Taekyoon

Taekyoon

cls_model.load_weights(checkpoint_path)

results = cls_model.evaluate(test_data_sents, test_data_labels, batch_size=1024)
print("test loss, test acc: ", results)

Taekyoon

model_name = "tf2_gpt2_naver_movie"

earlystop_callback = EarlyStopping(monitor='val_accuracy', min_delta=0.0001, patience=2)

checkpoint_path = os.path.join(DATA_OUT_PATH, model_name, 'weights.h5')
checkpoint_dir = os.path.dirname(checkpoint_path)

if os.path.exists(checkpoint_dir):
    print("{} -- Folder already exists \n".format(checkpoint_dir))
else:

Taekyoon

class TFBertClassifier(TFBertPreTrainedModel):
    def __init__(self, dir_path, num_class=2):
        super().__init__(config)
        
        self.bert = TFBertModel(name="bert").from_pretrained('bert-base-multilingual-cased', cache_dir=dir_path)
        self.num_class = num_class
        self.dropout = tf.keras.layers.Dropout(self.bert.config.hidden_dropout_prob)
        self.classifier = tf.keras.layers.Dense(self.num_class, 
                                                kernel_initializer=tf.keras.initializers.TruncatedNormal(config.initializer_range), 

Taekyoon

def bert_tokenizer_v2(sent1, sent2, MAX_LEN):
    
    # For Two setenece input
    
    encoded_dict = tokenizer.encode_plus(
        text = sent1,
        text_pair = sent2,
        add_special_tokens = True, # Add '[CLS]' and '[SEP]'
        max_length = MAX_LEN,           # Pad & truncate all sentences.
        pad_to_max_length = True,

Taekyoon

class SquadExample:
    def __init__(self, question, context, start_char_idx, answer_text, all_answers):
        self.question = " ".join(str(question).split())
        self.context = " ".join(str(context).split())
        self.answer_text = " ".join(str(answer_text).split())
        self.start_char_idx = start_char_idx
        self.all_answers = all_answers
        self.skip = False
        
    def get_answer_position(self, context, tokenized_context, answer, start_char_idx):

Taekyoon

import parselmouth
import time
import concurrent.futures

NO_PITCH_VALUE = -999
NO_INTENSITY_VALUE = -999
MIN_PITCH = 75.0


def extract_pitch(path, window_size=0.01):

Taekyoon

import torch
import math
import time


SLEEP_TIME = 0.1

class MyMapDataset(torch.utils.data.Dataset):
    def __init__(self, size):
        self.dataset = [i for i in range(size)]

Taekyoon

def base_to_char_level_bio_tags(sent, tag, empty_tag='O'):
    char_level_tags = list()
    sent_list, tag_list = sent.split(), tag.split()

    if len(sent_list) != len(tag_list):
        raise ValueError()

    for i, (s, t) in enumerate(zip(sent_list, tag_list)):
        char_level_tags += [t for _ in range(len(s))]

Taekyoon

import copy
import json
from typing import List
from collections import Counter
from pathlib import Path


class Vocabulary(object):
    def __init__(self,
                 max_size=None,