modeling_utils.py

import copy
import json
import os
from typing import List, Optional

import safetensors

from tensorrt_llm.plugin.plugin import init_all_reduce_helper

from .._common import default_net
from .._utils import str_dtype_to_trt
from ..functional import PositionEmbeddingType, Tensor, gather_last_token_logits
from ..layers import AttentionParams, KeyValueCacheParams, LoraParams
from ..mapping import Mapping
from ..module import Module, ModuleList
from ..quantization import QuantMode
from ..quantization.quantize import quantize
from .generation_mixin import GenerationMixin


class PretrainedConfig:

    def __init__(self,
                 architecture: str,
                 dtype: str,
                 logits_dtype: str,
                 vocab_size: int,
                 max_position_embeddings: int,
                 hidden_size: int,
                 num_hidden_layers: int,
                 num_attention_heads: int,
                 num_key_value_heads: int,
                 hidden_act: str,
                 intermediate_size: int,
                 norm_epsilon: float,
                 position_embedding_type: str,
                 world_size: int,
                 tp_size: int,
                 pp_size: int,
                 quant_mode: QuantMode,
                 quant_kwargs: dict,
                 use_prompt_tuning: bool = False,
                 use_parallel_embedding: bool = False,
                 embedding_sharding_dim: int = 0,
                 share_embedding_table: bool = False,
                 **kwargs):
        self.architecture = architecture
        self.dtype = dtype
        self.logits_dtype = logits_dtype

        self.vocab_size = vocab_size
        self.max_position_embeddings = max_position_embeddings
        self.hidden_size = hidden_size
        self.num_hidden_layers = num_hidden_layers
        self.num_attention_heads = num_attention_heads
        self.num_key_value_heads = num_key_value_heads
        self.hidden_act = hidden_act
        self.intermediate_size = intermediate_size
        self.norm_epsilon = norm_epsilon
        self.position_embedding_type = PositionEmbeddingType.from_string(
            position_embedding_type)
        self.use_prompt_tuning = use_prompt_tuning
        self.use_parallel_embedding = use_parallel_embedding
        self.embedding_sharding_dim = embedding_sharding_dim
        self.share_embedding_table = share_embedding_table
        self.mapping = Mapping(world_size=world_size,
                               tp_size=tp_size,
                               pp_size=pp_size)
        self.quant_mode = quant_mode
        self.quant_kwargs = quant_kwargs
        self.kv_dtype = self.dtype
        if self.quant_mode.has_int8_kv_cache():
            self.kv_dtype = 'int8'
        elif self.quant_mode.has_fp8_kv_cache():
            self.kv_dtype = 'fp8'

        for key, value in kwargs.items():
            try:
                setattr(self, key, value)
            except AttributeError as err:
                raise err

    def set_if_not_exist(self, key, value):
        if not hasattr(self, key):
            setattr(self, key, value)

    @classmethod
    def from_dict(cls, config):
        architecture = config.pop('architecture')
        dtype = config.pop('dtype')
        vocab_size = config.pop('vocab_size')
        hidden_size = config.pop('hidden_size')
        num_hidden_layers = config.pop('num_hidden_layers')
        num_attention_heads = config.pop('num_attention_heads')
        hidden_act = config.pop('hidden_act')
        norm_epsilon = config.pop('norm_epsilon', 1e-5)
        position_embedding_type = config.pop('position_embedding_type',
                                             'learned_absolute')
        logits_dtype = config.pop('logits_dtype', 'float32')
        num_key_value_heads = config.pop('num_key_value_heads',
                                         num_attention_heads)
        intermediate_size = config.pop('intermediate_size', None)
        max_position_embeddings = config.pop('max_position_embeddings', None)
        use_prompt_tuning = config.pop('use_prompt_tuning', False)
        use_parallel_embedding = config.pop('use_parallel_embedding', False)
        embedding_sharding_dim = config.pop('embedding_sharding_dim', 0)
        share_embedding_table = config.pop('share_embedding_table', False)

        mapping = config.pop('mapping', {
            'world_size': 1,
            'tp_size': 1,
            'pp_size': 1
        })
        world_size = mapping.get('world_size', 1)
        tp_size = mapping.get('tp_size', 1)
        pp_size = mapping.get('pp_size', 1)

        if share_embedding_table and mapping.tp_size > 1:
            if (not use_parallel_embedding) or (use_parallel_embedding and
                                                embedding_sharding_dim == 1):
                raise NotImplementedError(
                    "For multiple-processes cases, sharing the embedding table must set" \
                        "use_parallel_embedding=True and embedding_sharding_dim=0"
                )

        quantization = config.pop(
            'quantization', {
                'use_smooth_quant': False,
                'per_channel': False,
                'per_token': False,
                'per_group': False,
                'group_size': 128,
                'zero': False,
                'pre_quant_scale': False,
                'exclude_modules': None,
                'int8_kv_cache': False,
                'enable_fp8': False,
                'fp8_kv_cache': False,
                'use_weight_only': False,
                'weight_only_precision': 'int8'
            })
        use_smooth_quant = quantization.get('use_smooth_quant', False)
        per_channel = quantization.get('per_channel', False)
        per_token = quantization.get('per_token', False)
        per_group = quantization.get('per_group', False)
        group_size = quantization.get('group_size', 128)
        zero = quantization.get('zero', False)
        pre_quant_scale = quantization.get('pre_quant_scale', False)
        exclude_modules = quantization.get('exclude_modules', None)
        int8_kv_cache = quantization.get('int8_kv_cache', False)
        enable_fp8 = quantization.get('enable_fp8', False)
        fp8_kv_cache = quantization.get('fp8_kv_cache', False)
        use_weight_only = quantization.get('use_weight_only', False)
        weight_only_precision = quantization.get('weight_only_precision',
                                                 'int8')

        quantize_weights, quantize_activations = False, False
        if use_smooth_quant:
            quantize_weights = True
            quantize_activations = True
        elif use_weight_only:
            quantize_weights = True
            per_token = False
            per_channel = False

        quant_mode = QuantMode.from_description(
            quantize_weights=quantize_weights,
            quantize_activations=quantize_activations,
            per_token=per_token,
            per_channel=per_channel,
            per_group=per_group,
            use_int4_weights=(weight_only_precision == 'int4'),
            use_int8_kv_cache=int8_kv_cache,
            use_fp8_kv_cache=fp8_kv_cache,
            use_fp8_qdq=enable_fp8,
        )

        quant_kwargs = {
            'group_size': group_size,
            'zero': zero,
            'pre_quant_scale': pre_quant_scale,
            'exclude_modules': exclude_modules,
        }

        return cls(architecture, dtype, logits_dtype, vocab_size,
                   max_position_embeddings, hidden_size, num_hidden_layers,
                   num_attention_heads, num_key_value_heads, hidden_act,
                   intermediate_size, norm_epsilon, position_embedding_type,
                   world_size, tp_size, pp_size, quant_mode, quant_kwargs,
                   use_prompt_tuning, use_parallel_embedding,
                   embedding_sharding_dim, share_embedding_table, **config)

    @classmethod
    def from_json_file(cls, config_file: str):
        with open(config_file) as f:
            config = json.load(f)
            return PretrainedConfig.from_dict(config)

    def to_dict(self):
        output = copy.deepcopy(self.__dict__)

        output['position_embedding_type'] = str(self.position_embedding_type)
        output['mapping'] = {
            'world_size': self.mapping.world_size,
            'tp_size': self.mapping.tp_size,
            'pp_size': self.mapping.pp_size,
        }
        output.pop('quant_mode')
        output.pop('quant_kwargs')
        output['quantization'] = {
            'use_smooth_quant':
            self.quant_mode.has_act_and_weight_quant(),
            'per_channel':
            self.quant_mode.has_per_channel_scaling(),
            'per_token':
            self.quant_mode.has_per_token_dynamic_scaling(),
            'per_group':
            self.quant_mode.has_per_group_scaling(),
            'group_size':
            self.quant_kwargs.get('group_size', 128),
            'zero':
            self.quant_kwargs.get('zero', False),
            'pre_quant_scale':
            self.quant_kwargs.get('pre_quant_scale', False),
            'exclude_modules':
            self.quant_kwargs.get('exclude_modules', None),
            'int8_kv_cache':
            self.quant_mode.has_int8_kv_cache(),
            'enable_fp8':
            self.quant_mode.has_fp8_qdq(),
            'fp8_kv_cache':
            self.quant_mode.has_fp8_kv_cache(),
            'use_weight_only':
            self.quant_mode.is_weight_only(),
            'weight_only_precision':
            'int8' if self.quant_mode.is_int8_weight_only() else 'int4',
        }

        return output

    def set_rank(self, rank):
        self.mapping = Mapping(self.mapping.world_size,
                               rank=rank,
                               tp_size=self.mapping.tp_size,
                               pp_size=self.mapping.pp_size)


class DecoderLayerList(ModuleList):

    def __init__(self, cls, config):
        self.layer_list = config.mapping.pp_layers(config.num_hidden_layers)
        super().__init__([cls(config, idx) for idx in self.layer_list])

    def forward(self,
                hidden_states,
                use_cache=False,
                attention_mask=None,
                kv_cache_params=None,
                attention_params=None,
                lora_params=None):
        kv_cache_params.fill_none_tensor_list(len(self.layer_list))

        if use_cache:
            presents = []

        for layer_idx, (
                layer, past, pointer, host_pointer,
                max_attention_window_size) in enumerate(
                    zip(self, kv_cache_params.past_key_value,
                        kv_cache_params.kv_cache_block_pointers,
                        kv_cache_params.host_kv_cache_block_pointers,
                        kv_cache_params.host_max_attention_window_sizes)):

            lora_param = None
            if lora_params is not None and lora_params.lora_ranks is not None:
                lora_param = lora_params.get_layer_params(layer_idx)

            kwargs = {}
            if lora_param is not None:
                kwargs['lora_param'] = lora_param
            hidden_states = layer(
                hidden_states,
                use_cache=use_cache,
                attention_mask=attention_mask,
                kv_cache_params=KeyValueCacheParams(
                    past_key_value=[past],
                    host_past_key_value_lengths=kv_cache_params.
                    host_past_key_value_lengths,
                    host_max_attention_window_sizes=max_attention_window_size,
                    host_sink_token_length=kv_cache_params.
                    host_sink_token_length,
                    kv_cache_block_pointers=[pointer],
                    host_kv_cache_block_pointers=[host_pointer],
                    cache_indirection=kv_cache_params.cache_indirection),
                attention_params=attention_params,
                **kwargs)

            if use_cache:
                presents.append(hidden_states[1])
                hidden_states = hidden_states[0]

        if use_cache:
            return hidden_states, presents
        return hidden_states


class PostInitCaller(type):

    def __call__(cls, *args, **kwargs):
        obj = type.__call__(cls, *args, **kwargs)
        obj.__post_init__()
        return obj


class PretrainedModel(Module, GenerationMixin, metaclass=PostInitCaller):

    def __init__(self, config):
        super().__init__()
        self.config = config

    def __post_init__(self):
        quantize(self, self.config.quant_mode, **self.config.quant_kwargs)

    @classmethod
    def from_config(cls, config: PretrainedConfig):
        return cls(config)

    @classmethod
    def from_checkpoint(cls,
                        ckpt_dir: str,
                        rank: int = 0,
                        config: PretrainedConfig = None):
        if config is None:
            config = PretrainedConfig.from_json_file(
                os.path.join(ckpt_dir, 'config.json'))
            config.set_rank(rank)
        model = cls.from_config(config)

        weights = {}
        with safetensors.safe_open(os.path.join(ckpt_dir,
                                                f'rank{rank}.safetensors'),
                                   framework='pt',
                                   device='cpu') as f:
            for key in f.keys():
                weights[key] = f.get_tensor(key)

        model.load(weights)

        return model

    def load(self, weights):
        expected_names = set([name for name, param in self.named_parameters()])
        provided_names = set(weights.keys())
        if provided_names != expected_names:
            err_msg = "Provided tensor names are different from those expected by the engine."
            if expected_names.difference(provided_names):
                err_msg += f"\nExpected but not provided tensors: {expected_names.difference(provided_names)}"
            if provided_names.difference(expected_names):
                err_msg += f"\nProvided but not expected tensors: {provided_names.difference(expected_names)}"
            raise RuntimeError(err_msg)

        for name, param in self.named_parameters():
            param.value = weights[name]

    def prepare_inputs(self,
                       max_batch_size,
                       max_input_len,
                       max_new_tokens,
                       use_cache,
                       max_beam_width: int = 1,
                       max_num_tokens: int = None,
                       prompt_embedding_table_size: int = 0,
                       gather_context_logits: bool = False,
                       gather_generation_logits: bool = False,
                       lora_target_modules: List[str] = None):
        '''@brief: Prepare inputs Tensors for the model, the given sizes are used to determine the
            ranges of the dimensions of when using TRT dynamic shapes.

            @return: a list contains values which can be fed into the self.forward()
        '''

        # Prepare inputs
        head_size = self.config.hidden_size // self.config.num_attention_heads
        remove_input_padding = default_net().plugin_config.remove_input_padding
        use_gpt_attention_plugin = default_net(
        ).plugin_config.gpt_attention_plugin
        use_gemm_plugin = default_net().plugin_config.gemm_plugin
        paged_kv_cache = default_net().plugin_config.paged_kv_cache
        tokens_per_block = default_net().plugin_config.tokens_per_block
        use_custom_all_reduce = default_net(
        ).plugin_config.use_custom_all_reduce
        use_lora_plugin = default_net().plugin_config.lora_plugin

        model_inputs = self.prepare_basic_inputs(
            max_batch_size=max_batch_size,
            max_beam_width=max_beam_width,
            max_input_len=max_input_len,
            max_new_tokens=max_new_tokens,
            num_kv_heads=self.config.num_key_value_heads,
            head_size=head_size,
            num_layers=self.config.num_hidden_layers,
            kv_dtype=str_dtype_to_trt(self.config.kv_dtype),
            remove_input_padding=remove_input_padding,
            use_gpt_attention_plugin=use_gpt_attention_plugin,
            use_gemm_plugin=use_gemm_plugin,
            paged_kv_cache=paged_kv_cache,
            tokens_per_block=tokens_per_block,
            num_heads=self.config.num_attention_heads,
            max_num_tokens=max_num_tokens,
            dtype=str_dtype_to_trt(self.config.dtype),
            prompt_embedding_table_size=prompt_embedding_table_size,
            mapping=self.config.mapping,
            gather_context_logits=gather_context_logits,
            gather_generation_logits=gather_generation_logits,
            use_custom_all_reduce=use_custom_all_reduce,
            use_lora_plugin=use_lora_plugin,
            lora_target_modules=lora_target_modules)

        result = {
            'input_ids':
            model_inputs['input_ids'],
            'position_ids':
            model_inputs['position_ids'],
            'use_cache':
            True,
            'last_token_ids':
            model_inputs['last_token_ids'],
            'attention_mask':
            model_inputs['attention_mask'],
            'kv_cache_params':
            KeyValueCacheParams(
                past_key_value=model_inputs['past_key_value'],
                host_past_key_value_lengths=model_inputs[
                    'host_past_key_value_lengths'],
                host_max_attention_window_sizes=model_inputs[
                    'host_max_attention_window_sizes'],
                host_sink_token_length=model_inputs['host_sink_token_length'],
                kv_cache_block_pointers=model_inputs[
                    'kv_cache_block_pointers_list'],
                host_kv_cache_block_pointers=model_inputs[
                    'host_kv_cache_block_pointers_list'],
                cache_indirection=model_inputs['cache_indirection'],
            ),
            'attention_params':
            AttentionParams(
                sequence_length=model_inputs['sequence_length'],
                context_lengths=model_inputs['context_lengths'],
                host_context_lengths=model_inputs['host_context_lengths'],
                max_context_length=max_input_len,
                host_request_types=model_inputs['host_request_types'])
        }

        if prompt_embedding_table_size > 0:
            result['prompt_embedding_table'] = model_inputs[
                'prompt_embedding_table']
            result['prompt_tasks'] = model_inputs['tasks']
            result['prompt_vocab_size'] = model_inputs['prompt_vocab_size']
        if model_inputs['hidden_states_input'] is not None:
            result['hidden_states'] = model_inputs['hidden_states_input']
        if use_lora_plugin:
            result['lora_params'] = LoraParams(
                model_inputs['lora_ranks'],
                model_inputs['lora_weights_pointers'],
                host_context_lengths=model_inputs['host_context_lengths'],
                max_context_length=max_input_len,
                host_request_types=model_inputs['host_request_types'])

        return result


class DecoderModelForCausalLM(PretrainedModel):

    def __init__(self, config, transformer, lm_head):
        super().__init__(config)
        self.transformer = transformer
        self.lm_head = lm_head

    def forward(self,
                input_ids: Tensor,
                position_ids=None,
                use_cache=False,
                last_token_ids=None,
                attention_mask=None,
                kv_cache_params=None,
                attention_params=None,
                hidden_states=None,
                prompt_embedding_table: Optional[Tensor] = None,
                prompt_tasks: Optional[Tensor] = None,
                prompt_vocab_size: Optional[Tensor] = None,
                lora_params=None):
        init_all_reduce_helper()
        kwargs = {
            'input_ids': input_ids,
            'position_ids': position_ids,
            'use_cache': use_cache,
            'attention_mask': attention_mask,
            'kv_cache_params': kv_cache_params,
            'attention_params': attention_params,
        }
        if lora_params is not None:
            kwargs['lora_params'] = lora_params
        if hidden_states is not None:
            kwargs['hidden_states'] = hidden_states
        if prompt_embedding_table is not None:
            kwargs['prompt_embedding_table'] = prompt_embedding_table
        if prompt_tasks is not None:
            kwargs['prompt_tasks'] = prompt_tasks
        if prompt_vocab_size is not None:
            kwargs['prompt_vocab_size'] = prompt_vocab_size

        hidden_states = self.transformer.forward(**kwargs)

        if use_cache:
            hidden_states, presents = hidden_states

        if self.config.mapping.is_last_pp_rank():
            hidden_states = gather_last_token_logits(
                hidden_states, last_token_ids,
                default_net().plugin_config.remove_input_padding)

            # [batch_size, hidden_size] -> [batch_size, vocab_size]
            lm_logits = self.lm_head(hidden_states)
            lm_logits.mark_output('logits', self.config.logits_dtype)
        else:
            hidden_states.mark_output('hidden_states_output', self.config.dtype)

        if use_cache and default_net().plugin_config.paged_kv_cache == False:
            for i, present in zip(
                    self.config.mapping.pp_layers(
                        self.config.num_hidden_layers), presents):
                present.mark_output(f'present_key_value_{i}',
                                    self.config.kv_dtype)
            if self.config.mapping.is_last_pp_rank():
                return (lm_logits, presents)
            return (hidden_states, presents)
        else:
            if self.config.mapping.is_last_pp_rank():
                return lm_logits
            return hidden_states