Stack Overflow
Stack Overflow

Questions tagged [imagenet]

200 questions
0
votes
1 answer

Cannot run imagenet download and preprocess script (suggestion in issue 202 did not work)

I am following the instruction here: https://github.com/tensorflow/models/tree/master/inception After running bazel-bin/inception/imagenet_train --num_gpus=1 --batch_size=32 --train_dir=/tmp/imagenet_train --data_dir=/tmp/imagenet_data I get the…
phojjat
  • 1
0
votes
1 answer

Why are my inception-v3 (in Keras) predictions all wrong?

I am an ML beginner and simply implementing inception-v3 using the ImageNet weights. This is my first run at it. My implementation is in Keras. My predictions are all wrong and I need a little foot up, to see what the problem is. It is actaully…
GhostRider
  • 2,109
  • 7
  • 35
  • 53
0
votes
0 answers

Cannot find mean file while training on Imagenet

I am trying to train and validate a network on Imagenet. The validation process works without any problems (with the pretrained weights). However, when I try to perform the training, there appears an error that the imagenet_mean.binaryproto file is…
cerebrou
  • 5,353
  • 15
  • 48
  • 80
0
votes
1 answer

caffe convert_imageset pending

When I convert ImageNet dataset to lmdb, convert_imageset will suspend and no longer process image data.Can you tell me how to solve this problem? I have tried it many time.I checked the log. Every time the number of images that was processed is…
Tyan
  • 1
0
votes
1 answer

Change the output layer of AlexNet/GoogleNet/ImageNet?

I have a question regarding changing the output layer of the nets (AlexNet/GoogleNet/ImageNet). So the standard output is a 1x1000 Vector so one value for each class. I know I can change the output to e.g 5 so I would get 1x5 Vector, if I only have…
Biba
  • 1,595
  • 1
  • 12
  • 17
0
votes
1 answer

Cannot predict input from ImageNet classes [Keras + Tensorflow]

I followed this repo (https://github.com/iamgroot42/keras-finetuning), I've done with the training. Now, I want to predict my input image both of my own dataset (containing 2 classes, Avocado & Mango) and ImageNet set. But the prediction result…
Justinus Hermawan
  • 1,194
  • 1
  • 23
  • 44
0
votes
0 answers

How to make Inception V3 model defined with slim do prediction on GPU

I use the Inception V3 model defined by the slim to predict my own images. The model variables are restored from a checkpoint file. When I run it, I find that GPU is not used. I have set tf.device('/gpu:0'), but it does not work. So, I am confused…
songqiang
  • 31
  • 4
0
votes
1 answer

Test accuracy cannot improve when learning ZFNet on ILSVRC12

I've implemented a home-brewed ZFNet (prototxt) for my research. After 20k iterations with the definition, the test accuracy stays at ~0.001 (i.e., 1/1000), the test loss at ~6.9, and training loss at ~6.9, which seems that the net keeps playing…
stoneyang
  • 179
  • 1
  • 2
  • 12
0
votes
1 answer

Does Caffe "ImageNet" support .jpg or .bmp images?

Does Caffe "ImageNet" support .jpg or .bmp images as input? In ILSVRC2012, I see only .JPEG files. I tried to train models on .jpg and .bmp files (different train for each type), but it doesn't give reasonable results.
Z.Kal
  • 426
  • 4
  • 18
0
votes
1 answer

Why does googlenet (inception) work well on the ImageNet dataset?

Some people said that the reason that inception works well on the ImageNet dataset is that:the original images in the ImageNet dataset have different resolutions, and they are resized to the same size when they are used. So the inception which can…
lu tingting
  • 41
  • 1
  • 3
0
votes
0 answers

Confusing predictions when implementing GoogleNet

I'm user the Caffe implementation of GoogleNet available from https://github.com/BVLC/caffe/tree/master/models/bvlc_googlenet in a detection and localization process. I've had great success using this network for transfer learning on my own…
aaron
  • 6,339
  • 12
  • 54
  • 80
0
votes
1 answer

VGG: order of categories?

I'm using VGG-16 network trained on ILSVRC with Fast-RCNN. You can get a 1000-dimensional matrix for probability of each of 1,000 classes in ILSVRC. However, I can't find out which column in the matrix corresponds to which class. For example, I'd…
ytrewq
  • 3,670
  • 9
  • 42
  • 71
0
votes
1 answer

ImageNet ILSVRC2012 truncated tar archive

I have downloaded the image data from ImageNet ILSVRC2012 and save into a hard disk on this site : http://www.image-net.org/challenges/LSVRC/2012/nonpub-downloads . I am working on OSX. When I do tar -xvf ILSVRC2012_img_train.tar, I have this error…
A. Attia
  • 1,630
  • 3
  • 20
  • 29
0
votes
0 answers

Caffe: Why in Imagenet model activation dimension of fc6 is 4096?

I was looking at pool5 layer in the Imagenet trained model (ILSVRC challenge) whose output size is 256*6*6 (~9000), after which there is a fc6 layer whose num_outputs is 4096. Can anyone please explain how is 4096 chosen?
Deven
  • 617
  • 2
  • 6
  • 20
-1
votes
0 answers

Do we need to resize to 224 when using pretrained Resnet for Tiny imageNet?

I'd like to fine-tune a resnet50 net = resnet50(weights=ResNet50_Weights.DEFAULT) models on Tinyimage Net data. As the resnet50 in pytorch has been trained on 3x256x256 images and Tinyimagetnet is sized at 3x64x64 I wonder if I need to have,…
linkho
  • 99
  • 2
1 2 3
…
13
14