florishoogenboom / keras-han-for-docla Goto Github PK

WARNING:tensorflow:
The following Variables were used a Lambda layer's call (lambda_4), but
are not present in its tracked objects:
<tf.Variable 'sentence_attention/W:0' shape=(100, 100) dtype=float32>
<tf.Variable 'sentence_attention/context_vector:0' shape=(100, 1) dtype=float32>
It is possible that this is intended behavior, but it is more likely
an omission. This is a strong indication that this layer should be
formulated as a subclassed Layer rather than a Lambda layer.

Hi, how do I predict with the model?
Tried this but didn't work

print('Test...')
test_data = pd.read_csv('C:/Users/UOU/Desktop/ebifolder/mpm/keras-han-for-docla/trialdata/testData.tsv', sep='\t')
test_data = test_data['review'].values

result = han_model.predict(test_data)

from keras_han.models import HAN
Should be changed for
from keras_han.model import HAN

This is the error I got

TypeError Traceback (most recent call last)
in
132 # Load the embeddings from a file
133 embeddings = {}
--> 134 with open('C:/Users/UOU/Desktop/ebifolder/mpm/keras-han-for-docla/trialdata/glove.6B.100d.txt' % GLOVE_DIM, encoding='utf-8') as file:
135 for line in file:
136 values = line.split()

TypeError: not all arguments converted during string formatting

Any ideas?

Hello and thanks for sharing your code!
I have a question, what do the input layers represent in this model? I think the first input represents the word indexes, so in that case what does input_2 represent? Is it the encoded states of the sentences? In that case shouldn't it be a float?