legolas123 / cv-tricks.com Goto Github PK

In the folder, Tensorflow-tutorials/Tensorflow-slim-run-prediction when I run python run_inference_on_v3.py http://cv-tricks.com/wp-content/uploads/2017/03/pexels-photo-280207.jpeg
To run the prediction for the model inception_v3 as per the blog post http://cv-tricks.com/tensorflow-tutorial/understanding-alexnet-resnet-squeezenetand-running-on-tensorflow/, I get the following error:

TypeError: Expected int32, got list containing Tensors of type '_Message' instead.
Tensorflow version: 1.1

I'm running 2.2.4..

  File "./1_vgg16_pretrain.py", line 1, in <module>
    from network import VGG16
ModuleNotFoundError: No module named 'network'

my issue is in imagnet.py where my tring to access url am getting the url error as follows
filename, _ = urllib.urlretrieve(synset_url,"imagenet_lsvrc_2015_synsets.txt")
File "/usr/lib/python2.7/urllib.py", line 94, in urlretrieve
return _urlopener.retrieve(url, filename, reporthook, data)
File "/usr/lib/python2.7/urllib.py", line 240, in retrieve
fp = self.open(url, data)
and my code is
base_url = 'https://github.com/tensorflow/models/tree/master/research/inception/inception/data'
synset_url = 'https://github.com/tensorflow/models/tree/master/research/inception/inception/data/imagenet_lsvrc_2015_synsets.txt'.format(base_url)
synset_to_human_url = 'https://github.com/tensorflow/models/tree/master/research/inception/inception/data/imagenet_metadata.txt'.format(base_url)

filename, _ = urllib.urlretrieve(synset_url,"imagenet_lsvrc_2015_synsets.txt")
synset_list = [s.strip() for s in open(filename).readlines()]
num_synsets_in_ilsvrc = len(synset_list)
assert num_synsets_in_ilsvrc == 1000

filename, _ = urllib.urlretrieve(synset_to_human_url,"imagenet_metadata.txt")
synset_to_human_list = open(filename).readlines()
num_synsets_in_all_imagenet = len(synset_to_human_list)
assert num_synsets_in_all_imagenet == 21842
please help me am totally frustraed bu this issue

Hello,
Thank you for sharing this excellent work. I am trying to port the Crop Layer to cpp.
I am not very familiar with Python, can you elaborate on the meaning of this Python operation :

return [inputs[0][:,:,self.ystart:self.yend,self.xstart:self.xend]]

Is inputs[0] a 4D cube and I want to take only the sub-cube delimited by x and y indexes?
Cheers,
Renaud

Hi,
Thanks for putting this up on github. It got me started with tensorflow.
However some comments;
I think the dataset shows a lot of underfitting. I get a validation score of 0.5 at end, or even below.
I connected tensorboard and the learning graph shows classic underfitting signs (straight learning rate).
Perhaps it doesn't help that the images have wide ranging aspect ratios?
I tried to insert horizontal flipped images as well, but it doesn't help that much.

Getting an error while running the code.

python3 run_pose.py --input sample.jpg --proto pose/mpi/pose_deploy_linevec_faster_4_stages.prototxt --model pose/mpi/pose_iter_160000.caffemodel --dataset MPI
time is 2.5539000034332275
: cannot connect to X server

Hello,

I am trying to add data augmentation to increase the number of images I have to at least 2,000 images per class. How can I achieve this?
Any help will be really appreciated

Hi!

I get the following error when I want to run edge detection on an image. Any idea why I get that specific dnn layer error?

cv2.error: OpenCV(4.6.0) /io/opencv/modules/dnn/src/layers/convolution_layer.cpp:377: error: (-215:Assertion failed) !blobs.empty() || inputs.size() > 1 in function 'getMemoryShapes'

Hi,

I am getting the following error when using code from your repository for Image classifier CNN:

Traceback (most recent call last):
  File "E:/Machine Learning/TensorFlow/Tutorials & Practice/CV Tricks/train.py", line 292, in <module>
    optimize(num_iterations=3000)
  File "E:/Machine Learning/TensorFlow/Tutorials & Practice/CV Tricks/train.py", line 261, in optimize
    x_batch, y_true_batch, _, cls_batch = data.train.next_batch(train_batch_size)
  File "E:\Machine Learning\TensorFlow\Tutorials & Practice\CV Tricks\dataset.py", line 117, in next_batch
    assert batch_size <= self._num_examples
AssertionError

I am not sure what am I missing, Can you help me please?

OS: Windows 10
IDE: Pycharm
Python: Python 3.5.2 :: Anaconda 4.1.1 (64-bit)

Quantization code is never called since it is commented. Binary convolution and activation is already implemented. Do I need to uncomment them (a few of them are given below)? During the forward pass I always quantize the weights and activation. Why do I need "quantization.py" ?

# if args.quantize:
    #     global bin_op
    #     bin_op = quantization.Binarize(model)

# if args.quantize:
        #     bin_op.restore()
        #     bin_op.updateBinaryGradWeight()

if I uncomment them, I get the following error.

  File "main2.py", line 88, in main
    bin_op = quantization.Binarize(model)
  File "/media/mmrg/inci/files/quantization.py", line 33, in __init__
    param_grp_1.append(m.conv.weight)
  File "/home/mmrg/anaconda3/envs/pytorch_mmdet/lib/python3.7/site-packages/torch/nn/modules/module.py", line 779, in __getattr__
    type(self).__name__, name))
torch.nn.modules.module.ModuleAttributeError: 'BinConv2d' object has no attribute 'conv'

I run the code of image-classifier, it's a little slow in my laptop.
My computer:
Intel Core i7 6700-HQ
NVIDIA GeForce GTX960M
run in GPU
Then I try move the save code into the if rather than out of if
It become much faster.....
And I think it is useless to do so much save and waste time in IO
Maybe this would be helpful if you want to accelerate your code

~/git/cv-tricks.com/Tensorflow-tutorials/tutorial-2-image-classifier$ python train.py --train training_data/ --val testing_data/ --num_classes 2
Going to read training images
Now going to read dogs files (Index: 0)
Now going to read cats files (Index: 1)
Complete reading input data. Will Now print a snippet of it
Number of files in Training-set: 800
Number of files in Validation-set: 200
WARNING:tensorflow:From train.py:44: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the axis argument instead
WARNING:tensorflow:From train.py:160: softmax_cross_entropy_with_logits (from tensorflow.python.ops.nn_ops) is deprecated and will be removed in a future version.
Instructions for updating:

Future major versions of TensorFlow will allow gradients to flow
into the labels input on backprop by default.

See @{tf.nn.softmax_cross_entropy_with_logits_v2}.

Training Epoch 1 --- Training Accuracy: 50.0%, Validation Accuracy: 43.8%, Validation Loss: 0.705
2018-08-17 16:52:14.024307: W tensorflow/core/framework/op_kernel.cc:1275] OP_REQUIRES failed at save_restore_v2_ops.cc:109 : Not found: ; No such file or directory
Traceback (most recent call last):
File "train.py", line 206, in
train(num_iteration=3000)
File "train.py", line 201, in train
saver.save(session, 'dogs-cats-model')
File "/home/Wukong2/l00176142/.local/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1637, in save
raise exc
ValueError: Parent directory of dogs-cats-model doesn't exist, can't save.

Hi Ankit,

I am new to image classification. I cloned your repository and tried to run the code train.py in my system. I just keep getting an error message saying:
"2018-06-17 00:19:55.817737: W tensorflow/core/framework/op_kernel.cc:1318] OP_REQUIRES failed at save_restore_v2_ops.cc:109 : Not found: ; No such file or directory
Traceback (most recent call last):
File "train.py", line 206, in
train(num_iteration=3000)
File "train.py", line 201, in train
saver.save(session, 'dogs-cats-model')
File "/Users/z001txn/anaconda/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1720, in save
raise exc
ValueError: Parent directory of dogs-cats-model doesn't exist, can't save."

I think the problem is in saving the model but I am not able to figure out how to solve it. Can you please help me out here.

Thanks,
Aishwarya

Hi!
In predict.py I am always getting the same output even though the image are of cats(for example ), the probability is very less for that image.

Could you add a way to save the model?
Also, could you add a file to load the saved model and infer the prediction on a test image?

Laurent

I am not able to get the image_path correctly.
The output is showing ---------------------list index out of range by running your code predict.py
dir_path = os.path.dirname(os.path.abspath(__file__)) image_path=sys.argv[1]

due to which image = cv2.resize(image, (image_size, image_size),0,0, cv2.INTER_LINEAR) is also not working.

output of dir_path D:\WORKSPACE\Deep_learning\others\cv_tricks.com_new\cv-tricks.com\Tensorflow-tutorials\tutorial-2-image-classifier
output of image_path D:/WORKSPACE/Deep_learning/others/cv_tricks.com_new/cv-tricks.com/Tensorflow-tutorials/tutorial-2-image-classifier/predict.py

how can this be solved?

i have built a cnn network which im trying to train with my own data. but when i run the code i get the next error:

File "C:/linear regression/new_net.py", line 92, in <module>

ValueError: Cannot feed value of shape (500, 99999) for Tensor 'X:0', which has shape '(500, 4392)'

and i dont know what value i need to fix. my data is built of 100,000 rows on 4392 column. i reshape each row to 61*72 pixels which means i have 100,000 images. i take 80,000 to the training and 20,000 to the test. and using batch size of 500. i dont know what is wrong with the code if someone can help please!!!

here is my code:

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import csv
import time
import os
import urllib

import numpy as np
import tensorflow as tf
#input images- chek if i need the transpose command
data_in = np.transpose(np.loadtxt(open("images.csv"), delimiter=","))

images_training="images.csv"

train_X=data_in[0:80000,:]
test_X=data_in[80001:100000,:]

test_images=test_X[:,1]
test_labels=test_X[:,0]

n = train_X.shape[1]
l = test_X.shape[0]

n_epochs = 10
batch_size = 500
regions=50
height = 61
width = 72
channels = 1

n_inputs = height * width #4392

X = tf.placeholder(tf.float32, shape=[batch_size, n_inputs], name="X")
X_reshaped = tf.reshape(X, shape=[-1, height, width, channels])
y = tf.placeholder(tf.int32, shape=[None], name="y")
labels=tf.placeholder(tf.int32, shape=[batch_size], name="labels")

conv1 = tf.layers.conv2d(inputs=X_reshaped,filters=32,kernel_size=[5, 
5],padding="same",activation=tf.nn.relu)

pool1 = tf.layers.max_pooling2d(inputs=conv1, pool_size=[2, 2], strides=2) 
#need to change pool size and strides according to image pixels

conv2 = tf.layers.conv2d(inputs=pool1,filters=64,kernel_size=[5, 
5],padding="same",activation=tf.nn.relu)

pool2 = tf.layers.max_pooling2d(inputs=conv2, pool_size=[2, 2], 
strides=2)#need to change pool size and strides according to image pixels

pool2_flat = tf.reshape(pool2, [-1, 15 * 18 * 64]) #change it according to 
the pooling layers size:[batch_size,18432]

dense = tf.layers.dense(inputs=pool2_flat, units=1024, 
activation=tf.nn.relu)

# Add dropout operation; 0.6 probability that element will be kept
dropout = tf.layers.dropout(inputs=dense, rate=0.4)

# Logits layer
# Input Tensor Shape: [batch_size, 1024]
# Output Tensor Shape: [batch_size, 50]
logits = tf.layers.dense(inputs=dropout, units=50) #size=[batch_size,50]

predictions = {
  # Generate predictions (for PREDICT and EVAL mode)
  "classes": tf.argmax(input=logits, axis=1),
  # Add `sigmoid_tensor` to the graph. It is used for PREDICT and by the
  # `logging_hook`.
  "probabilities": tf.nn.sigmoid(logits, name="sigmoid_tensor")}


 onehot_labels = tf.one_hot(indices=tf.cast(labels, tf.int32), depth=50)
 #loss = tf.losses.sigmoid_cross_entropy(multi_class_labels=labels, 
 logits=logits)
 loss = tf.losses.sigmoid_cross_entropy(onehot_labels, logits=logits) 


optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001)
train_op = optimizer.minimize(loss=loss, 
global_step=tf.train.get_global_step())

# Initializing the variables
init = tf.global_variables_initializer()

with tf.Session() as sess:
     sess.run(init)
     for epoch in range(n_epochs):
     for iteration in range(len(train_X) // batch_size):
          # this cycle is for dividing step by step the heavy work of each neuron
          X_batch = train_X[iteration * batch_size:iteration * batch_size + batch_size, 1:]
          y_batch = train_X[iteration * batch_size:iteration * batch_size + batch_size, 0]
          sess.run(optimizer,feed_dict={X:X_batch, y:y_batch})
     sess.run(train_op, feed_dict={X: X_batch, y: y_batch})
     acc_train =accuracy.eval(feed_dict={X: X_batch, y: y_batch})
     acc_test  =accuracy.eval(feed_dict={X: test_images, y: test_labels})
     print("Epoch:", epoch + 1, "Train accuracy:", acc_train, "test accuracy:", acc_test)
     print("Optimization Finished!")

in addition i dont understand why python takes my data as 99,999 and not 100,000.

thank you!!!

is there a edge detection model which is not in caffe ?
as caffe merged into pytorch, it's almost not used anymore. instaling caffe "just for that" is too bad.

Where could I get the code for passing an image to a CNN Cifar10 TensorFlow Model and getting the result in terms of prediction scores
I had Build the CNN Cifar10 TensorFlow Model following the tutorial at
https://www.tensorflow.org/tutorials/deep_cnn
Code:
https://github.com/tensorflow/models/tree/master/tutorials/image/cifar10

WARNING:tensorflow:From train.py:43: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the axis argument instead
WARNING:tensorflow:From train.py:156: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the axis argument instead
Going to read training images
Now going to read dogs files (Index: 0)
Now going to read cats files (Index: 1)
Complete reading input data. Will Now print a snippet of it
Number of files in Training-set: 800
Number of files in Validation-set: 200
Training Epoch 1 --- Training Accuracy: 50.0%, Validation Accuracy: 43.8%, Validation Loss: 0.703
Traceback (most recent call last):
File "C:\Users\USER\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1327, in _do_call
return fn(*args)
File "C:\Users\USER\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1306, in _run_fn
status, run_metadata)
File "C:\Users\USER\AppData\Local\Programs\Python\Python36\lib\contextlib.py", line 89, in exit
next(self.gen)
File "C:\Users\USER\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\errors_impl.py", line 466, in raise_exception_on_not_ok_status
pywrap_tensorflow.TF_GetCode(status))
tensorflow.python.framework.errors_impl.NotFoundError: Failed to create a directory:
[[Node: save/SaveV2 = SaveV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_arg_save/Const_0_0, save/SaveV2/tensor_names, save/SaveV2/shape_and_slices, Variable, Variable/Adam, Variable/Adam_1, Variable_1, Variable_1/Adam, Variable_1/Adam_1, Variable_2, Variable_2/Adam, Variable_2/Adam_1, Variable_3, Variable_3/Adam, Variable_3/Adam_1, Variable_4, Variable_4/Adam, Variable_4/Adam_1, Variable_5, Variable_5/Adam, Variable_5/Adam_1, Variable_6, Variable_6/Adam, Variable_6/Adam_1, Variable_7, Variable_7/Adam, Variable_7/Adam_1, Variable_8, Variable_8/Adam, Variable_8/Adam_1, Variable_9, Variable_9/Adam, Variable_9/Adam_1, beta1_power, beta2_power)]]

I am getting the below outputs for Cat and Dog images respectively. how do i interpret it?

run : python predict.py testing_data/cats/cat.1041.jpg
output: [[3.3456912e-05 9.9996650e-01]]

run: python predict.py testing_data/dogs/dog.1000.jpg
output: [[5.817143e-05 9.999418e-01]]

Hi,
I am getting error "frozen importlib._bootstrap>(219)_call_with_frames_removed()" after line 93.
93 line is class DataSets(object):
I am new in Deep Learning as well as in Tensorflow.
Please help to resolve this issue

What is the meaning of "Using real valued downsample"?

I did the training, it had 95% confidence.

Somehow my prediction is always the same, even when giving a different image to test (changing cats to dogs):

[[0.5592177  0.44078225]]
[[0.5690457  0.43095428]]

Anyway to debug or help out what is going wrong?