TF-Slim is a lightweight library for defining, training and evaluating complex models in TensorFlow.
Questions tagged [tf-slim]
117 questions
0
votes
1 answer
TF Slim and inception prediction probabilities
I've trained the final layers of inception v4 using tf slim (v1.3) and I'm trying to integrate processing into an existing tool but I'm having trouble figuring out how to generate predictions. I want all the prediction values, not just the argmax.…
Trevor
- 1,369
- 2
- 13
- 28
0
votes
0 answers
distributed tensorflow clarification
Is my understanding correct that model_deploy lets the user train a model using multiple devices on a single machine? The basic premise seems that the clone devices do variable sharing and variables get distributed to param servers in a round-robin…
deniz
- 23
- 5
0
votes
1 answer
Understanding tf.metrics and slims streaming metric
I'm not sure if I understand tf.metrics and tf.contrib.slim.metrics correctly.
Here is the general flow of the program:
# Setup of the neural network...
# Adding some metrics
dict_metrics[name] = compute_metric_and_update_op()
# Getting a list of…
user667804
- 740
- 6
- 25
0
votes
1 answer
Tf-slim: ValueError: Variable vgg_19/conv1/conv1_1/weights already exists, disallowed. Did you mean to set reuse=True in VarScope?
I am using tf-slim to extract features from several batches of images. The problem is my code works for the first batch , after that I get the error in the title.My code is something like this:
for i in range(0, num_batches):
#Obtain the…
HIMANSHU RAI
- 305
- 1
- 4
- 13
0
votes
0 answers
"TypeError: Cannot create initializer for non-floating point type." when using extract features in tf-slim with VGG19 model
I want to extract features from the pool5 layer from vgg19. I am using tf-slim library for that . I get a very weird error saying
TypeError: Cannot create initializer for non-floating point type.
My images are ndarrays of dtype=uint8 . So in order…
HIMANSHU RAI
- 305
- 1
- 4
- 13
0
votes
1 answer
Can i fetch a tensor when i am using slim.learning.train?
Thanks for TF-Slim, it is easier to build new models
But when comes about training the model, it seems that slim.learning.train can't fetch a tensor which i need to check my model.
It can only return the loss (the training op), the code below is in…
ycd
- 36
- 1
- 7
0
votes
1 answer
does tf slim fine tuning require gpu?
Actually I am trying to fine tune inceptionV3 model using tf slim fine tuning example on git hub it is giving me this error :
InvalidArgumentError (see above for traceback): Cannot assign a device to node…
Anuj Rohilla
- 31
- 2
- 6
0
votes
1 answer
Tensorflow CNN Shape error
I'm trying to use a CNN on some data but I'm getting an error in my loss function since the output of my model is the shape [1000, 1000, 4000] when it should be [1000, 4000]. The first 1000 in this case is the batch size, and the 4000 is the number…
dooder
- 529
- 1
- 6
- 14
0
votes
1 answer
How to read the top of a Queue multiple times before dequeueing in tensorflow
In the following example, every time I run sess.run([image, label]), a different sample from the queue is returned, thus a different np_image is returned.
Is there a way that I can let the slim.queues.QueueRunners know that I want to use (run) the…
Alex Allen
- 39
- 7
0
votes
1 answer
TF-slim : FIFOQueue is closed and has insufficient elements (requested 1, current size 0)
I trying to use slim to make my own dataset and to read into it.
When I tried to read into it I get the following error:
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.OutOfRangeError: FIFOQueue
…
BashIntosh
- 101
- 8
0
votes
1 answer
Not sure why my build_image_data.py script for tensforflow isn't working
I'm trying to convert a set of jpg images to TFRecord for use with a CNN in Tensorflow.I'm using the script build_image_data.py from here:
https://github.com/tensorflow/models/blob/master/inception/inception/data/build_image_data.py
I've created a…
Jaafa
- 141
- 2
- 12
0
votes
0 answers
TF-Slim: computation of accuracy
I am using TensorFlow-Slim and I've added some code lines in eval_image_classifier.py (located in /models/slim/) for computing TP, TN, FP and FN. However, computing Accuracy = (TP + TN) / (TP + FP + FN + TN) is not equal to the accuracy given by…
Elternhaus
- 45
- 2
- 9
0
votes
1 answer
Multi-GPU training using tf.slim takes more time than single GPU
I'm fine-tuning ResNet50 on the CIFAR10 dataset using tf.slim's train_image_classifier.py script:
python train_image_classifier.py \
--train_dir=${TRAIN_DIR}/all \
…
Anas
- 866
- 1
- 13
- 23
0
votes
1 answer
Is it possible to show classification error together with continuous error during slim train?
i want to watch not only continuous error, which should be optimized during training, but also another not differentiable metrics(like top1 or top5 classification error) during training. Is it possible?
Example:
outputs =…
svetlov.vsevolod
- 191
- 8
0
votes
1 answer
How to train inception v3 model for multilabel classification?
I'm trying to build a model to detect multiple attributes in an image.
I'm using the pre-trained inception V3 model. I understand that we have to change the final softmax layer to sigmoid.
I'm loading the model as follows:
with…
gibbidi
- 183
- 1
- 8