Stack Overflow
Stack Overflow

Questions tagged [tensorflow-slim]

TensorFlow-Slim is a lightweight library for defining, training and evaluating complex models in TensorFlow. Components of tf-slim can be freely mixed with native tensorflow, as well as other frameworks, such as tf.contrib.learn.

TensorFlow-Slim is a lightweight library for defining, training and evaluating complex models in TensorFlow. Components of tf-slim can be freely mixed with native tensorflow, as well as other frameworks, such as tf.contrib.learn.

Usage: import tensorflow.contrib.slim as slim

Github: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/slim

87 questions
14
votes
1 answer

Interpreting tensorboard plots

I'm still newbie in tensorflow and I'm trying to understand what's happenning in details while my models' training goes on. Briefly, I'm using the slim models pretrained on ImageNet to do the finetuning on my dataset. Here are some plots extracted…
Maystro
  • 2,907
  • 8
  • 36
  • 71
8
votes
3 answers

tf object detection api - extract feature vector for each detection bbox

I'm using Tensorflow object detection API and working on pretrainedd ssd-mobilenet model. is there a way to extact the last global pooling of the mobilenet for each bbox as a feature vector? I can't find the name of the operation holding this…
dotannn
  • 131
  • 1
  • 5
7
votes
1 answer

How to augment data in tensorflow tfrecords?

I am storing my data using tfrecords and I read them as tensors using Dataset API and then I use the Estimator API to perform training. Now, I want to do online data-augmentation on each item in the dataset, but after trying for a while I cannot…
deep_jandu
  • 183
  • 3
  • 8
4
votes
0 answers

Saving tensorflow-slim model using steps intervals instead of time interval (seconds)

I am using object detection in tensorflow api. In my previous work I used to check the current step and save my model every n steps, something like the approach mentioned here. In this case though the authors use TensorFlow-Slim to perform the…
Eypros
  • 5,370
  • 6
  • 42
  • 75
4
votes
2 answers

What does tf.contrib.slim.prefetch_queue do in tensorflow

I saw an example of training cifar10 data using tensorflow: https://github.com/tensorflow/models/tree/master/tutorials/image/cifar10 the code generate a batch of images from several single image using tf.train.batch and create a queue of batchs…
Ao Li
  • 101
  • 5
3
votes
1 answer

Tensorflow Hub Image Modules: Clarity on Preprocessing and Output values

Many thanks for support! I currently use TF Slim - and TF Hub seems like a very useful addition for transfer learning. However the following things are not clear from the documentation: 1. Is preprocessing done implicitly? Is this based on…
Maverick
  • 45
  • 3
3
votes
1 answer

Re-train pre-trained ResNet-50 model with tf slim for classification purposes

I would like to re-train a pre-trained ResNet-50 model with TensorFlow slim, and use it later for classifying purposes. The ResNet-50 is designed to 1000 classes, but I would like just 10 classes (land cover types) as output. First, I try to code…
poetyi
  • 236
  • 4
  • 13
3
votes
1 answer

RGB or BGR for Tensorflow-slim-ResNet V2 pre-trained model?

For CNN training, the exact order of input image channels can be different from library to library, even model to model. For Caffe, the input image is usually expected to be in BGR order, while in Tensorflow, the order can be arbitrary. So does…
user4225701
3
votes
2 answers

Can't train my Tensorflow detector model in Google Cloud

I'm trying to train my own Detector model based on Tensorflow sample and this post. And I did succeed on training locally on a Macbook Pro. The problem is that I don't have a GPU and doing it on the CPU is too slow (about 25s per iteration). This…
Minoru
  • 1,680
  • 3
  • 20
  • 44
2
votes
1 answer

OpenVino Model Optimizer Error when converting TensorFlow model

I have created a custom image classification .pb model file using the python scripts in the TensorFlow for Poets 2 repo (https://github.com/googlecodelabs/tensorflow-for-poets-2). I tried converting it to Intermediate Representation using the…
Gino
  • 21
  • 4
2
votes
0 answers

tf.keras equivalent code block written in tf.contrib.slim

I'm trying to re-implement a research paper code in tf.keras, in init block it was written as: with slim.arg_scope([slim.conv2d,separable_conv],activation_fn=tf.nn.relu6, normalizer_fn=slim.batch_norm): with slim.arg_scope([slim.batch_norm],…
PRNV JB
  • 51
  • 6
2
votes
2 answers

Tensorflow object detection API: Custom VGG 16 model

I am in the process of creating a Custom VGG model as a feature extractor of Faster RCNN model in Tensorflow object detection API. As mentioned on in the document…
ReInvent_IO
  • 477
  • 3
  • 13
2
votes
1 answer

Unable to understand tf.nn.raw_rnn

In the official documentation of tf.nn.raw_rnn we have emit structure as the third output of loop_fn when the loop_fn is run for the first time. Later on the emit_structure is used to copy tf.zeros_like(emit_structure) to the minibatch entries that…
figs_and_nuts
  • 4,870
  • 2
  • 31
  • 56
2
votes
1 answer

A generic way to retrain the last/top layers in a pre-trained TensorFlow graph

I am experimenting with different pre-trained TensorFlow models from "Tensorflow detection model zoo" on my data - I want to retrain (fine-tune) only the deepest N layers in each model to find the optimal configuration for my data. Is there a way to…
Jenny
  • 333
  • 2
  • 11
2
votes
0 answers

How to load parameters from two tensorflow models and output one model file?

Let's say I have two tensorflow models(A and B) with exactly the same graph but different weights in each layer. I want to merge layers conv1, conv2, conv3 from A and layers conv4, conv5, conv6 from B into one model C. Model C has the same structure…
Eric Li
  • 21
  • 1
1
2 3 4 5 6