File size: 12,198 Bytes
b321543 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 | import sentencepiece as spm
import os, json, numpy as np, tensorflow as tf
from tensorflow.keras import layers, Model
import requests
from tensorflow.keras import mixed_precision
import glob
tf.get_logger().setLevel("ERROR")
SEED = 42
tf.random.set_seed(SEED)
np.random.seed(SEED)
try:
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu="local")
tf.tpu.experimental.initialize_tpu_system(resolver)
strategy = tf.distribute.TPUStrategy(resolver)
print("✅ TPU 초기화 완료")
on_tpu = True
except Exception as e:
print("⚠️ TPU 미사용, GPU/CPU로 진행:", e)
strategy = tf.distribute.get_strategy()
on_tpu = False
policy_name = "mixed_bfloat16" if on_tpu else "float32"
policy = mixed_precision.Policy(policy_name)
mixed_precision.set_global_policy(policy)
print(f"✅ Mixed precision: {policy_name}")
def download_file(url, save_path):
if not os.path.exists(save_path):
r = requests.get(url, stream=True)
r.raise_for_status()
with open(save_path, "wb") as f:
for chunk in r.iter_content(8192*2):
f.write(chunk)
print(f"✅ {save_path} 저장됨")
TOKENIZER_PATH = "tokenizer.model"
download_file("https://huggingface.co/datasets/OpenLab-NLP/tiny-corpus/resolve/main/tokenizer.model?download=true", TOKENIZER_PATH)
DATA_DIR = "/kaggle/input/lm-pretrain"
FILE_PATTERN = os.path.join(DATA_DIR, "tokenized_variable_part_*.txt")
sp = spm.SentencePieceProcessor(TOKENIZER_PATH)
pad_id = sp.piece_to_id("<pad>") if sp.piece_to_id("<pad>") != -1 else 0
end_id = sp.piece_to_id("[EOS]")
vocab_size = sp.get_piece_size()
max_len = 512
batch_size = 768
import random
def prepare_packed_dataset(file_pattern, max_len, batch_size):
file_list = tf.io.gfile.glob(file_pattern)
# [수정] 파일 경로 리스트 자체를 무작위로 섞습니다.
# 이렇게 하면 매 에포크마다 파일을 읽는 순서가 달라집니다.
random.shuffle(file_list)
print(f"🔄 파일 로드 순서 섞기 완료 (첫 번째 파일: {file_list[0]})")
dataset = tf.data.TextLineDataset(file_list)
def parse_tokens(line):
return tf.strings.to_number(tf.strings.split(line), tf.int32)
dataset = dataset.map(parse_tokens, num_parallel_calls=tf.data.AUTOTUNE)
dataset = dataset.unbatch()
dataset = dataset.batch(max_len + 1, drop_remainder=True)
def split_input_target(chunk):
return chunk[:-1], chunk[1:]
dataset = dataset.map(split_input_target, num_parallel_calls=tf.data.AUTOTUNE)
dataset = dataset.shuffle(20000)
dataset = dataset.batch(batch_size, drop_remainder=True)
return dataset.prefetch(tf.data.AUTOTUNE)
with strategy.scope():
dataset = prepare_packed_dataset(FILE_PATTERN, max_len, batch_size)
dist_dataset = strategy.experimental_distribute_dataset(dataset)
print("✅ 데이터 패킹 및 TPU 분산 파이프라인 준비 완료")
class TimeMix(layers.Layer):
def __init__(self, d_model, layer_id, n_layers):
super().__init__()
self.d_model = d_model
ratio = (layer_id / (n_layers - 1)) if n_layers > 1 else 0.5
decay_speed = np.arange(d_model)
self.time_decay = tf.Variable(
-5 + 8 * (decay_speed / (d_model - 1)) ** (0.7 + 1.3 * ratio),
dtype=tf.float32, name="time_decay"
)
self.time_first = tf.Variable(
np.ones(d_model) * np.log(0.3),
dtype=tf.float32, name="time_first"
)
self.time_mix_k = tf.Variable(1 - (ratio ** 0.5), dtype=tf.float32)
self.time_mix_v = tf.Variable(1 - (ratio ** 0.5), dtype=tf.float32)
self.time_mix_r = tf.Variable(1 - (ratio ** 0.2), dtype=tf.float32)
self.key = layers.Dense(d_model, use_bias=False)
self.value = layers.Dense(d_model, use_bias=False)
self.receptance = layers.Dense(d_model, use_bias=False)
self.output_projection = layers.Dense(d_model, use_bias=False)
def call(self, x, training=False):
t_type = x.dtype
tm_k = tf.cast(self.time_mix_k, t_type)
tm_v = tf.cast(self.time_mix_v, t_type)
tm_r = tf.cast(self.time_mix_r, t_type)
xx = tf.pad(x[:, :-1, :], [[0, 0], [1, 0], [0, 0]])
k = self.key(x * tm_k + xx * (1 - tm_k))
v = self.value(x * tm_v + xx * (1 - tm_v))
r = self.receptance(x * tm_r + xx * (1 - tm_r))
wkv = self.parallel_wkv(k, v)
return self.output_projection(tf.nn.sigmoid(r) * wkv)
def parallel_wkv(self, k, v):
t_type = k.dtype
w = tf.cast(tf.exp(self.time_decay), t_type)
u = tf.cast(self.time_first, t_type)
t = tf.shape(k)[1]
t_index = tf.cast(tf.range(t), t_type)[:, None]
s = k - (t_index * w)
kv = tf.exp(s) * v
k_exp = tf.exp(s)
num = tf.cumsum(kv, axis=1) - kv + tf.exp(s + u) * v
den = tf.cumsum(k_exp, axis=1) - k_exp + tf.exp(s + u)
return num / (den + 1e-8)
class ChannelMix(layers.Layer):
def __init__(self, d_model, layer_id, n_layers, num_experts=8, top_k=2):
super().__init__()
self.d_model = d_model
self.num_experts = num_experts
self.top_k = top_k
ratio = (layer_id / (n_layers - 1)) if n_layers > 1 else 0.5
self.time_mix_k = tf.Variable(1 - (ratio ** 0.5), dtype=tf.float32)
self.time_mix_r = tf.Variable(1 - (ratio ** 0.5), dtype=tf.float32)
# Gate: 전문가 선택을 위한 로직
self.gate = layers.Dense(num_experts, use_bias=False)
# Experts 가중치
self.key_weight = self.add_weight(
name="expert_key",
shape=(num_experts, d_model, int(d_model * 4)),
initializer="glorot_uniform"
)
self.value_weight = self.add_weight(
name="expert_value",
shape=(num_experts, int(d_model * 4), d_model),
initializer="glorot_uniform"
)
self.receptance = layers.Dense(d_model, use_bias=False)
def call(self, x, training=False):
t_type = x.dtype
b, t, d = tf.shape(x)[0], tf.shape(x)[1], tf.shape(x)[2]
xx = tf.pad(x[:, :-1, :], [[0, 0], [1, 0], [0, 0]])
k_in = x * tf.cast(self.time_mix_k, t_type) + xx * (1 - tf.cast(self.time_mix_k, t_type))
# 1. Gate Logits 및 Top-K 선택
gate_logits = self.gate(k_in) # (B, T, num_experts)
# Top-K 전문가와 가중치 추출
raw_weights, indices = tf.math.top_k(gate_logits, k=self.top_k)
gate_weights = tf.nn.softmax(tf.cast(raw_weights, tf.float32))
gate_weights = tf.cast(gate_weights, t_type) # (B, T, top_k)
# 2. Sparse 연산을 위한 Mask 생성 (수치적 안정성 및 로드 밸런싱용)
# 실제로는 모든 전문가를 다 계산한 뒤 Masking하는 방식이 TPU MXU 활용에 유리할 수 있음
# 여기서는 einsum을 활용하되 선택된 전문가의 영향력만 남김
masks = tf.one_hot(indices, depth=self.num_experts, dtype=t_type) # (B, T, top_k, num_experts)
final_mask = tf.reduce_sum(masks * tf.expand_dims(gate_weights, -1), axis=2) # (B, T, num_experts)
# 3. Auxiliary Loss (전문가 균등 분배)
if training:
# Load Balancing Loss: gate_logits의 확률 분포가 균등하도록 유도
prob_dist = tf.nn.softmax(tf.cast(gate_logits, tf.float32), axis=-1)
importance = tf.reduce_sum(prob_dist, axis=[0, 1])
load = tf.reduce_sum(tf.cast(final_mask > 0, tf.float32), axis=[0, 1])
aux_loss = tf.reduce_sum(importance * load) * (self.num_experts / (tf.cast(b * t, tf.float32) ** 2))
self.add_loss(0.01 * aux_loss)
# 4. 전문가 연산 (Einsum 활용)
# 모든 전문가를 계산하되, mask를 통해 필요한 정보만 남김 (Sparse Approximation)
k_experts = tf.einsum('btd,edh->bteh', k_in, self.key_weight)
k_experts = tf.square(tf.nn.relu(k_experts))
v_experts = tf.einsum('bteh,ehd->bted', k_experts, self.value_weight) # (B, T, E, D)
# 5. 가중 합산 (최종 선택된 전문가의 결과만 결합)
kv = tf.reduce_sum(v_experts * tf.expand_dims(final_mask, -1), axis=2)
# Receptance (Gate) 연산
r_in = x * tf.cast(self.time_mix_r, t_type) + xx * (1 - tf.cast(self.time_mix_r, t_type))
r = self.receptance(r_in)
return tf.nn.sigmoid(r) * kv
class Block(layers.Layer):
def __init__(self, d_model, layer_id, n_layers):
super().__init__()
self.ln = layers.LayerNormalization(epsilon=1e-5)
self.time_mix = TimeMix(d_model, layer_id, n_layers)
self.channel_mix = ChannelMix(d_model, layer_id, n_layers)
def call(self, x, training=False):
ln_x = self.ln(x)
return x + self.time_mix(ln_x, training=training) + self.channel_mix(ln_x)
class Head(tf.keras.Model):
def __init__(self, vocab_size):
super().__init__()
self.lm_head = layers.Dense(vocab_size, use_bias=False, name="output_head", dtype=policy)
def call(self, x, training=False):
logits = self.lm_head(x)
return tf.cast(logits, tf.float32)
class LM(tf.keras.Model):
def __init__(self, d_model, n_layers, dropout_rate=0.1):
super().__init__()
self.token_embedding = layers.Embedding(vocab_size, d_model)
self.blocks = [Block(d_model, i, n_layers) for i in range(n_layers)]
self.ln_f = layers.LayerNormalization(epsilon=1e-5, dtype=tf.float32)
def call(self, x, training=False):
x = self.token_embedding(x)
for block in self.blocks:
x = block(x, training=training)
x = tf.cast(x, tf.float32)
x = self.ln_f(x)
return x
def smoothed_loss_keras(y_true, y_pred, eps=0.1):
y_true = tf.cast(y_true, tf.int32)
mask = tf.cast(tf.not_equal(y_true, pad_id), tf.float32)
vocab = tf.shape(y_pred)[-1]
y_true_oh = tf.one_hot(y_true, depth=vocab, dtype=tf.float32)
y_true_ls = (1.0 - eps) * y_true_oh + eps / tf.cast(vocab, tf.float32)
log_probs = tf.nn.log_softmax(y_pred, axis=-1)
per_tok = -tf.reduce_sum(y_true_ls * log_probs, axis=-1)
return tf.reduce_sum(per_tok * mask) / (tf.reduce_sum(mask) + 1e-8)
with strategy.scope():
blocklm = LM(d_model=512, n_layers=16)
head = Head(vocab_size=vocab_size)
inputs = layers.Input(shape=(max_len,), dtype=tf.int32)
x = blocklm(inputs)
outputs = head(x)
model = tf.keras.Model(inputs=inputs, outputs=outputs)
optimizer = tf.keras.optimizers.AdamW(learning_rate=1e-4, weight_decay=0.01)
model.compile(optimizer=optimizer, loss=smoothed_loss_keras)
dummy_input = np.zeros((1, max_len), dtype=np.int32)
model(dummy_input)
model.summary()
def get_training_stats(file_pattern, max_len, batch_size):
total_tokens = 0
files = glob.glob(file_pattern)
for f in files:
with open(f, 'r') as file:
for line in file:
total_tokens += len(line.split())
total_chunks = total_tokens // (max_len + 1)
steps_per_epoch = total_chunks // batch_size
return total_tokens, total_chunks, steps_per_epoch
#total_tokens, total_chunks, steps_per_epoch = get_training_stats(FILE_PATTERN, max_len, batch_size)
#print(f"✅ 총 토큰 수: {total_tokens}")
#print(f"✅ 생성된 총 덩어리(Chunk) 수: {total_chunks}")
#print(f"✅ steps_per_epoch: {steps_per_epoch}")
model.fit(dist_dataset, epochs=1, steps_per_epoch=14582)
blocklm.save_weights("blocklm.weights.h5")
head.save_weights("head.weights.h5")
print("저장됨")
|