TF Hub Fine Tuning Error : ValueError: Failed to find data adapter that can handle input

Question

I am running a simple fine tuning using Cosine Similiarity loss code and getting following error. Can you please help me to understand the issue and resolve.

Tensorflow Version: 2.0

Error:

ValueError: Failed to find data adapter that can handle input: (class 'list' containing values of types {"class 'str'"}), class 'numpy.ndarray'>

Code

import tensorflow as tf
import tensorflow_hub as hub
from tensorflow import keras
huburl='https://tfhub.dev/google/universal-sentence-encoder-large/5'
left_loaded_module_obj = hub.load(huburl)

left_input = keras.Input(shape=(), dtype=tf.string)


# left_trainable_embedding_layer = hub.KerasLayer(huburl,trainable=True)
left_trainable_embedding_layer = hub.KerasLayer(left_loaded_module_obj,trainable=True)

embedding_left_output= left_trainable_embedding_layer(left_input)


model = tf.keras.Model(left_input, embedding_left_output)
model.compile('sgd', loss=tf.keras.losses.CosineSimilarity(axis=-1))
model.summary()

import tensorflow_hub as hub
import tensorflow as tf
tf.enable_eager_execution()
embed = hub.load("https://tfhub.dev/google/universal-sentence-encoder-large/5")
text_list=["The quick brown fox jumps over the lazy dog.",
    "I am a sentence for which I would like to get its embedding"]
embeddings = embed(text_list)

print (embeddings.numpy())

embed_target=embeddings.numpy()
print (type(embed_target),embed_target.shape)

model.fit(x=text_list,y=embed_target)

Answer 1

Simply convert the list y into a numpy array, and it should do the trick. Something like this:

y = np.array(y)