| import os |
| import pandas as pd |
| import tensorflow as tf |
| import numpy as np |
|
|
| from tensorflow.keras.layers import TextVectorization |
| import gradio as gr |
| from tensorflow.keras.layers import TextVectorization |
|
|
| modelbaru = tf.keras.models.load_model('toxicity.h5') |
|
|
| MAX_FEATURES = 200000 |
| data = pd.read_csv(os.path.join('jigsaw-toxic-comment-classification-challenge', 'train.csv', 'train.csv')) |
| x = data['comment_text'] |
| y = data[data.columns[2:]].values |
|
|
| vectorizer = TextVectorization(max_tokens=MAX_FEATURES, output_sequence_length=1800, output_mode='int') |
| vectorizer.adapt(x.values) |
| vectorizer('Yo Whats up')[:3] |
| vectorized_text = vectorizer(x.values) |
| vectorized_text |
|
|
| input_str = vectorizer('yo i fuckin hate you') |
| res = modelbaru.predict(np.expand_dims(input_str,0)) |
| res > 0.5 |
| data.columns[2:] |
| data.columns[2:-1] |
|
|
| def score_comment(comment): |
| vectorized_comment = vectorizer([comment]) |
| results = modelbaru.predict(vectorized_comment) |
|
|
| text = '' |
| for idx, col in enumerate(data.columns[2:-1]): |
| text += '{}: {}\n'.format(col, results[0][idx]>0.5) |
| |
| return text |
|
|
| interface = gr.Interface(fn=score_comment, inputs=gr.inputs.Textbox(lines=2, placeholder='Toxic Detector by: AezersX'), outputs='text') |
| interface.launch() |
