TensorRT-LLMs/tensorrt_llm/models/gemma/weight.py
2024-04-01 16:39:43 +08:00

854 lines
38 KiB
Python

# SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import configparser
import os
import time
from pathlib import Path
from typing import Union
import numpy as np
import torch
from ..._utils import (numpy_to_torch, pad_vocab_size, str_dtype_to_torch,
torch_to_numpy)
from ...logger import logger
from ...mapping import Mapping
from ...quantization import QuantMode
def gen_suffix(rank, use_smooth_quant, quant_per_channel):
suffix = f"{rank}.bin"
if use_smooth_quant:
sq_prefix = "int8."
if quant_per_channel:
sq_prefix += "col."
suffix = sq_prefix + suffix
return suffix
def extract_layer_idx(name):
ss = name.split('.')
for s in ss:
if s.isdigit():
return s
return None
def split(v: Union[np.ndarray, torch.Tensor],
tp_size: int,
tp_rank: int,
dim=0):
if tp_size == 1:
return v
assert len(v.shape) > 1 or dim == 0
if isinstance(v, np.ndarray):
return np.ascontiguousarray(
np.split(v, tp_size, axis=dim)[tp_rank].copy())
else:
assert v.shape[dim] % tp_size == 0, \
'Unable to split: shape={v.shape} (dim={dim}) tp_size={tp_size}.'
split_size = v.shape[dim] // tp_size
return v.split(split_size, dim=dim)[tp_rank].clone().detach()
def dup_kv_weight(v, num_head, tp_size):
assert tp_size % num_head == 0
reps = tp_size // num_head
head_size = v.shape[0] // num_head
v = v.reshape(num_head, head_size,
-1)[:, None, :, :].expand(num_head, reps, head_size,
v.shape[1])
return v.reshape(num_head * reps * head_size, -1).clone().detach()
def parse_bin_config(ini_file):
model_config = configparser.ConfigParser()
model_config.read(ini_file)
n_embd = model_config.getint('gemma', 'hidden_size')
n_head = model_config.getint('gemma', 'num_attention_heads')
n_head_size = model_config.getint('gemma',
'head_size',
fallback=n_embd // n_head)
n_layer = model_config.getint('gemma', 'num_hidden_layers')
n_positions = model_config.getint('gemma', 'max_position_embeddings')
vocab_size = model_config.getint('gemma', 'vocab_size')
hidden_act = model_config.get('gemma', 'hidden_act')
inter_size = model_config.getint('gemma',
'intermediate_size',
fallback=None)
n_kv_head = model_config.getint('gemma',
'num_key_value_heads',
fallback=None)
if inter_size is None:
inter_size = 4 * n_embd
return n_embd, n_head, n_layer, n_positions, vocab_size, hidden_act, inter_size, n_kv_head, n_head_size
def load_from_binary(tensorrt_llm_gemma,
dir_path,
mapping=Mapping(),
fp16=False,
multi_query_mode=False):
logger.info('Loading weights from binary...')
tik = time.time()
quant_mode = getattr(tensorrt_llm_gemma, 'quant_mode', QuantMode(0))
n_embd, n_head, n_layer, n_positions, vocab_size, hidden_act, inter_size, n_kv_head, n_head_size = parse_bin_config(
Path(dir_path) / 'config.ini')
np_dtype = np.float16 if fp16 else np.float32
def fromfile(dir_path, name, shape=None, dtype=None):
dtype = np_dtype if dtype is None else dtype
p = dir_path + '/' + name
if Path(p).exists():
t = np.fromfile(p, dtype=dtype)
if shape is not None:
t = t.reshape(shape)
return t
return None
def set_smoothquant_scale_factors(module,
pre_scale_weight,
dir_path,
basename,
shape,
per_tok_dyn,
per_channel,
is_qkv=False,
rank=None):
suffix = "bin"
if per_channel:
if rank is not None:
suffix = f"{rank}." + suffix
suffix = "col." + suffix
col_shape = shape if (per_channel or is_qkv) else [1, 1]
if per_tok_dyn:
if pre_scale_weight is not None:
pre_scale_weight.value = np.array([1.0], dtype=np.float32)
if is_qkv and not per_channel:
t = fromfile(dir_path,
f"{basename}scale_w_quant_orig.{rank}.{suffix}",
col_shape, np.float32)
else:
t = fromfile(dir_path, f"{basename}scale_w_quant_orig.{suffix}",
col_shape, np.float32)
module.per_channel_scale.value = t
else:
t = fromfile(dir_path, f"{basename}scale_x_orig_quant.bin", [1],
np.float32)
pre_scale_weight.value = t
if is_qkv:
t = fromfile(dir_path,
f"{basename}scale_y_accum_quant.{rank}.{suffix}",
col_shape, np.float32)
else:
t = fromfile(dir_path,
f"{basename}scale_y_accum_quant.{suffix}",
col_shape, np.float32)
module.per_channel_scale.value = t
t = fromfile(dir_path, f"{basename}scale_y_quant_orig.bin", [1, 1],
np.float32)
module.act_scale.value = t
def set_smoother(module, dir_path, base_name, shape, rank):
suffix = f"{rank}.bin"
t = fromfile(dir_path, f"{base_name}.smoother.{suffix}", shape,
np.float32)
module.smoother.value = t
# Determine the quantization mode.
quant_mode = getattr(tensorrt_llm_gemma, "quant_mode", QuantMode(0))
if quant_mode.is_int8_weight_only():
plugin_weight_only_quant_type = torch.int8
elif quant_mode.is_int4_weight_only():
plugin_weight_only_quant_type = torch.quint4x2
# Do we use SmoothQuant?
use_smooth_quant = quant_mode.has_act_and_weight_quant()
# Do we use quantization per token?
quant_per_token_dyn = quant_mode.has_per_token_dynamic_scaling()
# Do we use quantization per channel?
quant_per_channel = quant_mode.has_per_channel_scaling()
# Do we use INT4/INT8 weight-only?
use_weight_only = quant_mode.is_weight_only()
# Int8 KV cache
use_int8_kv_cache = quant_mode.has_int8_kv_cache()
# Debug
suffix = gen_suffix(mapping.tp_rank, use_smooth_quant, quant_per_channel)
# The type of weights.
w_type = np_dtype if not use_smooth_quant else np.int8
if mapping.is_first_pp_rank():
tensorrt_llm_gemma.vocab_embedding.weight.value = (fromfile(
dir_path, 'vocab_embedding.weight.bin', [vocab_size, n_embd]))
if mapping.is_last_pp_rank():
tensorrt_llm_gemma.ln_f.weight.value = (fromfile(
dir_path, 'ln_f.weight.bin'))
# share input embedding
lm_head_weight = fromfile(dir_path, 'lm_head.weight.bin',
[vocab_size, n_embd])
if vocab_size % mapping.tp_size != 0:
# padding
vocab_size_padded = tensorrt_llm_gemma.lm_head.out_features * mapping.tp_size
pad_width = vocab_size_padded - vocab_size
lm_head_weight = np.pad(lm_head_weight, ((0, pad_width), (0, 0)),
'constant',
constant_values=0)
if mapping.is_last_pp_rank():
tensorrt_llm_gemma.lm_head.weight.value = np.ascontiguousarray(
split(lm_head_weight, mapping.tp_size, mapping.tp_rank))
num_hidden_layers = tensorrt_llm_gemma.num_layers
layers_range = mapping.pp_layers(num_hidden_layers)
# This code does not support the case where the number of ranks is greater than the number of K/V heads for GQA.
assert (n_kv_head % mapping.tp_size == 0) or (n_kv_head == 1)
# Compute the number of K/V heads per rank. It's 1 for MQA.
kv_heads_per_rank = min(1, n_kv_head // mapping.tp_size)
# The N-dimension for each rank of the QKV matrix is number of columns for Q + 2 * number of columns for K/V.
if multi_query_mode:
c_attn_out_dim = n_head * n_head_size // mapping.tp_size + 2 * kv_heads_per_rank * n_head_size
else:
c_attn_out_dim = 3 * (n_head * n_head_size) // mapping.tp_size
for i in layers_range:
idx = i - layers_range[0]
tensorrt_llm_gemma.layers[idx].input_layernorm.weight.value = (fromfile(
dir_path, 'model.layers.' + str(i) + '.input_layernorm.weight.bin'))
t = fromfile(
dir_path, 'model.layers.' + str(i) +
'.attention.query_key_value.weight.' + suffix,
[n_embd, c_attn_out_dim], w_type)
if t is not None:
dst = tensorrt_llm_gemma.layers[idx].attention.qkv.weight
if use_smooth_quant:
dst.value = np.ascontiguousarray(np.transpose(t, [1, 0]))
set_smoothquant_scale_factors(
tensorrt_llm_gemma.layers[idx].attention.qkv,
tensorrt_llm_gemma.layers[idx].input_layernorm.scale_to_int,
dir_path,
'model.layers.' + str(i) + '.attention.query_key_value.',
[1, c_attn_out_dim],
quant_per_token_dyn,
quant_per_channel,
rank=mapping.tp_rank,
is_qkv=True)
elif use_weight_only:
processed_torch_weights, torch_weight_scales = torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
torch.tensor(t), plugin_weight_only_quant_type)
dst.value = processed_torch_weights.numpy()
scales = tensorrt_llm_gemma.layers[
idx].attention.qkv.per_channel_scale
scales.value = torch_weight_scales.numpy()
else:
dst.value = np.ascontiguousarray(np.transpose(t, [1, 0]))
dst = tensorrt_llm_gemma.layers[idx].attention.dense.weight
t = fromfile(
dir_path,
'model.layers.' + str(i) + '.attention.dense.weight.' + suffix,
[(n_head * n_head_size) // mapping.tp_size, n_embd], w_type)
if use_smooth_quant:
dst.value = np.ascontiguousarray(np.transpose(t, [1, 0]))
dense_scale = getattr(tensorrt_llm_gemma.layers[idx].attention,
"quantization_scaling_factor", None)
set_smoothquant_scale_factors(
tensorrt_llm_gemma.layers[idx].attention.dense, dense_scale,
dir_path, 'model.layers.' + str(i) + '.attention.dense.',
[1, n_embd], quant_per_token_dyn, quant_per_channel)
set_smoother(tensorrt_llm_gemma.layers[idx].attention.dense,
dir_path,
'model.layers.' + str(i) + '.attention.dense',
[1, n_embd // mapping.tp_size], mapping.tp_rank)
elif use_weight_only:
processed_torch_weights, torch_weight_scales = torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
torch.tensor(t), plugin_weight_only_quant_type)
dst.value = processed_torch_weights.numpy()
scales = tensorrt_llm_gemma.layers[
idx].attention.dense.per_channel_scale
scales.value = torch_weight_scales.numpy()
else:
dst.value = np.ascontiguousarray(np.transpose(t, [1, 0]))
dst = tensorrt_llm_gemma.layers[idx].post_layernorm.weight
dst.value = fromfile(
dir_path, 'model.layers.' + str(i) + '.post_layernorm.weight.bin')
t = fromfile(dir_path,
'model.layers.' + str(i) + '.mlp.fc.weight.' + suffix,
[n_embd, inter_size // mapping.tp_size], w_type)
if use_smooth_quant:
tensorrt_llm_gemma.layers[
idx].mlp.fc.weight.value = np.ascontiguousarray(
np.transpose(t, [1, 0]))
set_smoothquant_scale_factors(
tensorrt_llm_gemma.layers[idx].mlp.fc,
tensorrt_llm_gemma.layers[idx].post_layernorm.scale_to_int,
dir_path,
'model.layers.' + str(i) + '.mlp.fc.',
[1, inter_size // mapping.tp_size],
quant_per_token_dyn,
quant_per_channel,
rank=mapping.tp_rank)
elif use_weight_only:
dst = tensorrt_llm_gemma.layers[idx].mlp.fc.weight
processed_torch_weights, torch_weight_scales = torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
torch.tensor(t), plugin_weight_only_quant_type)
dst.value = processed_torch_weights.numpy()
scales = tensorrt_llm_gemma.layers[idx].mlp.fc.per_channel_scale
scales.value = torch_weight_scales.numpy()
else:
tensorrt_llm_gemma.layers[
idx].mlp.fc.weight.value = np.ascontiguousarray(
np.transpose(t, [1, 0]))
t = fromfile(dir_path,
'model.layers.' + str(i) + '.mlp.gate.weight.' + suffix,
[n_embd, inter_size // mapping.tp_size], w_type)
if use_smooth_quant:
tensorrt_llm_gemma.layers[
idx].mlp.gate.weight.value = np.ascontiguousarray(
np.transpose(t, [1, 0]))
set_smoothquant_scale_factors(
tensorrt_llm_gemma.layers[idx].mlp.gate,
tensorrt_llm_gemma.layers[idx].post_layernorm.scale_to_int,
dir_path,
'model.layers.' + str(i) + '.mlp.gate.',
[1, inter_size // mapping.tp_size],
quant_per_token_dyn,
quant_per_channel,
rank=mapping.tp_rank)
elif use_weight_only:
dst = tensorrt_llm_gemma.layers[idx].mlp.gate.weight
processed_torch_weights, torch_weight_scales = torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
torch.tensor(t), plugin_weight_only_quant_type)
dst.value = processed_torch_weights.numpy()
scales = tensorrt_llm_gemma.layers[idx].mlp.gate.per_channel_scale
scales.value = torch_weight_scales.numpy()
else:
tensorrt_llm_gemma.layers[
idx].mlp.gate.weight.value = np.ascontiguousarray(
np.transpose(t, [1, 0]))
t = fromfile(dir_path,
'model.layers.' + str(i) + '.mlp.proj.weight.' + suffix,
[inter_size // mapping.tp_size, n_embd], w_type)
if use_smooth_quant:
tensorrt_llm_gemma.layers[
idx].mlp.proj.weight.value = np.ascontiguousarray(
np.transpose(t, [1, 0]))
proj_scale = getattr(tensorrt_llm_gemma.layers[idx].mlp,
"quantization_scaling_factor", None)
set_smoothquant_scale_factors(
tensorrt_llm_gemma.layers[idx].mlp.proj, proj_scale, dir_path,
'model.layers.' + str(i) + '.mlp.proj.', [1, n_embd],
quant_per_token_dyn, quant_per_channel)
set_smoother(tensorrt_llm_gemma.layers[idx].mlp.proj, dir_path,
'model.layers.' + str(i) + '.mlp.proj',
[1, inter_size // mapping.tp_size], mapping.tp_rank)
elif use_weight_only:
dst = tensorrt_llm_gemma.layers[idx].mlp.proj.weight
processed_torch_weights, torch_weight_scales = torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
torch.tensor(t), plugin_weight_only_quant_type)
dst.value = processed_torch_weights.numpy()
scales = tensorrt_llm_gemma.layers[idx].mlp.proj.per_channel_scale
scales.value = torch_weight_scales.numpy()
else:
tensorrt_llm_gemma.layers[idx].mlp.proj.weight.value = (
np.ascontiguousarray(np.transpose(t, [1, 0])))
if use_int8_kv_cache:
t = fromfile(
dir_path, 'model.layers.' + str(i) +
'.attention.query_key_value.scale_y_quant_orig.bin', [1],
np.float32)
tensorrt_llm_gemma.layers[
idx].attention.kv_cache_scaling_factor.value = t
tok = time.time()
t = time.strftime('%H:%M:%S', time.gmtime(tok - tik))
logger.info(f'Weights loaded. Total time: {t}')
def load_from_hf_gemma(tensorrt_llm_llama: 'GemmaForCausalLM',
hf_gemma,
mapping=Mapping(),
dtype='float32',
use_gemm_woq_plugin=True):
logger.info('Loading weights from HF Gemma...')
tik = time.time()
quant_mode = getattr(tensorrt_llm_llama, 'quant_mode', QuantMode(0))
if quant_mode.is_int8_weight_only():
plugin_weight_only_quant_type = torch.int8
elif quant_mode.is_int4_weight_only():
plugin_weight_only_quant_type = torch.quint4x2
use_weight_only = quant_mode.is_weight_only()
num_kv_heads = tensorrt_llm_llama.config.num_key_value_heads
mha_mode = (num_kv_heads == tensorrt_llm_llama.config.num_attention_heads)
model_params = dict(hf_gemma.named_parameters())
# concatenate, duplicate and reshape q, k, v -> qkv
for l in range(hf_gemma.config.num_hidden_layers):
prefix = f'model.layers.{l}.self_attn.'
q_weight = model_params[prefix + 'q_proj.weight']
k_weight = model_params[prefix + 'k_proj.weight']
v_weight = model_params[prefix + 'v_proj.weight']
if not mha_mode:
head_size = tensorrt_llm_llama.config.hidden_size // tensorrt_llm_llama.config.num_attention_heads
if num_kv_heads < mapping.tp_size:
# duplicate the KV heads up to tensor_parallel
k_weight = dup_kv_weight(k_weight, num_kv_heads,
mapping.tp_size)
v_weight = dup_kv_weight(v_weight, num_kv_heads,
mapping.tp_size)
assert (k_weight.shape[0] % (mapping.tp_size * head_size)) == 0
assert (v_weight.shape[0] % (mapping.tp_size * head_size)) == 0
qkv_weight = [q_weight, k_weight, v_weight]
else:
qkv_weight = torch.cat([q_weight, k_weight, v_weight], dim=0)
model_params[prefix + 'qkv_proj.weight'] = qkv_weight
torch_dtype = str_dtype_to_torch(dtype)
layers_range = mapping.pp_layers(hf_gemma.config.num_hidden_layers)
vocab_size = hf_gemma.config.vocab_size
weights = {}
for k, v in model_params.items():
t_dtype = torch_dtype if "block_sparse_moe.gate" not in k else torch.float32
if isinstance(v, list):
v = [torch_to_numpy(vv.to(t_dtype).detach().cpu()) for vv in v]
else:
v = torch_to_numpy(v.to(t_dtype).detach().cpu())
if 'model.embed_tokens.weight' in k:
if hf_gemma.config.tie_word_embeddings:
# lm_head.weight has the same weights as embedding
if mapping.is_last_pp_rank():
if vocab_size % mapping.tp_size != 0:
# padding
vocab_size_padded = pad_vocab_size(
vocab_size, mapping.tp_size)
pad_width = vocab_size_padded - vocab_size
v = torch.from_numpy(
np.pad(v.detach().cpu().numpy(),
((0, pad_width), (0, 0)),
'constant',
constant_values=0))
weights['lm_head.weight'] = split(v, mapping.tp_size,
mapping.tp_rank)
if tensorrt_llm_llama.config.use_parallel_embedding:
v = split(v, mapping.tp_size, mapping.tp_rank,
tensorrt_llm_llama.config.embedding_sharding_dim)
if mapping.is_first_pp_rank():
weights['transformer.vocab_embedding.weight'] = torch_to_numpy(
numpy_to_torch(v).to(torch.float32) *
np.sqrt(tensorrt_llm_llama.config.hidden_size))
elif 'model.norm.weight' in k:
if mapping.is_last_pp_rank():
weights['transformer.ln_f.weight'] = torch_to_numpy(
numpy_to_torch(v) + 1.0)
elif 'lm_head.weight' in k:
if mapping.is_last_pp_rank():
if vocab_size % mapping.tp_size != 0:
# padding
vocab_size_padded = tensorrt_llm_llama.lm_head.out_features * mapping.tp_size
pad_width = vocab_size_padded - vocab_size
v = np.pad(v, ((0, pad_width), (0, 0)),
'constant',
constant_values=0)
weights['lm_head.weight'] = split(v, mapping.tp_size,
mapping.tp_rank)
else:
layer_idx = extract_layer_idx(k)
if layer_idx is None or int(layer_idx) not in layers_range:
continue
idx = int(layer_idx) - layers_range[0]
if 'input_layernorm.weight' in k:
weights['transformer.layers.{}.input_layernorm.weight'.format(
idx)] = torch_to_numpy(numpy_to_torch(v) + 1.0)
elif 'post_attention_layernorm.weight' in k:
weights['transformer.layers.{}.post_layernorm.weight'.format(
idx)] = torch_to_numpy(numpy_to_torch(v) + 1.0)
elif 'self_attn.qkv_proj.weight' in k:
if not mha_mode:
assert isinstance(v, list) and len(v) == 3
wq = split(v[0], mapping.tp_size, mapping.tp_rank)
wk = split(v[1], mapping.tp_size, mapping.tp_rank)
wv = split(v[2], mapping.tp_size, mapping.tp_rank)
split_v = np.concatenate((wq, wk, wv))
else:
q_emb = v.shape[0] // 3
model_emb = v.shape[1]
v = v.reshape(3, q_emb, model_emb)
split_v = split(v, mapping.tp_size, mapping.tp_rank, dim=1)
split_v = split_v.reshape(3 * (q_emb // mapping.tp_size),
model_emb)
if use_weight_only:
v = np.ascontiguousarray(split_v.transpose())
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
if not use_gemm_woq_plugin:
weights['transformer.layers.{}.attention.qkv.weight'.
format(idx)] = v
else:
weights['transformer.layers.{}.attention.qkv.weight'.
format(idx)] = processed_torch_weights
weights[
'transformer.layers.{}.attention.qkv.per_channel_scale'.
format(idx)] = torch_weight_scales
else:
weights['transformer.layers.{}.attention.qkv.weight'.format(
idx)] = split_v
elif 'self_attn.o_proj.weight' in k:
# dst = tensorrt_llm_llama.layers[idx].attention.dense.weight
split_v = split(v, mapping.tp_size, mapping.tp_rank, dim=1)
if use_weight_only:
v = np.ascontiguousarray(split_v.transpose())
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
if not use_gemm_woq_plugin:
weights['transformer.layers.{}.attention.dense.weight'.
format(idx)] = v
else:
weights['transformer.layers.{}.attention.dense.weight'.
format(idx)] = processed_torch_weights
weights[
'transformer.layers.{}.attention.dense.per_channel_scale'
.format(idx)] = torch_weight_scales
else:
weights['transformer.layers.{}.attention.dense.weight'.
format(idx)] = split_v
elif 'mlp.up_proj.weight' in k:
split_v = split(v, mapping.tp_size, mapping.tp_rank, dim=0)
if use_weight_only:
v = np.ascontiguousarray(split_v.transpose())
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
if not use_gemm_woq_plugin:
weights['transformer.layers.{}.mlp.gate.weight'.format(
idx)] = v
else:
weights['transformer.layers.{}.mlp.gate.weight'.format(
idx)] = processed_torch_weights
weights['transformer.layers.{}.mlp.gate.per_channel_scale'.
format(idx)] = torch_weight_scales
else:
weights['transformer.layers.{}.mlp.gate.weight'.format(
idx)] = split_v
elif 'mlp.down_proj.weight' in k:
# dst = tensorrt_llm_llama.layers[idx].mlp.proj.weight
split_v = split(v, mapping.tp_size, mapping.tp_rank, dim=1)
if use_weight_only:
v = np.ascontiguousarray(split_v.transpose())
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
if not use_gemm_woq_plugin:
weights['transformer.layers.{}.mlp.proj.weight'.format(
idx)] = v
else:
weights['transformer.layers.{}.mlp.proj.weight'.format(
idx)] = processed_torch_weights
weights['transformer.layers.{}.mlp.proj.per_channel_scale'.
format(idx)] = torch_weight_scales
else:
weights['transformer.layers.{}.mlp.proj.weight'.format(
idx)] = split_v
elif 'mlp.gate_proj.weight' in k:
# dst = tensorrt_llm_llama.layers[idx].mlp.fc.weight
split_v = split(v, mapping.tp_size, mapping.tp_rank, dim=0)
if use_weight_only:
v = np.ascontiguousarray(split_v.transpose())
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
if not use_gemm_woq_plugin:
weights['transformer.layers.{}.mlp.fc.weight'.format(
idx)] = v
else:
weights['transformer.layers.{}.mlp.fc.weight'.format(
idx)] = processed_torch_weights
weights['transformer.layers.{}.mlp.fc.per_channel_scale'.
format(idx)] = torch_weight_scales
else:
# dst.value = np.ascontiguousarray(split_v)
weights['transformer.layers.{}.mlp.fc.weight'.format(
idx)] = split_v
elif 'experts.w2.weight' in k:
# Note: no need for splitting, it's already been done above
split_v = v
if use_weight_only:
v = np.ascontiguousarray(
np.transpose(split_v, axes=(0, 2, 1)))
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
weights['transformer.layers.{}.mlp.experts_weight_2'.format(
idx)] = processed_torch_weights
weights['transformer.layers.{}.mlp.experts_scale_2'.format(
idx)] = torch_weight_scales
else:
weights['transformer.layers.{}.mlp.experts_weight_2'.format(
idx)] = v
elif 'experts.w3w1.weight' in k:
# Note: no need for splitting, it's already been done above
split_v = v
if use_weight_only:
v = np.ascontiguousarray(
np.transpose(split_v, axes=(0, 2, 1)))
processed_torch_weights, torch_weight_scales = \
torch.ops.trtllm.symmetric_quantize_last_axis_of_batched_matrix(
numpy_to_torch(v), plugin_weight_only_quant_type)
weights['transformer.layers.{}.mlp.experts_weight_1'.format(
idx)] = processed_torch_weights
weights['transformer.layers.{}.mlp.experts_scale_1'.format(
idx)] = torch_weight_scales
else:
weights['transformer.layers.{}.mlp.experts_weight_1'.format(
idx)] = v
elif 'block_sparse_moe.gate' in k:
v = split(v, mapping.tp_size, mapping.tp_rank, dim=-1)
weights['transformer.layers.{}.mlp.router.weight'.format(
idx)] = v
tok = time.time()
t = time.strftime('%H:%M:%S', time.gmtime(tok - tik))
logger.info(f'Weights loaded. Total time: {t}')
return weights
def quantize_fp8_weights(weights, num_layers, mapping):
def get_scaling_factor(weight):
amax = weight.max()
scale = 448.0 / amax
return scale
layers_range = mapping.pp_layers(num_layers)
scaling_factors = {}
scaled_weights = {}
trt_llm_prefix = "transformer.layers"
for l in layers_range:
# attention.qkv.weight
for name in [
"attention.qkv", "attention.dense", "mlp.fc", "mlp.gate",
"mlp.proj"
]:
trt_llm_name = ".".join((trt_llm_prefix, str(l), name, "weight"))
scale_name = ".".join(
(trt_llm_prefix, str(l), name, "weights_scaling_factor"))
weight = weights[trt_llm_name]
dtype = weights[trt_llm_name].dtype
scale = get_scaling_factor(weight)
scaled_weights[trt_llm_name] = np.ascontiguousarray(
(weight * scale).astype(dtype))
scaling_factors[scale_name] = np.asarray([1 / scale
]).astype(np.float32)
return scaling_factors
def load_from_fp8_llama(quant_ckpt_path: str, num_layers: int, mapping: Mapping,
fp8_kv_cache: bool, weight_scales: dict):
"""
Get the fp8 scaling factors.
"""
fake_fp8_sf_dt = torch.float32
if quant_ckpt_path is not None and os.path.isfile(quant_ckpt_path):
fp8_llama = np.load(quant_ckpt_path)
else:
fp8_llama = None
logger.info(
f"There is not quantized checkpoint, use dummy fp8 scaling factors instead."
)
weights = {}
def get_fp8_llama(name):
if fp8_llama is not None:
return fp8_llama[name]
else:
return torch.tensor([1.0], dtype=fake_fp8_sf_dt).numpy()
layers_range = mapping.pp_layers(num_layers)
for l in layers_range:
prefix = f'_np:layers:{l}'
tllm_prex = f'transformer.layers.{l-layers_range[0]}'
weights[f'{tllm_prex}.attention.qkv.activation_scaling_factor'] = max(
get_fp8_llama(
f'{prefix}:attention:qkv:q:activation_scaling_factor'),
get_fp8_llama(
f'{prefix}:attention:qkv:k:activation_scaling_factor'),
get_fp8_llama(
f'{prefix}:attention:qkv:v:activation_scaling_factor'))
weights[f'{tllm_prex}.attention.qkv.weights_scaling_factor'] = max(
get_fp8_llama(f'{prefix}:attention:qkv:q:weights_scaling_factor'),
get_fp8_llama(f'{prefix}:attention:qkv:k:weights_scaling_factor'),
get_fp8_llama(f'{prefix}:attention:qkv:v:weights_scaling_factor'))
weights[
f'{tllm_prex}.attention.dense.activation_scaling_factor'] = get_fp8_llama(
f'{prefix}:attention:dense:activation_scaling_factor')
weights[
f'{tllm_prex}.attention.dense.weights_scaling_factor'] = get_fp8_llama(
f'{prefix}:attention:dense:weights_scaling_factor')
weights[
f'{tllm_prex}.mlp.fc.activation_scaling_factor'] = get_fp8_llama(
f'{prefix}:mlp:fc:activation_scaling_factor')
weights[f'{tllm_prex}.mlp.fc.weights_scaling_factor'] = get_fp8_llama(
f'{prefix}:mlp:fc:weights_scaling_factor')
weights[
f'{tllm_prex}.mlp.gate.activation_scaling_factor'] = get_fp8_llama(
f'{prefix}:mlp:gate:activation_scaling_factor')
weights[f'{tllm_prex}.mlp.gate.weights_scaling_factor'] = get_fp8_llama(
f'{prefix}:mlp:gate:weights_scaling_factor')
weights[
f'{tllm_prex}.mlp.proj.activation_scaling_factor'] = get_fp8_llama(
f'{prefix}:mlp:proj:activation_scaling_factor')
weights[f'{tllm_prex}.mlp.proj.weights_scaling_factor'] = get_fp8_llama(
f'{prefix}:mlp:proj:weights_scaling_factor')
if fp8_kv_cache:
# Not calibrating KV cache.
scaling_factor = 1.0
weights[
f'{tllm_prex}.attention.kv_cache_scaling_factor'] = torch.tensor(
[scaling_factor], dtype=fake_fp8_sf_dt).numpy()
if fp8_llama is None:
weights.update(weight_scales)
return weights
def dummy_scaling_factor_sq(weights):
for name in list(weights):
if any([
_name in name for _name in [
'mlp.proj.weight', 'mlp.gate.weight', 'mlp.fc.weight',
'attention.qkv.weight', 'attention.dense.weight'
]
]):
print("Processing:", name)
weight = weights[name]
out_dim, in_dim = weight.shape
weights_scaling_factor = (np.abs(weight).max(1, keepdims=True) /
127.)
prequant_scaling_factor = np.ones([in_dim], dtype=weight.dtype)
activation_scaling_factor = np.array([0.1], dtype=np.float32)
int_weight = (weight / weights_scaling_factor).round().astype(
np.int8)
weights[name.replace(
'weight', 'prequant_scaling_factor')] = prequant_scaling_factor
weights[name.replace(
'weight',
'weights_scaling_factor')] = weights_scaling_factor.astype(
np.float32).squeeze(1)
weights[name.replace(
'weight',
'activation_scaling_factor')] = activation_scaling_factor
weights[name] = int_weight
return weights
def dummy_scaling_factor_kv_cache(weights):
for name in list(weights):
if 'attention.qkv.weight' in name:
kv_cache_scaling_factor = np.array([0.1], dtype=np.float32)
weights[name.replace(
'qkv.weight',
'kv_cache_scaling_factor')] = kv_cache_scaling_factor
def dummy_weights_awq(weights, precision, trt_llm_config, group_size):
packer = torch.ops.trtllm.pack_int8_tensor_to_packed_int4
use_fp8_kv_cache = trt_llm_config.quant_mode.has_fp8_kv_cache()
use_int8_kv_cache = trt_llm_config.quant_mode.has_int8_kv_cache()
num_layers = trt_llm_config.num_hidden_layers
for name in list(weights):
if any([
_name in name for _name in [
'mlp.proj.weight', 'mlp.gate.weight', 'mlp.fc.weight',
'attention.qkv.weight', 'attention.dense.weight'
]
]):
print("Processing:", name)
weight = np.ascontiguousarray(weights[name].T)
in_dim, out_dim = weight.shape
scale = np.amax(weight) / 7
weights_scaling_factor = np.ones([out_dim, in_dim // group_size
]) * scale.astype(np.float32)
weight_smoothed = (weight.astype(np.float32) / scale).astype(
np.int8)
weight_smoothed[weight_smoothed < -8] = -8
weight_smoothed[weight_smoothed > 7] = 7
prequant_scaling_factor = np.ones([in_dim], dtype=weight.dtype)
weights[name] = packer(
torch.from_numpy(weight_smoothed)).T.contiguous().numpy()
weights[name.replace(
'weight', 'prequant_scaling_factor')] = prequant_scaling_factor
weights[name.replace(
'weight',
'weights_scaling_factor')] = weights_scaling_factor.astype(
weight.dtype)
if precision == "w4a8_awq":
alpha = np.array([1], dtype=np.float32)
weights[name.replace('weight', 'alpha')] = alpha
if use_fp8_kv_cache or use_int8_kv_cache:
for l in range(num_layers):
t = np.array([1], dtype=np.float32)
weights[
f"transformer.layers.{l}.attention.kv_cache_scaling_factor"] = t
return weights