Hvass-Labs / TensorFlow-Tutorials

TensorFlow Tutorials with YouTube Videos

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Problem loading saved model

miftahihsan opened this issue · comments

(This problem is from NLP number 20)

Hello,
I was having problem saving and loading this model
I used model.save(path.h5) to save the model

used the code below to load the model

==========================================================

import tensorflow as tf
from tensorflow.python.keras.preprocessing.text import Tokenizer
from tensorflow.python.keras.preprocessing.sequence import pad_sequences

pad = 'pre'
num_words = 10000
max_tokens = 544

tokenizer = Tokenizer(num_words)

text = ["Great movie!", "Worst Movie ever!"]

tokens = tokenizer.texts_to_sequences(text)

tokens_pad = pad_sequences(tokens, maxlen=max_tokens,
padding=pad, truncating=pad)

new_model = tf.keras.models.load_model("path.h5")

new_model.predict(tokens_pad)

===========================================================

my all my results are the same
rather then having two different +ve and -ve sentiment

array([[0.57864296],
[0.57864296]], dtype=float32)

If possible please do tell me what I am doing wrong, and how to fix this issue
Thank You!