错误节点：“binary_crossentropy/Cast”在训练模型时不支持将字符串转换为浮点数

Question

我想训练我的数据，我已经将我的数据与 word2vec 预训练模型进行字符串连接https://dl.fbaipublicfiles.com/fasttext/vectors-crawl/cc.id.300.vec.gz，并成功创建模型，但是当我想训练数据集时，我收到了这样的错误

UnimplementedError                        Traceback (most recent call last)
<ipython-input-28-85ce60cd1ded> in <module>()
      1 history = model.fit(X_train, y_train, epochs=6,
      2                     validation_data=(X_test, y_test),
----> 3                     validation_steps=30)

1 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)
     53     ctx.ensure_initialized()
     54     tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
---> 55                                         inputs, attrs, num_outputs)
     56   except core._NotOkStatusException as e:
     57     if name is not None:

UnimplementedError: Graph execution error:

#skiping error

Node: 'binary_crossentropy/Cast'
Cast string to float is not supported
     [[{{node binary_crossentropy/Cast}}]] [Op:__inference_train_function_21541]

代码 ：

file = gzip.open(urlopen('https://dl.fbaipublicfiles.com/fasttext/vectors-crawl/cc.id.300.vec.gz'))
vocab_and_vectors = {}
# put words as dict indexes and vectors as words values
for line in file:
    values = line.split()
    word = values [0].decode('utf-8')
    vector = np.asarray(values[1:], dtype='float32')
    vocab_and_vectors[word] = vector

from sklearn.model_selection import train_test_split
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.utils import to_categorical

# how many features should the tokenizer extract
features = 500
tokenizer = Tokenizer(num_words = features)
# fit the tokenizer on our text
tokenizer.fit_on_texts(df["Comment"].tolist())

# get all words that the tokenizer knows
word_index = tokenizer.word_index
print(len(word_index))
# put the tokens in a matrix
X = tokenizer.texts_to_sequences(df["Comment"].tolist())
X = pad_sequences(X)

# prepare the labels
y = df["sentiments"].values

# split in train and test
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, stratify=y, random_state=30)
print(X_train.shape, y_train.shape)
print(X_test.shape, y_test.shape)

embedding_matrix = np.zeros((len(word_index) + 1, 300))
for word, i in word_index.items():
    embedding_vector = vocab_and_vectors.get(word)
    # words that cannot be found will be set to 0
    if embedding_vector is not None:
        embedding_matrix[i] = embedding_vector

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense, Embedding

model = tf.keras.Sequential([
    tf.keras.layers.Embedding(len(word_index)+1, 300, input_length=X.shape[1], weights=[embedding_matrix], trainable=False),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(64,  return_sequences=True)),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(32)),
    tf.keras.layers.Dense(64, activation='relu'),
    tf.keras.layers.Dropout(0.5),
    tf.keras.layers.Dense(1, activation='sigmoid')
])
model.summary()
model.compile(loss='binary_crossentropy',
              optimizer=tf.keras.optim

izers.Adam(1e-4),
              metrics=['accuracy'])

history = model.fit(X_train, y_train, epochs=6,
                    validation_data=(X_test, y_test), 
                    validation_steps=30)

我的数据：

Answer 1

我认为问题是，您将字符串作为类（“positif”，“negativ”），如数据摘录中所示。只需将它们转换为

负数 = 0正数 = 1

它应该有效。如果不是，我们将不得不更深入地研究您的代码。

编辑： 所以怀疑字符串作为类的问题。您可以将它们更改为浮动：

df["sentiments"].loc[df["sentiments"]=="positif"]=1.0
df["sentiments"].loc[df["sentiments"]=="negatif"]=0.0

之后，您还应该将 的 dtype 更改y-nparry为 float：

y = np.asarray(y).astype("float64")

这应该可以解决问题。作为参考，另请参阅我的colab 代码