Keras error - ValueError: could not convert string to float

Question

I have already been through questions dealing with similar problems. However, they have not been able to answer my specific issue. Hence, I would sincerely appreciate any suggestions to overcome the following issue that I am facing.

I am trying to implement an RNN model for a text classification problem. I have a csv file of sentences(triple) and a class label[0, 1] (truth) in my triple.csv file.

A Sample of the triple.csv file

triple,truth
sportsteam hawks teamplaysincity city atlanta,1
stadiumoreventvenue hondacenter stadiumlocatedincity city anaheim,1
sportsteam ducks teamplaysincity city anaheim,1
sportsteam n1985chicagobears teamplaysincity city chicago,1
...

I am trying to train the sentences (triples) using an RNN and their word2vec embeddings. However, I keep getting the following error.

ValueError: could not convert string to float: 'sportsleague nfl leaguestadiums stadiumoreventvenue heinzfield'

My main code

import string
import nltk
from nltk.tokenize import word_tokenize
from nltk.corpus import stopwords
import numpy as np
import gensim
import pandas as pd
import os
from tensorflow.python.keras.preprocessing.text import Tokenizer
from tensorflow.python.keras.preprocessing.sequence import pad_sequences
from gensim.models import Word2Vec, KeyedVectors
from keras.models import Sequential
from keras.layers import Dense, Embedding, LSTM, GRU
from keras.layers.embeddings import Embedding
from keras.initializers import Constant
from sklearn.metrics import precision_recall_fscore_support
from sklearn.model_selection import train_test_split
from termcolor import colored
from keras.utils import to_categorical

nltk.download('stopwords')
# one hot encode

df = pd.DataFrame()
df = pd.read_csv('data/triple.csv')
triple_lines = list()
lines = df['triple'].values.tolist()

for line in lines:
    tokens = word_tokenize(line)
    tokens = [w.lower() for w in tokens]
    table = str.maketrans('','',string.punctuation)
    stripped = [w.translate(table) for w in tokens]
    words = [word for word in stripped if word.isalpha()]
    stop_words = set(stopwords.words('english'))
    words = [w for w in words if not w in stop_words]
    triple_lines.append(words)

print(colored(len(triple_lines),'green'))
EMBEDDING_DIM = 100
model = gensim.models.Word2Vec(sentences=triple_lines, size=EMBEDDING_DIM, window =5, workers=4, min_count=1)
words = list(model.wv.vocab)
print(colored('Vocabulary size: %d' % len(words),'green'))

filename = 'embedding_word2vec.txt'
model.wv.save_word2vec_format(filename,binary=False)

embedding_index = {}
f = open(os.path.join('', 'embedding_word2vec.txt'), encoding="utf-8")
for line in f:
    values = line.split()
    word = values[0]
    coefs = np.asarray(values[1:])
    embedding_index[word] = coefs
f.close()

#Vectorize the text samples into a S2 integer tensor
tokenizer_obj = Tokenizer()
tokenizer_obj.fit_on_texts(triple_lines)
sequences = tokenizer_obj.texts_to_sequences(triple_lines)

#pad sequences
word_index = tokenizer_obj.word_index
print(colored('Found %s unique tokens.'% len(word_index),'magenta'))

max_length = 9

triple_pad = pad_sequences(sequences, maxlen=max_length)
truth = df['triple'].values
print('Shape of triple tensor: ', triple_pad.shape)
print('Shape of truth tensor: ', truth.shape)

#map embeddings from loaded word2vec model for each word to the tokenizer_obj.word_index vocabulary & create a wordvector matrix

num_words = len(word_index)+1
embedding_matrix = np.zeros((num_words, EMBEDDING_DIM))

for word,i in word_index.items():
    if i>num_words:
        continue
    embedding_vector = embedding_index.get(word)
    if embedding_vector is not None:
        # words not found in the embedding index will be all-zero
        embedding_matrix[i] = embedding_vector

print(colored(num_words,'cyan'))

# Define Model
model = Sequential()
embedding_layer = Embedding(num_words,
                            EMBEDDING_DIM,
                            embeddings_initializer=Constant(embedding_matrix),
                            input_length=max_length,
                            trainable=False)

model.add(embedding_layer)
model.add(GRU(units=32, dropout=0.2, recurrent_dropout=0.2))
model.add(Dense(1, activation='sigmoid'))

model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
print(colored(model.summary(),'cyan'))

#Split the data into training set and validation set
VALIDATION_SPLIT = 0.2

indices = np.arange(triple_pad.shape[0])
np.random.shuffle(indices)
triple_pad = triple_pad[indices]
truth = truth[indices]
num_validation_samples = int(VALIDATION_SPLIT * triple_pad.shape[0])

X_train_pad = triple_pad[:-num_validation_samples]
y_train = truth[:-num_validation_samples]
X_test_pad = triple_pad[-num_validation_samples:]
y_test = truth[-num_validation_samples:]

print('Shape of X_train_pad tensor: ',X_train_pad.shape)
print('Shape of y_train tensor: ',y_train.shape)
print('Shape of X_test_pad tensor: ',X_test_pad.shape)
print('Shape of y_test tensor: ',y_test.shape)

print(colored('Training...','green'))
model.fit(X_train_pad, y_train, batch_size=128, epochs=25, validation_data=(X_test_pad, y_test), verbose=2)

Any help on how to overcome this issue will be much appreciated.

Answer 1

I encountered this error as I was passing strings y_train to the model.fit().

Instead of defining the boolean truths as the target class values, I had defined the triples as the target class, which was passing strings into the model.fit() as shown below.

truth = df['triple'].values

So, simply modifying the above line as follows solved this issue.

truth = df['truth'].values

It's crazy how I miss these trivial details. Silly me!