i have a directory of png files . there is a train folder and test folder . In the train folder i have 10 folders as 10 labels [ 0 -9 ] .Each folder contains png files of that label . I want to load them in tensor flow for training . I am new in tensor flow i am having a very hard time getting this done
i am using anaconda ( py ver 3.5 )
import tensorflow as tf
filename_queue = tf.train.string_input_producer(
tf.train.match_filenames_once("./images/*.jpg"))
image_reader = tf.WholeFileReader()
i have tried using this but can make it work . it only loads 1 image
It's working for me though. Can you run this script ? (Updated to get Labels as well)
import tensorflow as tf
filename_queue = tf.train.string_input_producer(
tf.train.match_filenames_once("/home/xxx/Desktop/stackoverflow/images/*/*.png"))
image_reader = tf.WholeFileReader()
key, image_file = image_reader.read(filename_queue)
S = tf.string_split([key],'/')
length = tf.cast(S.dense_shape[1],tf.int32)
# adjust constant value corresponding to your paths if you face issues. It should work for above format.
label = S.values[length-tf.constant(2,dtype=tf.int32)]
label = tf.string_to_number(label,out_type=tf.int32)
image = tf.image.decode_png(image_file)
# Start a new session to show example output.
with tf.Session() as sess:
# Required to get the filename matching to run.
tf.initialize_all_variables().run()
# Coordinate the loading of image files.
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(coord=coord)
for i in xrange(6):
# Get an image tensor and print its value.
key_val,label_val,image_tensor = sess.run([key,label,image])
print(image_tensor.shape)
print(key_val)
print(label_val)
# Finish off the filename queue coordinator.
coord.request_stop()
coord.join(threads)
File Directory
./images/1/1.png
./images/1/2.png
./images/3/1.png
./images/3/2.png
./images/2/1.png
./images/2/2.png
Output:
(881, 2079, 3)
/home/xxxx/Desktop/stackoverflow/images/3/1.png
3
(155, 2552, 3)
/home/xxxx/Desktop/stackoverflow/images/2/1.png
2
(562, 1978, 3)
/home/xxxx/Desktop/stackoverflow/images/3/2.png
3
(291, 2558, 3)
/home/xxxx/Desktop/stackoverflow/images/1/1.png
1
(157, 2554, 3)
/home/xxxx/Desktop/stackoverflow/images/1/2.png
1
(866, 936, 3)
/home/xxxx/Desktop/stackoverflow/images/2/2.png
2
Related
I am trying to run a facial recognition service on my Raspberry Pi, but it has suddenly stopped detecting faces
I am using python-opencv for this and the last time I tested it everything worked fine. The below code is for the the training the system on new faces
`
import cv2
import os
import numpy as np
from PIL import Image
import sqlite3
recognizer = cv2.createLBPHFaceRecognizer()
detector= cv2.CascadeClassifier("haarcascade_frontalface_default.xml");
cam = cv2.VideoCapture(0)
con = sqlite3.connect('Users.db')
cur = con.cursor()
Id=raw_input('Enter your id: ')
name=raw_input('Enter your name: ')
sampleNum=0
while True:
ret, img = cam.read()
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = detector.detectMultiScale(gray, 1.3, 5)
for (x,y,w,h) in faces:
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
#incrementing sample number
sampleNum=sampleNum+1
#saving the captured face in the dataset folder
cv2.imwrite("dataSet/User."+Id +'.'+ str(sampleNum) + ".jpg", gray[y:y+h,x:x+w])
cv2.imshow('frame',img)
#wait for 100 miliseconds
if cv2.waitKey(100) & 0xFF == ord('q'):
break
# break if the sample number is more than 20
elif sampleNum>30:
break
cam.release()
cv2.destroyAllWindows()
def getImagesAndLabels(path):
#get the path of all the files in the folder
imagePaths=[os.path.join(path,f) for f in os.listdir(path)]
#create empth face list
faceSamples=[]
#create empty ID list
Ids=[]
#now looping through all the image paths and loading the Ids and the images
for imagePath in imagePaths:
#loading the image and converting it to gray scale
pilImage=Image.open(imagePath).convert('L')
#Now we are converting the PIL image into numpy array
imageNp=np.array(pilImage,'uint8')
#getting the Id from the image
Id=int(os.path.split(imagePath)[-1].split(".")[1])
# extract the face from the training image sample
faces=detector.detectMultiScale(imageNp)
#If a face is there then append that in the list as well as Id of it
for (x,y,w,h) in faces:
faceSamples.append(imageNp[y:y+h,x:x+w])
Ids.append(Id)
return faceSamples,Ids
os.system("sudo rm trainer/trainer.yml")
faces,Ids = getImagesAndLabels('dataSet')
recognizer.train(faces, np.array(Ids))
recognizer.save('trainer/trainer.yml')
`
Normally it would open a window and show 30 photos in succession of the user's face, but now nothing shows up after it asks for a name. I have run other OpenCV applications and it can find faces in static images.
I had to replace the haar_cascade_frontalface.xml file
I started writing a python code for creating HDF5 for image segmentation tasks. I used the code in this link and the link provided by Shai. my images are one channel and in .mat format. I have written the following code, I only want to check with experts whether this code is correct or not. Could experts please have a look? Thanks
import os, h5py
import caffe
import numpy as np
import scipy
import scipy.io as sio
from array import array
import cv2
import matplotlib.pyplot as plt
caffe_root='/home/ss/caffe/'
import sys
sys.path.insert(0,caffe_root+'python')
def img_to_hdf5(paths_src_file,paths_lbl_file,path_dst,msg):
"""
paths_src_file : path to the image paths in a txt file
paths_lbl_file : path to the image paths in a txt file
path_dst = path to the hdf5 file
"""
print(msg)
arrays = {}
SIZE=256 #fixed size of all images
#read the lines of img and lbl path from text file and save into paths_src and paths_lbl
paths_src = []
with open(paths_src_file) as f:
for line in f.readlines():
line = line.strip('\n')
paths_src.append(line)
paths_lbl=[]
with open(paths_lbl_file) as f:
for line in f.readlines():
line=line.strip('\n')
paths_lbl.append(line)
data = np.zeros( (len(paths_src), 1, SIZE, SIZE), dtype='f4' ) #1 channel grayscale image
label = np.zeros( (len(paths_lbl), 1, SIZE, SIZE), dtype='f4' ) #1 channel label image
for in_idx, in_ in enumerate(paths_src):
print in_idx,in_
f=h5py.File(in_,'r')
mat=f['image'].value
im=np.array(mat,dtype=np.float32)
#im = cv2.cvtColor(im,cv2.COLOR_GRAY2RGB)
#im = im[:,:,::-1] #switch from RGB to BGR
im = im.reshape(im.shape[0], im.shape[1], 1)
im = im.transpose((2,0,1)) # convert to CxHxW
data[in_idx]=im
for in_idx, in_ in enumerate(paths_lbl):
print in_idx,in_
f=h5py.File(in_,'r')
mat=f['image'].value
im=np.array(mat,dtype=np.float32)
#im = cv2.cvtColor(im,cv2.COLOR_GRAY2RGB)
#im = im[:,:,::-1] #switch from RGB to BGR
im = im.reshape(im.shape[0], im.shape[1], 1)
im = im.transpose((2,0,1)) # convert to CxHxW
label[in_idx]=im
h5_train = os.path.join(path_dst, 'train_data.h5')
with h5py.File(h5_train,'w') as H:
H.create_dataset( 'data', data=data ) # note the name X given to the dataset!
H.create_dataset( 'label', data=label ) # note the name y given to the dataset!
text_train = os.path.join(path_dst, 'train-path.txt')
with open(text_train,'w') as L:
L.write(h5_train) # list all h5 files you are going to use
train_img_paths = './train_img.txt' #text file of paths to images
train_label_paths = './train_label.txt' #text file of paths to label images (ground truth)
train_img_hdf5 = '/home/ss/workspace/create_hdf5/' # Your path to h5 file
st='Creating Training Data HDF5 File .....'
img_to_hdf5(train_img_paths, train_label_paths,train_img_hdf5,st)
print('DONE...')
This is what I have so far: the output of the script doesn't even print the difference values yet.
import os
import sys
from skimage.measure import compare_ssim
import cv2
im1 = cv2.imread("7.jpeg");
dir= '/Users/Desktop/images'
//I'm trying to search the specified directory and compare each image
in the "images" folder to im1. Then compute the difference between the
two images for each image in the folder. Lastly I want the program to
output the name of the image that has the smallest difference with im1.
def get_nb_files(dir):
for r in dir:
grayA = cv2.cvtColor(im1, cv2.COLOR_BGR2GRAY)
grayB = cv2.cvtColor(r, cv2.COLOR_BGR2GRAY)
diff = (diff * 255).astype("uint8")
(score, diff) = compare_ssim(im1, r, full=True)
print("SSIM: {}".format(score))
My experience with python is very limited so I don't fully understand what the code does in this instance. This is part of the code for poets lab from the tensorflow framework.
import os, sys
import tensorflow as tf
import sys
import numpy as np
from PIL import Image
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
# change this as you see fit
image_path = sys.argv[1]
# Read in the image_data
image_data = tf.gfile.FastGFile(image_path, 'rb').read()
image = Image.open(image_path)
image_array = image.convert('RGB')
# Loads label file, strips off carriage return
label_lines = [line.rstrip() for line
in tf.gfile.GFile("retrained_labels.txt")]
# Unpersists graph from file
with tf.gfile.FastGFile("retrained_graph.pb", 'rb') as f:
graph_def = tf.GraphDef()
graph_def.ParseFromString(f.read())
_ = tf.import_graph_def(graph_def, name='')
with tf.Session() as sess:
# Feed the image_data as input to the graph and get first prediction
softmax_tensor = sess.graph.get_tensor_by_name('final_result:0')
predictions = sess.run(softmax_tensor,{'DecodeJpeg:0': image_array})
# Sort to show labels of first prediction in order of confidence
top_k = predictions[0].argsort()[-len(predictions[0]):][::-1]
for node_id in top_k:
human_string = label_lines[node_id]
score = predictions[0][node_id]
print('%s (score = %.5f)' % (human_string, score))
filename = "results.txt"
with open(filename, 'a+') as f:
f.write('\n**%s**\n' % (image_path))
for node_id in top_k:
human_string = label_lines[node_id]
score = predictions[0][node_id]
f.write('%s (score = %.5f)\n' % (human_string, score))
I want the above code to read in a directory instead of a single image and then process them all and output the scores to the results.txt file.
Currently I can call this like so:
python this_file.py /root/images/1.jpg
How would I get this code to take the following input and processes it
python this_file.py /root/images/
Use os.listdir to list all files in the directory. Qualify it with a filter as well. Join the resulting files to their directory. Read them from the list with a for loop.
python this_file.py /root/images/
image_path = sys.argv[1]
image_paths = [os.path.join(image_path,img) for img in os.listdir(image_path) if '.jpg' in img]
I also recommend re-examining your training function and strategy. It is also good practice to abstract your entire network with tf variable placeholders as far as you can. In addition it would be much more efficient to implement batching, as well as possibly convert your dataset to tf records.
[short summary: how to use TF high-level Estimator on Python with an external file reader? or with feed_dict?]
Been struggling with this for few days, couldn't find any solution on-line...
I'm using TF high-level modules (tf.contrib.learn.Estimator on tf1.0, or tf.estimator.Estimator on tf1.1),
features and targets (x/y) inputted through an input_fn, and the graph built on the model_fn.
Already trained a nn on 'small' data sets, in which the whole input is the part of the graph, using slice_input_producer etc. (I can push an example to github if it serves ppl here).
I try to train a larger nn on 'heavier' data-sets (10s-100s GB).
I have an external Python reader that does some nasty binary file reading, which I really don't want to get into.
This reader has its own queue.Queue with m1 samples. When I use it to extract the m1 {features} & {targets}, the net simply saves all these samples as const. in the first layer of the graph... completely undesired.
I try to either -
feed the output of the external file reader as input to my graph.
define a proper tf queue object that will keep updating the queue (each time a sample is dequeued, i want a completely other sample to be enqueued).
Reminding that I use the "high level", e.g.
self.Estimator = tf.contrib.learn.Estimator(
model_fn=self.model_fn,
model_dir=self.config['model_dir'],
config=tf.contrib.learn.RunConfig( ... ) )
def input_fn(self, mode):
batch_data = self.data[mode].next() # pops out a batch of samples, as numpy 4D matrices
... # some processing of batch data
features_dict = dict(data=batch_data.pop('data'))
targets_dict = batch_data
return features_dict, targets_dict
self.Estimator.fit(input_fn=lambda: self.input_fn(modekeys.TRAIN))
Attached is a final solution for integrating an external reader into the high-level TF api (tf.contrib.learn.Estimator / tf.estimator.Estimator).
Please note:
the architecture and "logic" is not important. it's a stupid simple net.
the external reader outputs a dictionary of numpy matrices.
the input_fn is using this reader.
In order to verify that the reader "pulls new values", I both
save the recent value to self.status (should be > 1.0)
save a summary, to be viewed in tensorboard.
Code example is in gist, and below.
import tensorflow as tf
import numpy as np
modekeys = tf.contrib.learn.ModeKeys
tf.logging.set_verbosity(tf.logging.DEBUG)
# Tested on python 2.7.9, tf 1.1.0
class inputExample:
def __init__(self):
self.status = 0.0 # tracing which value was recently 'pushed' to the net
self.model_dir = 'temp_dir'
self.get_estimator()
def input_fn(self):
# returns features and labels dictionaries as expected by tf Estimator's model_fn
data, labels = tf.py_func(func=self.input_fn_np, inp=[], Tout=[tf.float32, tf.float32], stateful=True)
data.set_shape([1,3,3,1]) # shapes are unknown and need to be set for integrating into the network
labels.set_shape([1,1,1,1])
return dict(data=data), dict(labels=labels)
def input_fn_np(self):
# returns a dictionary of numpy matrices
batch_data = self.reader()
return batch_data['data'], batch_data['labels']
def model_fn(self, features, labels, mode):
# using tf 2017 convention of dictionaries of features/labels as inputs
features_in = features['data']
labels_in = labels['labels']
pred_layer = tf.layers.conv2d(name='pred', inputs=features_in, filters=1, kernel_size=3)
tf.summary.scalar(name='label', tensor=tf.squeeze(labels_in))
tf.summary.scalar(name='pred', tensor=tf.squeeze(pred_layer))
loss = None
if mode != modekeys.INFER:
loss = tf.losses.mean_squared_error(labels=labels_in, predictions=pred_layer)
train_op = None
if mode == modekeys.TRAIN:
train_op = tf.contrib.layers.optimize_loss(
loss=loss,
learning_rate = 0.01,
optimizer = 'SGD',
global_step = tf.contrib.framework.get_global_step()
)
predictions = {'estim_exp': pred_layer}
return tf.contrib.learn.ModelFnOps(mode=mode, predictions=predictions, loss=loss, train_op=train_op)
def reader(self):
self.status += 1
if self.status > 1000.0:
self.status = 1.0
return dict(
data = np.random.randn(1,3,3,1).astype(dtype=np.float32),
labels = np.sin(np.ones([1,1,1,1], dtype=np.float32)*self.status)
)
def get_estimator(self):
self.Estimator = tf.contrib.learn.Estimator(
model_fn = self.model_fn,
model_dir = self.model_dir,
config = tf.contrib.learn.RunConfig(
save_checkpoints_steps = 10,
save_summary_steps = 10,
save_checkpoints_secs = None
)
)
if __name__ == '__main__':
ex = inputExample()
ex.Estimator.fit(input_fn=ex.input_fn)
You can use tf.constant if you have the training data already in python memory as shown in the abalone TF example: https://github.com/tensorflow/tensorflow/blob/r1.1/tensorflow/examples/tutorials/estimators/abalone.py#L138-L141
Note: copying the data from disk to Python to TensorFlow is often less efficient than constructing an input pipeline in TensorFlow (i.e. loading data from disk directly into TensorFlow Tensors), such as using tf.contrib.learn.datasets.base.load_csv_without_header.