最新 | 最热门 | 最高评价

+0  Some tips about using google’s TPU

Tag: machine learning | tensorflow | TPU
Robin Dong 发于 2018年09月21日 16:22 | 点击: 218 | 展开摘要
About one month ago, I submit a request to Google Research Cloud for using TPU for free. Fortunately, I received the approvement yesterday. The approvement let me use 5 regular Cloud TPUs and 100 preemptible Cloud TPUs for free for 30 days

查看全文: http://www.udpwork.com/item/17097.html

+0  Some modifications about SSD-Tensorflow

Tag: machine learning | tensorflow
Robin Dong 发于 2018年09月13日 10:54 | 点击: 412 | 展开摘要
In the previous article, I introduced a new library for Object Detection. But yesterday, after I added slim.batch_norm() into ‘nets/ssd_vgg_512.py’ like this:

def ssd_arg_scope(weight_decay=0.0005, data_format='NHWC', is_traini

查看全文: http://www.udpwork.com/item/17088.html

+0  Choosing a Object Detection Framework written by Tensorflow

Tag: machine learning | Object Detection | tensorflow
Robin Dong 发于 2018年09月07日 15:23 | 点击: 402 | 展开摘要
Recently I need to train a DNN model for object detection task. In the past, I am using the object detection framework from tensorflows’s subject — models. But there are two reasons that I couldn’t use it to do my own proj

查看全文: http://www.udpwork.com/item/17081.html

+0  Do tf.random_crop() operation on GPU

Tag: machine learning | tensorflow
Robin Dong 发于 2018年07月27日 15:12 | 点击: 93 | 展开摘要
When I run code like:

with tf.device('/GPU:0'):
images = tf.random_crop(images, [IMAGE_HEIGHT, IMAGE_WIDTH, IMAGE_CHANNELS])
...
it reports:

Could not satisfy explicit device specification '/device:GPU:0' because no supported kernel for

查看全文: http://www.udpwork.com/item/17017.html

+0  Regularization loss in ‘slim’ library of Tensorflow

Tag: machine learning | tensorflow
Robin Dong 发于 2018年07月19日 15:16 | 点击: 104 | 展开摘要
My python code using slim library to train classification model in Tensorflow:

with tf.contrib.slim.arg_scope(mobilenet_v2.training_scope(weight_decay = 0.001)):
logits, _ = mobilenet_v2.mobilenet(images, NUM_CLASSES)

cross_entr

查看全文: http://www.udpwork.com/item/17018.html

+0  Using multi-GPUs for training in distributed environment of Tensorflow

Tag: machine learning | tensorflow
Robin Dong 发于 2018年07月13日 16:31 | 点击: 88 | 展开摘要
I am trying to write code for training on multi-GPUs. The code is mainly from the example of ‘Distributed Tensorflow‘. I have changed the code slightly for runing on GPU:

...
tf.train.replica_device_setter(
worker_devic

查看全文: http://www.udpwork.com/item/17019.html

+0  Problems and solutions about building Tensorflow-1.8 with TensorRT 4.0

Tag: machine learning | tensorflow | TensorRT
Robin Dong 发于 2018年06月29日 10:41 | 点击: 91 | 展开摘要
Problem:

When compiling Tensorflow-1.8 with CUDA-9.2, it reports:

bazel-out/host/bin/_solib_local/_U_S_Stensorflow_Spython_Cgen_Unn_Uops_Upy_Uwrappers_Ucc___Utensorflow/libtensorflow_framework.so: undefined reference to `cublasGemmEx@libc

查看全文: http://www.udpwork.com/item/17021.html

+0  Testing performance of Tensorflow’s fixed-point-quantization on x86_64 cpu

Tag: machine learning | quantization | tensorflow
Robin Dong 发于 2018年06月22日 11:49 | 点击: 100 | 展开摘要
Google has published their quantization method on this paper. It use int8 to run feed-forward but float32 for back-propagation, since back-propagation need more accurate to accumulate gradients. I got a question right after reading the pape

查看全文: http://www.udpwork.com/item/17022.html
|<<<1>>>| 一共1页, 8条记录