keras error on predict

Question

I am trying to use a keras neural network to recognize canvas images of drawn digits and output the digit. I have saved the neural network and use django to run the web interface. But whenever I run it, I get an internal server error and an error on the server side code. The error says Exception: Error when checking : expected dense_input_1 to have shape (None, 784) but got array with shape (784, 1). My only main view is

from django.shortcuts import render
from django.http import HttpResponse
import StringIO
from PIL import Image
import numpy as np
import re
from keras.models import model_from_json
def home(request):
    if request.method=="POST":
        vari=request.POST.get("imgBase64","")
        imgstr=re.search(r'base64,(.*)', vari).group(1)
        tempimg = StringIO.StringIO(imgstr.decode('base64'))
        im=Image.open(tempimg).convert("L")
        im.thumbnail((28,28), Image.ANTIALIAS)
        img_np= np.asarray(im)
        img_np=img_np.flatten()
        img_np.astype("float32")
        img_np=img_np/255
        json_file = open('model.json', 'r')
        loaded_model_json = json_file.read()
        json_file.close()
        loaded_model = model_from_json(loaded_model_json)
        # load weights into new model
        loaded_model.load_weights("model.h5")
        # evaluate loaded model on test data
        loaded_model.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])
        output=loaded_model.predict(img_np)
        score=output.tolist()
        return HttpResponse(score)
    else:
        return render(request, "digit/index.html")

The links I have checked out are:

Edit Complying with Rohan's suggestion, this is my stack trace

Internal Server Error: /home/
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/django/core/handlers/base.py", line 149, in get_response
    response = self.process_exception_by_middleware(e, request)
  File "/usr/local/lib/python2.7/dist-packages/django/core/handlers/base.py", line 147, in get_response
    response = wrapped_callback(request, *callback_args, **callback_kwargs)
  File "/home/vivek/keras/neural/digit/views.py", line 27, in home
output=loaded_model.predict(img_np)
  File "/usr/local/lib/python2.7/dist-packages/keras/models.py", line 671, in predict
return self.model.predict(x, batch_size=batch_size, verbose=verbose)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 1161, in predict
check_batch_dim=False)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 108, in standardize_input_data
str(array.shape))
Exception: Error when checking : expected dense_input_1 to have shape (None, 784) but got array with shape (784, 1)

Also, I have my model that I used to train the network initially.

import numpy
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Dropout
from keras.utils import np_utils
# fix random seed for reproducibility
seed = 7
numpy.random.seed(seed)
(X_train, y_train), (X_test, y_test) = mnist.load_data()
for item in y_train.shape:
    print item
num_pixels = X_train.shape[1] * X_train.shape[2]
X_train = X_train.reshape(X_train.shape[0], num_pixels).astype('float32')
X_test = X_test.reshape(X_test.shape[0], num_pixels).astype('float32')
# normalize inputs from 0-255 to 0-1
X_train = X_train / 255
X_test = X_test / 255
print X_train.shape
# one hot encode outputs
y_train = np_utils.to_categorical(y_train)
y_test = np_utils.to_categorical(y_test)
num_classes = y_test.shape[1]
# define baseline model
def baseline_model():
    # create model
    model = Sequential()
    model.add(Dense(num_pixels, input_dim=num_pixels, init='normal', activation='relu'))
    model.add(Dense(num_classes, init='normal', activation='softmax'))
    # Compile model
    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
    return model
# build the model
model = baseline_model()
# Fit the model
model.fit(X_train, y_train, validation_data=(X_test, y_test), nb_epoch=20, batch_size=200, verbose=1)
# Final evaluation of the model
scores = model.evaluate(X_test, y_test, verbose=0)
print("Baseline Error: %.2f%%" % (100-scores[1]*100))
# serialize model to JSON
model_json = model.to_json()
with open("model.json", "w") as json_file:
    json_file.write(model_json)
# serialize weights to HDF5
model.save_weights("model.h5")
print("Saved model to disk")

Edit I tried reshaping the img to (1,784) and it also failed, giving the same error as the title of this question

Thanks for the help, and leave comments on how I should add to the question.

Check if this works outside of django. Post full stack trace. How did you train the model? — Rohan, Oct 10 '16 at 03:35
@Rohan I added the traceback stack and the original keras file, so take I look — Superman, Oct 10 '16 at 13:34
Have you try to do something like: "img_np.reshape((None, 784))" before "loaded_model.predict(img_np)" ? — Dror Hilman, Oct 14 '16 at 03:32
@Dror Hillman I have tried this, but it then give the error "Value must be an integer" and fails, so this doesn't work — Superman, Oct 14 '16 at 15:15
I think your network needs to receive the pixels one by one, or at least you are passing the `X_train` as the training set as a dataset of shape (784,1). That is 784 samples of one pixel. But you have declared your `input_dim` as 784 and the network will have that number of neurons in the input layer. Or either you change the input_dim to 1 or make the train be an array of elements with length 784. — marquex, Oct 14 '16 at 21:25
@marquex so your saying the shape needs to be (784, 1) instead of (1,784). My goal is to pass one sample of 784 pixels — Superman, Oct 15 '16 at 20:38
https://stackoverflow.com/questions/47295025/valueerror-at-image-tensor-tensoractivation-5-softmax0-shape-4-dtyp/47300005?noredirect=1#comment81555441_47300005 any suggestions — Dexter, Nov 15 '17 at 13:16

score 31 · Accepted Answer · answered Nov 08 '16 at 14:14

You're asking the neural network to evaluate 784 cases with one input each instead of a single case with 784 inputs. I had the same problem and I solved it having an array with a single element which is an array of the inputs. See the example below, the first one works whereas the second one gives the same error you're experiencing.

model.predict(np.array([[0.5, 0.0, 0.1, 0.0, 0.0, 0.4, 0.0, 0.0, 0.1, 0.0, 0.0]]))
model.predict(np.array([0.5, 0.0, 0.1, 0.0, 0.0, 0.4, 0.0, 0.0, 0.1, 0.0, 0.0]))

hope this solves it for you as well :)

https://stackoverflow.com/questions/47295025/valueerror-at-image-tensor-tensoractivation-5-softmax0-shape-4-dtyp/47300005?noredirect=1#comment81555441_47300005 any suggesions — Dexter, Nov 15 '17 at 13:16

keras error on predict

1 Answers1

Linked