bhaveshoswal/cnn-text-classification-keras

Text Classification by Convolutional Neural Network in Keras

CNN-text-classification-keras

It is simplified implementation of Implementing a CNN for Text Classification in TensorFlow in Keras as functional api

Requirements

• Python 3.5.2
• Keras 2.1.2
• Tensorflow 1.4.1

Traning

Run the below command and it will run for 100 epochs if you want change it just open model.py

python model.py

For new data

You have to rebuild the vocabulary and then train.

For Citation

@misc{bhaveshoswal,
  author = {Bhavesh Vinod Oswal},
  title = {CNN-text-classification-keras},
  year = {2016},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/bhaveshoswal/CNN-text-classification-keras}},
}

cnn-text-classification-keras's Issues

How to cite it?

Very easy to use, thanks! Any idea how to cite this implementation?

Error in shape

Hi,
I have used the following code to define the model, in-line with your code

def define_model(max_length, vocab_size):
inputs = Input(shape=(max_length,))
embedding1 = Embedding(vocab_size, 50)(inputs)
reshape = Reshape((max_length, 50, 1))(embedding1)

conv1 = Conv2D(num_filters, filter_sizes[0], activation='relu')(reshape)
#conv1 = Conv2D(filters=100, filter_sizes[0], activation='relu')(reshape)
conv2 = Conv2D(num_filters, filter_sizes[1], activation='relu')(reshape)
conv3 = Conv2D(num_filters, filter_sizes[2], activation='relu')(reshape)
pool1 = MaxPooling2D(pool_size=(max_length - filter_sizes[0] + 1, 1), strides=(1,1), border_mode='valid', dim_ordering='tf')(conv1)
pool2 = MaxPooling2D(pool_size=(max_length - filter_sizes[1] + 1, 1), strides=(1,1), border_mode='valid', dim_ordering='tf')(conv2)
pool3 = MaxPooling2D(pool_size=(max_length - filter_sizes[2] + 1, 1), strides=(1,1), border_mode='valid', dim_ordering='tf')(conv3)
merged = merge([pool1, pool2, pool3], mode='concat', concat_axis=1)
flatten = Flatten()(merged)
dropout = Dropout(0.5)(flatten)
dense1 = Dense(10, activation='relu')(dropout)
outputs = Dense(1, activation='sigmoid')(dense1)
model = Model(inputs=inputs, outputs=outputs)
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
model.summary()
return model

However, i get the following error
"concat" mode can only merge layers with matching output shapes except for the concat axis. Layer shapes: [(None, 1, 48, 100), (None, 1, 47, 100), (None, 1, 46, 100)]

The filter_size i am using is 3,4,5. Max_length is the maximum length of the document.

In the Max pooling layer, since we subtract the max_length from different filter sizes, the shape will change. Can you please guide on how to rectify it?

Language Independent or not

Is this classification method language independent? I want to classify some bengali sentences written in transliterated form.

Exception ignored in: <bound method BaseSession.__del__ of <tensorflow.python.client.session.Session object at 0x7f1bf88e85f8>>

for testing I set epoch to 2
after epoch 2 passed it saved model weights.001-0.6057.hdf5 and this is the output:

Epoch 00002: val_acc improved from 0.60572 to 0.66245, saving model to weights.002-0.6624.hdf5
Exception ignored in: <bound method BaseSession.__del__ of <tensorflow.python.client.session.Session object at 0x7f1bf88e85f8>>
Traceback (most recent call last):
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 702, in __del__
TypeError: 'NoneType' object is not callable
marn@marn-HP-Compaq-Elite-8300-SFF ~/c/t/CNN-text-classification-keras> ls
_config.yml  data_helpers.py  __pycache__/  weights.001-0.6057.hdf5
data/        model.py         README.md     weights.002-0.6624.hdf5

is the error important? And how can I use this trained models?
thank you.

ValueError: Error when checking input: expected input_14 to have shape (None, 1850) but got array with shape (1480, 1)

Hi,
I used you helper code to prepare my dataset and then run the model. i have:
X_train=(1480,0)
y_train = (850,0)
X_test = (370,0)
y_test = (370,0)
sequence_length = 1850
the rest of parameters are the same, when i run the model, i got the error:ValueError: Error when checking input: expected input_14 to have shape (None, 1850) but got array with shape (1480, 1).

sorry, i am new, what should i do?
thanks,