(BOW, TF-IDF, Word2Vec, BERT) Word Embeddings + (SVM, Naive Bayes, Decision Tree, Random Forest) Base Classifiers + Pre-trained BERT on Tensorflow Hub + 1-D CNN and Bi-Directional LSTM on IMDB Movie Reviews Dataset
(BOW, TF-IDF, Word2Vec, BERT) Word Embeddings + (SVM, Naive Bayes, Decision Tree, Random Forest) Base Classifiers + Pre-trained BERT on Tensorflow Hub + 1-D CNN and Bi-Directional LSTM on IMDB Movie Reviews Dataset
Results for Base Classifiers
Rank
Word Embedding
Classifier
Accuracy
F1-Score
1
BERT Sentence Version (Mean Bert Features per Review)
SVM
90.35
0.90
2
BERT Sentence Version (Mean Bert Features per Review)
MLP
90.32
0.90
3
TFIDF with Stop Words
SVM
89.59
0.90
Results for Deep Neural Networks
Rank
Word Embedding
Model
Accuracy
1
BERT TensorFlow-HUB
Bi-Directional LSTM
91.34
2
BERT Sentence Version (Mean Bert Features per Review)
Hi @hmohebbi thanks for sharing your code. I have a questionI appreciate if you can help me with that. I can run the coder for BOW and TFIDF however for word2vec and Bert keeps raising this error:
530 else:
--> 531 array = np.asarray(array, order=order, dtype=dtype)
532 except ComplexWarning:
533 raise ValueError("Complex data not supported\n"
ValueError: setting an array element with a sequence.```
Do you have any idea of what could be the reason?
I searched the net and they mentioning that this error will arise whenever the list has not been prepared correctly for example [2, [2,3]].
But I could not check where could be wrong.
~\anaconda3\envs\gpu3\lib\zipfile.py in _RealGetContents(self)
1196 raise BadZipFile("File is not a zip file")
1197 if not endrec:
-> 1198 raise BadZipFile("File is not a zip file")
1199 if self.debug > 1:
1200 print(endrec)