tfdeploy
tfdeploy copied to clipboard
riga
Deploy tensorflow graphs for fast evaluation and export to tensorflow-less environments running numpy.
Hello,I found a performance issue in the definition of `create_models` , tests/perf/measure_runtimes.py, [tf_sess = tf.Session()](https://github.com/riga/tfdeploy/blob/22aea652fe12f081be43414e0f1f76c7d9aaf53c/tests/perf/measure_runtimes.py#L98) was repeatedly called and was not closed. I think it will increase the efficiency and...
Hello, I found a performance issue in the definition of `create_tf_model`, tests/perf/measure_runtimes.py, [`b = tf.Variable(tf.zeros([units]))`](https://github.com/riga/tfdeploy/blob/22aea652fe12f081be43414e0f1f76c7d9aaf53c/tests/perf/measure_runtimes.py#L85) will be created repeatedly during program execution, resulting in reduced efficiency. I think it should...
Currently, there's no support for tf.strided_slice() so if somewhere in your graph you have x[:,:,:,None] the graph won't execute. It's a complicated function, so it won't be easy to implement.
It seems that after saving a model, all subsequent saves are identical. I made a minimal example that reproduces the problem: ``` import tensorflow as tf import tfdeploy.tfdeploy as td...
Hi, I'm trying to restore a model saved with tfdeploy but during evaluation I have the following errors: ``` Traceback (most recent call last): File "ensemble_learning.py", line 71, in result...
I added tests for simple CNN-based Keras models (https://github.com/riga/tfdeploy/issues/18) as well as the pretrained models which then cover the popular image classification models (ResNet, VGG, Inception), currently all of the...
I am trying to deploy a combination of tf.contrib.slim nets including multiple inception_resnet_v1.conv2d and fully_connected that contain an op called Merge. The operation from the model looks like so `InceptionResnetV1/Conv2d_1a_3x3/BatchNorm/cond/Merge`...
Hello! I wonder, how does tfdeploy deal with tf queues? I don't use placeholder in my model but a graph like file_name_queue -> data_reader -> dnn. I get these error...
Hey. I'm trying to convert a keras model into a tfdeploy model. Below is the code ``` import tensorflow as tf import tfdeploy as td from keras.models import load_model td.setup(tf)...
