Gfyuki

I've used CNN model for text classification and I want to create web interactive application for rendering CNN output with powerful dash framework. I am new to dash and found difficulty how to wrap up my below implementation with simple dash app. Basically I want to render trained CNN model output with interactive plot in front end. How can I do that efficiently? can anyone point me out how to make this happen? any simple dash template that I can wrap up my below code with basic dash app? any idea?

data prep

import pandas as pd

pos_doc = pd.read_csv('imdb/train/pos.csv')

neg_doc = pd.read_csv('imdb/train/neg.csv')



docs   = negative_docs + positive_docs

labels = [0 for _ in range(len(negative_docs))] + [1 for _ in range(len(positive_docs))]

labels = to_categorical(labels)



tokenizer = Tokenizer(num_words=20000)

tokenizer.fit_on_texts(docs)

sequences = tokenizer.texts_to_sequences(docs)

word_index = tokenizer.word_index

data   = pad_sequences(sequences, maxlen=300, padding='post')



random_state = np.random.randint(1000)

X_train, X_val, y_train, y_val = train_test_split(data, labels, test_size=VAL_SIZE, random_state=random_state)

deep learning model for text classification

os.environ['KERAS_BACKEND']='theano'



from keras.preprocessing.text import Tokenizer

from keras.preprocessing.sequence import pad_sequences

from keras.utils.np_utils import to_categorical



from keras.layers import Embedding

from keras.layers import Dense, Input, Flatten

from keras.layers import Conv1D, MaxPooling1D, Embedding, Merge, Dropout

from keras.models import Model



convs = 

filter_sizes = [3,4,5]

sequence_input = Input(shape=(MAX_SEQUENCE_LENGTH,), dtype='int32')

embedded_sequences = embedding_layer(sequence_input)



for fsz in filter_sizes:

    l_conv = Conv1D(nb_filter=128,filter_length=fsz,activation='relu')(embedded_sequences)

    l_pool = MaxPooling1D(5)(l_conv)

    convs.append(l_pool)



l_merge = Merge(mode='concat', concat_axis=1)(convs)

l_cov1= Conv1D(128, 5, activation='relu')(l_merge)

l_pool1 = MaxPooling1D(5)(l_cov1)

l_cov2 = Conv1D(128, 5, activation='relu')(l_pool1)

l_pool2 = MaxPooling1D(30)(l_cov2)

l_flat = Flatten()(l_pool2)

l_dense = Dense(128, activation='relu')(l_flat)

preds = Dense(2, activation='softmax')(l_dense)



model = Model(sequence_input, preds)

model.compile(loss='categorical_crossentropy',

              optimizer='rmsprop',

              metrics=['acc'])



print("model fitting - more complex convolutional neural network")

model.summary()

model.fit(x_train, y_train, validation_data=(x_val, y_val),

          nb_epoch=20, batch_size=50)

to create my dash app for above implementation, I tried to follow the template of hello-world dash but it couldn't render my implementation above. Any idea to get this done? THANKS IN ADVANCE