I'm using Keras to implement a sentiment analysis code. I have my training data as follows:
I build my code in a similar fashion to here
The only difference is that their data is imported from Keras dataset while mine are text file
This is my code
# CNN for the IMDB problem
top_words = 5000
pos_file=open('pos.txt', 'r')
neg_file=open('neg.txt', 'r')
# Load data from files
pos = list(pos_file.readlines())
neg = list(neg_file.readlines())
x = pos + neg
total = numpy.array(x)
# Generate labels
positive_labels = [1 for _ in pos]
negative_labels = [0 for _ in neg]
y = numpy.concatenate([positive_labels, negative_labels], 0)
#Testing
pos_test=open('posTest.txt', 'r')
posT = list(pos_test.readlines())
print("pos length is",len(posT))
neg_test=open('negTest.txt', 'r')
negT = list(neg_test.readlines())
xTest = pos + negT
total2 = numpy.array(xTest)
# Generate labels
positive_labels2 = [1 for _ in posT]
negative_labels2 = [0 for _ in negT]
yTest = numpy.concatenate([positive_labels2, negative_labels2], 0)
#Create model
max_words = 1
model = Sequential()
model.add(Embedding(top_words, 32, input_length=max_words))
model.add(Conv1D(filters=32, kernel_size=3, padding='same', activation='relu'))
model.add(MaxPooling1D(pool_size=1))
model.add(Flatten())
model.add(Dense(250, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
print(model.summary())
#Fit the model
model.fit(total, y, validation_data=(xTest, yTest), epochs=2, batch_size=128, verbose=2)
# Final evaluation of the model
scores = model.evaluate(total2, yTest, verbose=0)
print("Accuracy: %.2f%%" % (scores[1]*100))
When I run my code , I get this error
File "C:\Users\\Anaconda3\lib\site-packages\keras\engine\training.py", line 70, in <listcomp>
data = [np.expand_dims(x, 1) if x is not None and x.ndim == 1 else x for x in data]
AttributeError: 'str' object has no attribute 'ndim'
You are feeding a list of strings to a model which is something it does not expect. You can use keras.preprocessing.text
module to convert the text to an integer sequence. More specifically you can prepare data like:
from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
tk = Tokenizer()
tk.fit_on_texts(texts)
index_list = tk.texts_to_sequences(texts)
x_train = pad_sequences(index_list, maxlen=maxlen)
Now x_train
(a n_samples * maxlen
ndarray of type np.int
) is a legitimate input for the model.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With