am severely simplifying things: In absolutely no way is a CPU core comparable to a cuda core! They are very very different things. And in no way can base clock frequencies be compared like this! It's just to give an idea of what's happening.
So, your devices needs to be able to process a lot of data in parallel in order to maximize their throughput. Luckily tensorflow already does this for you: It will automatically parallelize all those heavy matrix multiplications for maximum throughput. However this is only going to be fast if the matrices have a certain size. Your batch size is set to 128 which means that almost all of these matrices will have the first dimension set to 128. I don't know the details of your model, but if the other dimensions are not large either, then I suspect that most of your K40c is staying idle during those matrix multiplications. Try to increase the batch size and measure again. You should find that larger batch sizes will make the K40c faster in comparison with the GTX 960. The same should be true for increasing the model's capacity: increase the number of units in the fully-connected layers and the number of filters in the convolutional layers. Adding more layers will probably not help here. The output of the nvidia-smi tool is also very useful to see how busy a GPU really is.
Note however that changing the model's hyper-parameter and/or the batch size will of course have a huge impact on how the model is able to train successfully and naturally you might also hit memory limitations.
Perhaps if increasing the batch size or changing the model is not an option, you could also try to train two models on the K40c at the same time to make use of the idle cores. However I have never tried this, so it might not work at all.
I have Tensorflow 1.4 GPU version installed. Cuda8 is installed too.
I trained my pretty simple GAN network on MNIST data.
I have AMD FX 8320 CPU, 16Gb system memory and SSD hard drive.
It took about 17 seconds per epoch on GeForce 720 GPU with 1GB memory.
The training utilized about 25% of GPU and 99% of memory. CPU was loaded prettyhigh, close to 100%.
Then I inserted other video board with GeForce1050 Ti GPU and 4Gb memory instead of previous. The GPU was loaded only for 5-6%, memory was utilized for 93%.
But I still got about 17s per epoch and high load for CPU.
So maybe Tensorflow has some settings to utilize more GPU?
Or what is a cause of high CPU load and low GPU load?
If you are training a simple GAN network it is fairly likely that your old GPU was not the bottleneck in the first place. So, improving it had no effect. If the amount of work done per sess.run() call very small, the overheads (executing your Python code, copying the input data to GPU, starting and running the TensorFlow executor, scheduling all the operations to GPU, etc) can dominate your computation.
The only sure way of knowing what happens is to profile. You can take a look here https://www.tensorflow.org/performance/performance_guide as a starting point. The timeline tool it mentions can be fairly useful. See here for more details: Can I measure the execution time of individual operations with TensorFlow?.
Agree, for MNIST datasets, there are probably other bottlenecks in the system, not the GPU. I ran 2 side-by side TensorFlows,
Intel i7 4600M with NVIDIA Quadro K1100M GPU and 12 GB RAM, which is a 4th Gen Haswell Intel machine, and
Intel i5 8300U with No Cuda GPU and 16GB of RAM.
Basically 8th Gen Kaby Lake Intel CPU vs 4th Gen Intel, and I got:
4th Gen Intel chip with NVIDIA GPU:
311.5 sec, 315.9 sec, 313.0 sec to complete all 10 epocs on a MNIST run
8th Gen Intel chip with no GPU:
252.7 sec, 243.5 sec, 254.9 sec
So I'm running 20% faster with no GPU, just a newer generation of Intel chip.