Mixing
ValueError(“Tensor %s is not an element of this graph.” % obj)
0
First of all, English is not my mother language so excuse me, if I don't express myself really well, please feel free to correct me.
I'm making an emotion recognition system that uses rest services to send the image from the client's browser.
This is the code:
# hyper-parameters for bounding boxes shape
frame_window = 10
emotion_offsets = (20, 40)
# loading models
face_detection = load_detection_model(detection_model_path)
emotion_classifier = load_model(emotion_model_path, compile=False)
K.clear_session()
# getting input model shapes for inference
emotion_target_size = emotion_classifier.input_shape[1:3]
# starting lists for calculating modes
emotion_window =
And the function:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
I'm facing this error when i run my code using waitress:
File "c:usersafgirdocumentspythonprojectsface_recovenvlibsite-packagestensorflowpythonframeworkops.py", line 3569, in _as_graph_element_locked
raise ValueError("Tensor %s is not an element of this graph." % obj)
ValueError: Tensor Tensor("predictions_1/Softmax:0", shape=(?, 7), dtype=float32) is not an element of this graph.
It loads the image and does all the processing well, I'm pretty sure that the error is in the emotion_classifier.predict line, just don't know how to fix it.
I've tried with the two solutions in this question and none of them worked.
I'm really new using Tensorflow so I'm kinda stuck with this.
python rest tensorflow keras computer-vision
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
add a comment |
0
First of all, English is not my mother language so excuse me, if I don't express myself really well, please feel free to correct me.
I'm making an emotion recognition system that uses rest services to send the image from the client's browser.
This is the code:
# hyper-parameters for bounding boxes shape
frame_window = 10
emotion_offsets = (20, 40)
# loading models
face_detection = load_detection_model(detection_model_path)
emotion_classifier = load_model(emotion_model_path, compile=False)
K.clear_session()
# getting input model shapes for inference
emotion_target_size = emotion_classifier.input_shape[1:3]
# starting lists for calculating modes
emotion_window =
And the function:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
I'm facing this error when i run my code using waitress:
File "c:usersafgirdocumentspythonprojectsface_recovenvlibsite-packagestensorflowpythonframeworkops.py", line 3569, in _as_graph_element_locked
raise ValueError("Tensor %s is not an element of this graph." % obj)
ValueError: Tensor Tensor("predictions_1/Softmax:0", shape=(?, 7), dtype=float32) is not an element of this graph.
It loads the image and does all the processing well, I'm pretty sure that the error is in the emotion_classifier.predict line, just don't know how to fix it.
I've tried with the two solutions in this question and none of them worked.
I'm really new using Tensorflow so I'm kinda stuck with this.
python rest tensorflow keras computer-vision
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
why do you use tf.Graph()?
– Geeocode
Nov 21 '18 at 22:41
It was a "Solution" that i found in a GitHub forum, but it didn't work neither.
– Andrés Girón
Nov 21 '18 at 22:51
See my answer, please comment if I would miss something
– Geeocode
Nov 21 '18 at 22:53
add a comment |
0
0
0
First of all, English is not my mother language so excuse me, if I don't express myself really well, please feel free to correct me.
I'm making an emotion recognition system that uses rest services to send the image from the client's browser.
This is the code:
# hyper-parameters for bounding boxes shape
frame_window = 10
emotion_offsets = (20, 40)
# loading models
face_detection = load_detection_model(detection_model_path)
emotion_classifier = load_model(emotion_model_path, compile=False)
K.clear_session()
# getting input model shapes for inference
emotion_target_size = emotion_classifier.input_shape[1:3]
# starting lists for calculating modes
emotion_window =
And the function:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
I'm facing this error when i run my code using waitress:
File "c:usersafgirdocumentspythonprojectsface_recovenvlibsite-packagestensorflowpythonframeworkops.py", line 3569, in _as_graph_element_locked
raise ValueError("Tensor %s is not an element of this graph." % obj)
ValueError: Tensor Tensor("predictions_1/Softmax:0", shape=(?, 7), dtype=float32) is not an element of this graph.
It loads the image and does all the processing well, I'm pretty sure that the error is in the emotion_classifier.predict line, just don't know how to fix it.
I've tried with the two solutions in this question and none of them worked.
I'm really new using Tensorflow so I'm kinda stuck with this.
python rest tensorflow keras computer-vision
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
First of all, English is not my mother language so excuse me, if I don't express myself really well, please feel free to correct me.
I'm making an emotion recognition system that uses rest services to send the image from the client's browser.
This is the code:
# hyper-parameters for bounding boxes shape
frame_window = 10
emotion_offsets = (20, 40)
# loading models
face_detection = load_detection_model(detection_model_path)
emotion_classifier = load_model(emotion_model_path, compile=False)
K.clear_session()
# getting input model shapes for inference
emotion_target_size = emotion_classifier.input_shape[1:3]
# starting lists for calculating modes
emotion_window =
And the function:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
I'm facing this error when i run my code using waitress:
File "c:usersafgirdocumentspythonprojectsface_recovenvlibsite-packagestensorflowpythonframeworkops.py", line 3569, in _as_graph_element_locked
raise ValueError("Tensor %s is not an element of this graph." % obj)
ValueError: Tensor Tensor("predictions_1/Softmax:0", shape=(?, 7), dtype=float32) is not an element of this graph.
It loads the image and does all the processing well, I'm pretty sure that the error is in the emotion_classifier.predict line, just don't know how to fix it.
I've tried with the two solutions in this question and none of them worked.
I'm really new using Tensorflow so I'm kinda stuck with this.
python rest tensorflow keras computer-vision
python rest tensorflow keras computer-vision
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
edited Nov 21 '18 at 23:31
Geeocode
2,3501820
2,3501820
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
asked Nov 21 '18 at 22:19
Andrés GirónAndrés Girón
357
357
why do you use tf.Graph()?
– Geeocode
Nov 21 '18 at 22:41
It was a "Solution" that i found in a GitHub forum, but it didn't work neither.
– Andrés Girón
Nov 21 '18 at 22:51
See my answer, please comment if I would miss something
– Geeocode
Nov 21 '18 at 22:53
add a comment |
why do you use tf.Graph()?
– Geeocode
Nov 21 '18 at 22:41
It was a "Solution" that i found in a GitHub forum, but it didn't work neither.
– Andrés Girón
Nov 21 '18 at 22:51
See my answer, please comment if I would miss something
– Geeocode
Nov 21 '18 at 22:53
why do you use tf.Graph()?
– Geeocode
Nov 21 '18 at 22:41
why do you use tf.Graph()?
– Geeocode
Nov 21 '18 at 22:41
It was a "Solution" that i found in a GitHub forum, but it didn't work neither.
– Andrés Girón
Nov 21 '18 at 22:51
It was a "Solution" that i found in a GitHub forum, but it didn't work neither.
– Andrés Girón
Nov 21 '18 at 22:51
See my answer, please comment if I would miss something
– Geeocode
Nov 21 '18 at 22:53
See my answer, please comment if I would miss something
– Geeocode
Nov 21 '18 at 22:53
add a comment |
1 Answer
1
active
oldest
votes
0
I'm just trying to find out your real environment, but I guess you may use Keras amd some Keras model to predict emotions.
Your error message caused because of the line:
K.clear_session()
which, from the documentation: keras.backend.clear_session().
So you clear all graph it has been created, then you try to run the classifier's predict(), which lost all its context this way.
Thus just simply delete this line.
This section is was about some code the Op deleted:
In this task you don't need to use tf.Graph() at all. You just simply should call emotion_classifier.predict() as a simple python method outside and without of using any tensorflow graph:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
|
show 16 more comments
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
draft saved
draft discarded
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53421233%2fvalueerrortensor-s-is-not-an-element-of-this-graph-obj%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
0
I'm just trying to find out your real environment, but I guess you may use Keras amd some Keras model to predict emotions.
Your error message caused because of the line:
K.clear_session()
which, from the documentation: keras.backend.clear_session().
So you clear all graph it has been created, then you try to run the classifier's predict(), which lost all its context this way.
Thus just simply delete this line.
This section is was about some code the Op deleted:
In this task you don't need to use tf.Graph() at all. You just simply should call emotion_classifier.predict() as a simple python method outside and without of using any tensorflow graph:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
|
show 16 more comments
0
I'm just trying to find out your real environment, but I guess you may use Keras amd some Keras model to predict emotions.
Your error message caused because of the line:
K.clear_session()
which, from the documentation: keras.backend.clear_session().
So you clear all graph it has been created, then you try to run the classifier's predict(), which lost all its context this way.
Thus just simply delete this line.
This section is was about some code the Op deleted:
In this task you don't need to use tf.Graph() at all. You just simply should call emotion_classifier.predict() as a simple python method outside and without of using any tensorflow graph:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
|
show 16 more comments
0
0
0
I'm just trying to find out your real environment, but I guess you may use Keras amd some Keras model to predict emotions.
Your error message caused because of the line:
K.clear_session()
which, from the documentation: keras.backend.clear_session().
So you clear all graph it has been created, then you try to run the classifier's predict(), which lost all its context this way.
Thus just simply delete this line.
This section is was about some code the Op deleted:
In this task you don't need to use tf.Graph() at all. You just simply should call emotion_classifier.predict() as a simple python method outside and without of using any tensorflow graph:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
I'm just trying to find out your real environment, but I guess you may use Keras amd some Keras model to predict emotions.
Your error message caused because of the line:
K.clear_session()
which, from the documentation: keras.backend.clear_session().
So you clear all graph it has been created, then you try to run the classifier's predict(), which lost all its context this way.
Thus just simply delete this line.
This section is was about some code the Op deleted:
In this task you don't need to use tf.Graph() at all. You just simply should call emotion_classifier.predict() as a simple python method outside and without of using any tensorflow graph:
def detect_emotion(self, img):
# Convert RGB to BGR
bgr_image = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
faces = detect_faces(face_detection, gray_image)
for face_coordinates in faces:
x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets)
gray_face = gray_image[y1:y2, x1:x2]
try:
gray_face = cv2.resize(gray_face, (emotion_target_size))
except:
continue
gray_face = preprocess_input(gray_face, True)
gray_face = np.expand_dims(gray_face, 0)
gray_face = np.expand_dims(gray_face, -1)
emotion_classifier._make_predict_function()
emotion_prediction = emotion_classifier.predict(gray_face)
emotion_probability = np.max(emotion_prediction)
emotion_label_arg = np.argmax(emotion_prediction)
emotion_text = emotion_labels[emotion_label_arg]
emotion_window.append(emotion_text)
if len(emotion_window) > frame_window:
emotion_window.pop(0)
try:
emotion_mode = mode(emotion_window)
except:
continue
if emotion_text == 'angry':
color = emotion_probability * np.asarray((255, 0, 0))
elif emotion_text == 'sad':
color = emotion_probability * np.asarray((0, 0, 255))
elif emotion_text == 'happy':
color = emotion_probability * np.asarray((255, 255, 0))
elif emotion_text == 'surprise':
color = emotion_probability * np.asarray((0, 255, 255))
else:
color = emotion_probability * np.asarray((0, 255, 0))
color = color.astype(int)
color = color.tolist()
draw_bounding_box(face_coordinates, rgb_image, color)
draw_text(face_coordinates, rgb_image, emotion_mode,
color, 0, -45, 1, 1)
img = Image.fromarray(rgb_image)
return img
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
edited Nov 25 '18 at 1:16
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
answered Nov 21 '18 at 22:52
GeeocodeGeeocode
2,3501820
2,3501820
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
|
show 16 more comments
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
In the moment it detects a face, totally freezes and throws the same error, i'm using real time video streaming and when it receives the first image with a face in it, crashes.
– Andrés Girón
Nov 21 '18 at 23:05
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
@AndrésGirón first try to pass a single image to the predict function and see what happens and please reload the kernel to be sure. Simply you can't get the same error message
– Geeocode
Nov 21 '18 at 23:12
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
How do I reload the kernel?
– Andrés Girón
Nov 21 '18 at 23:18
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
@AndrésGirón How do you run your code? Command line or?
– Geeocode
Nov 21 '18 at 23:21
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
Command line with waitress-serve
– Andrés Girón
Nov 21 '18 at 23:32
|
show 16 more comments
draft saved
draft discarded
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
draft saved
draft discarded
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53421233%2fvalueerrortensor-s-is-not-an-element-of-this-graph-obj%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
why do you use tf.Graph()?
– Geeocode
Nov 21 '18 at 22:41
It was a "Solution" that i found in a GitHub forum, but it didn't work neither.
– Andrés Girón
Nov 21 '18 at 22:51
See my answer, please comment if I would miss something
– Geeocode
Nov 21 '18 at 22:53