From fc74e865589d65da4419ffb5529842c367b09439 Mon Sep 17 00:00:00 2001 From: Sean Lee Date: Sun, 7 Apr 2024 11:25:36 +0800 Subject: [PATCH] fix & add qwen --- README.md | 11 + src/billm/__init__.py | 1 + src/billm/modeling_llama.py | 409 +++++++++++++++++++++-- src/billm/modeling_mistral.py | 316 +++++++++++++++++- src/billm/modeling_qwen2.py | 592 +++++++++++++++++++++++++++++++++ tests/test_modeling_llama.py | 60 ++++ tests/test_modeling_mistral.py | 60 ++++ tests/test_modeling_qwen2.py | 75 +++++ 8 files changed, 1490 insertions(+), 34 deletions(-) create mode 100644 src/billm/modeling_qwen2.py create mode 100644 tests/test_modeling_qwen2.py diff --git a/README.md b/README.md index 92fd1b2..0952cf7 100644 --- a/README.md +++ b/README.md @@ -32,18 +32,29 @@ Tool for converting LLMs from uni-directional to bi-directional for tasks like c ```diff - from transformers import ( - LLamaModel, +- LLamaForCausalLM, - LLamaForSequenceClassification, - MistralModel, +- MistralForCausalLM, - MistralForSequenceClassification +- Qwen2Model, +- Qwen2ForCausalLM, +- Qwen2ForSequenceClassification - ) + from billm import ( + LLamaModel, ++ LLamaForCausalLM, + LLamaForSequenceClassification, + LLamaForTokenClassification, + MistralModel, ++ MistralForCausalLM, + MistralForSequenceClassification, + MistralForTokenClassification, ++ Qwen2Model, ++ Qwen2ForCausalLM, ++ Qwen2ForSequenceClassification, ++ Qwen2ForTokenClassification + ) ``` diff --git a/src/billm/__init__.py b/src/billm/__init__.py index d87d706..7b16d14 100644 --- a/src/billm/__init__.py +++ b/src/billm/__init__.py @@ -2,6 +2,7 @@ from .modeling_llama import * # NOQA from .modeling_mistral import * # NOQA +from .modeling_qwen2 import * # NOQA __version__ = '0.1.2' diff --git a/src/billm/modeling_llama.py b/src/billm/modeling_llama.py index 5d0fc5b..2307b43 100644 --- a/src/billm/modeling_llama.py +++ b/src/billm/modeling_llama.py @@ -7,6 +7,9 @@ from .config import BiLLM_START_INDEX, logger +_CONFIG_FOR_DOC = "LlamaConfig" + + @add_start_docstrings( "The bare LLaMA Model outputting raw hidden-states without any specific head on top.", LLAMA_START_DOCSTRING, @@ -28,16 +31,11 @@ def __init__(self, config: LlamaConfig): self.bidirectionas = [BiLLM_START_INDEX > -1 and layer_idx >= BiLLM_START_INDEX for layer_idx in range(config.num_hidden_layers)] self.layers = nn.ModuleList( - [LlamaDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]) + [LlamaDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)] + ) self.norm = LlamaRMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.gradient_checkpointing = False - # Register a causal mask to separate causal and padding mask creation. Merging happens in the attention class. - # NOTE: This is not friendly with TorchScript, ONNX, ExportedProgram serialization for very large `max_position_embeddings`. - causal_mask = torch.full( - (config.max_position_embeddings, config.max_position_embeddings), fill_value=True, dtype=torch.bool - ) - self.register_buffer("causal_mask", torch.triu(causal_mask, diagonal=1), persistent=False) # Initialize weights and apply final processing self.post_init() if BiLLM_START_INDEX > -1: @@ -102,7 +100,9 @@ def forward( if position_ids is None: position_ids = cache_position.unsqueeze(0) - causal_mask = self._update_causal_mask(attention_mask, inputs_embeds) + causal_mask = self._update_causal_mask( + attention_mask, inputs_embeds, cache_position, past_seen_tokens + inputs_embeds.shape[1] + ) bi_attention_mask = torch.zeros_like(causal_mask) # embed positions @@ -171,25 +171,27 @@ def forward( # KV cache is used. This is an issue for torch.compile which then recaptures cudagraphs at each decode steps due to the dynamic shapes. # (`recording cudagraph tree for symint key 13`, etc.), which is VERY slow. A workaround is `@torch.compiler.disable`, but this prevents using # `fullgraph=True`. See more context in https://github.com/huggingface/transformers/pull/29114 - def _update_causal_mask(self, attention_mask, input_tensor): + def _update_causal_mask(self, attention_mask, input_tensor, cache_position, current_length): if self.config._attn_implementation == "flash_attention_2": if attention_mask is not None and 0.0 in attention_mask: return attention_mask return None - batch_size, seq_length = input_tensor.shape[:2] - dtype = input_tensor.dtype - device = input_tensor.device - - # support going beyond cached `max_position_embedding` - if seq_length > self.causal_mask.shape[-1]: - causal_mask = torch.full((2 * self.causal_mask.shape[-1], 2 * self.causal_mask.shape[-1]), fill_value=1) - self.register_buffer("causal_mask", torch.triu(causal_mask, diagonal=1), persistent=False) - - # We use the current dtype to avoid any overflows + dtype, device = input_tensor.dtype, input_tensor.device min_dtype = torch.finfo(dtype).min - causal_mask = self.causal_mask[None, None, :, :].to(dtype=dtype, device=device) * min_dtype - causal_mask = causal_mask.expand(batch_size, 1, -1, -1) + sequence_length = input_tensor.shape[1] + if hasattr(getattr(self.layers[0], "self_attn", {}), "past_key_value"): # static cache + target_length = self.config.max_position_embeddings + else: # dynamic cache + target_length = ( + attention_mask.shape[-1] if isinstance(attention_mask, torch.Tensor) else current_length + 1 + ) + + causal_mask = torch.full((sequence_length, target_length), fill_value=min_dtype, dtype=dtype, device=device) + if sequence_length != 1: + causal_mask = torch.triu(causal_mask, diagonal=1) + causal_mask *= torch.arange(target_length, device=device) > cache_position.reshape(-1, 1) + causal_mask = causal_mask[None, None, :, :].expand(input_tensor.shape[0], 1, -1, -1) if attention_mask is not None: causal_mask = causal_mask.clone() # copy to contiguous memory for in-place edit if attention_mask.dim() == 2: @@ -197,30 +199,248 @@ def _update_causal_mask(self, attention_mask, input_tensor): padding_mask = causal_mask[..., :mask_length].eq(0.0) * attention_mask[:, None, None, :].eq(0.0) causal_mask[..., :mask_length] = causal_mask[..., :mask_length].masked_fill(padding_mask, min_dtype) elif attention_mask.dim() == 4: + # backwards compatibility: we allow passing a 4D attention mask shorter than the input length with + # cache. In that case, the 4D attention mask attends to the newest tokens only. + if attention_mask.shape[-2] < cache_position[0] + sequence_length: + offset = cache_position[0] + else: + offset = 0 mask_shape = attention_mask.shape mask_slice = (attention_mask.eq(0.0)).to(dtype=dtype) * min_dtype - causal_mask[: mask_shape[0], : mask_shape[1], : mask_shape[2], : mask_shape[3]] = mask_slice + causal_mask[ + : mask_shape[0], : mask_shape[1], offset : mask_shape[2] + offset, : mask_shape[3] + ] = mask_slice if ( self.config._attn_implementation == "sdpa" and attention_mask is not None and attention_mask.device.type == "cuda" ): - # TODO: For dynamo, rather use a check on fullgraph=True once this is possible (https://github.com/pytorch/pytorch/pull/120400). - is_tracing = ( - torch.jit.is_tracing() - or isinstance(input_tensor, torch.fx.Proxy) - or (hasattr(torch, "_dynamo") and torch._dynamo.is_compiling()) - ) - if not is_tracing and torch.any(attention_mask != 1): - # Attend to all tokens in fully masked rows in the causal_mask, for example the relevant first rows when - # using left padding. This is required by F.scaled_dot_product_attention memory-efficient attention path. - # Details: https://github.com/pytorch/pytorch/issues/110213 - causal_mask = AttentionMaskConverter._unmask_unattended(causal_mask, min_dtype) + # Attend to all tokens in fully masked rows in the causal_mask, for example the relevant first rows when + # using left padding. This is required by F.scaled_dot_product_attention memory-efficient attention path. + # Details: https://github.com/pytorch/pytorch/issues/110213 + causal_mask = AttentionMaskConverter._unmask_unattended(causal_mask, min_dtype) return causal_mask +class LlamaForCausalLM(LlamaPreTrainedModel): + _tied_weights_keys = ["lm_head.weight"] + + def __init__(self, config): + super().__init__(config) + self.model = LlamaModel(config) + self.vocab_size = config.vocab_size + self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + def get_output_embeddings(self): + return self.lm_head + + def set_output_embeddings(self, new_embeddings): + self.lm_head = new_embeddings + + def set_decoder(self, decoder): + self.model = decoder + + def get_decoder(self): + return self.model + + @add_start_docstrings_to_model_forward(LLAMA_INPUTS_DOCSTRING) + @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + cache_position: Optional[torch.LongTensor] = None, + ) -> Union[Tuple, CausalLMOutputWithPast]: + r""" + Args: + labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): + Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., + config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored + (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`. + + Returns: + + Example: + + ```python + >>> from transformers import AutoTokenizer, LlamaForCausalLM + + >>> model = LlamaForCausalLM.from_pretrained("meta-llama/Llama-2-7b-hf") + >>> tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-hf") + + >>> prompt = "Hey, are you conscious? Can you talk to me?" + >>> inputs = tokenizer(prompt, return_tensors="pt") + + >>> # Generate + >>> generate_ids = model.generate(inputs.input_ids, max_length=30) + >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] + "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you." + ```""" + output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions + output_hidden_states = ( + output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states + ) + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn) + outputs = self.model( + input_ids=input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + cache_position=cache_position, + ) + + hidden_states = outputs[0] + if self.config.pretraining_tp > 1: + lm_head_slices = self.lm_head.weight.split(self.vocab_size // self.config.pretraining_tp, dim=0) + logits = [F.linear(hidden_states, lm_head_slices[i]) for i in range(self.config.pretraining_tp)] + logits = torch.cat(logits, dim=-1) + else: + logits = self.lm_head(hidden_states) + logits = logits.float() + + loss = None + if labels is not None: + # Shift so that tokens < n predict n + shift_logits = logits[..., :-1, :].contiguous() + shift_labels = labels[..., 1:].contiguous() + # Flatten the tokens + loss_fct = CrossEntropyLoss() + shift_logits = shift_logits.view(-1, self.config.vocab_size) + shift_labels = shift_labels.view(-1) + # Enable model parallelism + shift_labels = shift_labels.to(shift_logits.device) + loss = loss_fct(shift_logits, shift_labels) + + if not return_dict: + output = (logits,) + outputs[1:] + return (loss,) + output if loss is not None else output + + return CausalLMOutputWithPast( + loss=loss, + logits=logits, + past_key_values=outputs.past_key_values, + hidden_states=outputs.hidden_states, + attentions=outputs.attentions, + ) + + def prepare_inputs_for_generation( + self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, **kwargs + ): + # With static cache, the `past_key_values` is None + # TODO joao: standardize interface for the different Cache classes and remove of this if + has_static_cache = False + if past_key_values is None: + past_key_values = getattr(getattr(self.model.layers[0], "self_attn", {}), "past_key_value", None) + has_static_cache = past_key_values is not None + + past_length = 0 + if past_key_values is not None: + if isinstance(past_key_values, Cache): + past_length = cache_position[0] if cache_position is not None else past_key_values.get_seq_length() + max_cache_length = ( + torch.tensor(past_key_values.get_max_length(), device=input_ids.device) + if past_key_values.get_max_length() is not None + else None + ) + cache_length = past_length if max_cache_length is None else torch.min(max_cache_length, past_length) + # TODO joao: remove this `else` after `generate` prioritizes `Cache` objects + else: + cache_length = past_length = past_key_values[0][0].shape[2] + max_cache_length = None + + # Keep only the unprocessed tokens: + # 1 - If the length of the attention_mask exceeds the length of input_ids, then we are in a setting where + # some of the inputs are exclusively passed as part of the cache (e.g. when passing input_embeds as + # input) + if attention_mask is not None and attention_mask.shape[1] > input_ids.shape[1]: + input_ids = input_ids[:, -(attention_mask.shape[1] - past_length) :] + # 2 - If the past_length is smaller than input_ids', then input_ids holds all input tokens. We can discard + # input_ids based on the past_length. + elif past_length < input_ids.shape[1]: + input_ids = input_ids[:, past_length:] + # 3 - Otherwise (past_length >= input_ids.shape[1]), let's assume input_ids only has unprocessed tokens. + + # If we are about to go beyond the maximum cache length, we need to crop the input attention mask. + if ( + max_cache_length is not None + and attention_mask is not None + and cache_length + input_ids.shape[1] > max_cache_length + ): + attention_mask = attention_mask[:, -max_cache_length:] + + position_ids = kwargs.get("position_ids", None) + if attention_mask is not None and position_ids is None: + # create position_ids on the fly for batch generation + position_ids = attention_mask.long().cumsum(-1) - 1 + position_ids.masked_fill_(attention_mask == 0, 1) + if past_key_values: + position_ids = position_ids[:, -input_ids.shape[1] :] + + # if `inputs_embeds` are passed, we only want to use them in the 1st generation step + if inputs_embeds is not None and past_key_values is None: + model_inputs = {"inputs_embeds": inputs_embeds} + else: + # The `contiguous()` here is necessary to have a static stride during decoding. torchdynamo otherwise + # recompiles graphs as the stride of the inputs is a guard. Ref: https://github.com/huggingface/transformers/pull/29114 + # TODO: use `next_tokens` directly instead. + model_inputs = {"input_ids": input_ids.contiguous()} + + input_length = position_ids.shape[-1] if position_ids is not None else input_ids.shape[-1] + if cache_position is None: + cache_position = torch.arange(past_length, past_length + input_length, device=input_ids.device) + else: + cache_position = cache_position[-input_length:] + + if has_static_cache: + past_key_values = None + + model_inputs.update( + { + "position_ids": position_ids, + "cache_position": cache_position, + "past_key_values": past_key_values, + "use_cache": kwargs.get("use_cache"), + "attention_mask": attention_mask, + } + ) + return model_inputs + + @staticmethod + def _reorder_cache(past_key_values, beam_idx): + reordered_past = () + for layer_past in past_key_values: + reordered_past += ( + tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past), + ) + return reordered_past + + @add_start_docstrings( """ The Bi-LLaMa Model transformer with a token classification head on top (linear layer). @@ -303,3 +523,126 @@ def forward( hidden_states=outputs.hidden_states, attentions=outputs.attentions, ) + + +@add_start_docstrings( + """ + The LLaMa Model transformer with a sequence classification head on top (linear layer). + + [`LlamaForSequenceClassification`] uses the last token in order to do the classification, as other causal models + (e.g. GPT-2) do. + + Since it does classification on the last token, it requires to know the position of the last token. If a + `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If + no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the + padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in + each row of the batch). + """, + LLAMA_START_DOCSTRING, +) +class LlamaForSequenceClassification(LlamaPreTrainedModel): + def __init__(self, config): + super().__init__(config) + self.num_labels = config.num_labels + self.model = LlamaModel(config) + self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + @add_start_docstrings_to_model_forward(LLAMA_INPUTS_DOCSTRING) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, SequenceClassifierOutputWithPast]: + r""" + labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): + Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., + config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If + `config.num_labels > 1` a classification loss is computed (Cross-Entropy). + """ + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + transformer_outputs = self.model( + input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + hidden_states = transformer_outputs[0] + logits = self.score(hidden_states) + + if input_ids is not None: + batch_size = input_ids.shape[0] + else: + batch_size = inputs_embeds.shape[0] + + if self.config.pad_token_id is None and batch_size != 1: + raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.") + if self.config.pad_token_id is None: + sequence_lengths = -1 + else: + if input_ids is not None: + # if no pad token found, use modulo instead of reverse indexing for ONNX compatibility + sequence_lengths = torch.eq(input_ids, self.config.pad_token_id).int().argmax(-1) - 1 + sequence_lengths = sequence_lengths % input_ids.shape[-1] + sequence_lengths = sequence_lengths.to(logits.device) + else: + sequence_lengths = -1 + + pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths] + + loss = None + if labels is not None: + labels = labels.to(logits.device) + if self.config.problem_type is None: + if self.num_labels == 1: + self.config.problem_type = "regression" + elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int): + self.config.problem_type = "single_label_classification" + else: + self.config.problem_type = "multi_label_classification" + + if self.config.problem_type == "regression": + loss_fct = MSELoss() + if self.num_labels == 1: + loss = loss_fct(pooled_logits.squeeze(), labels.squeeze()) + else: + loss = loss_fct(pooled_logits, labels) + elif self.config.problem_type == "single_label_classification": + loss_fct = CrossEntropyLoss() + loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1)) + elif self.config.problem_type == "multi_label_classification": + loss_fct = BCEWithLogitsLoss() + loss = loss_fct(pooled_logits, labels) + if not return_dict: + output = (pooled_logits,) + transformer_outputs[1:] + return ((loss,) + output) if loss is not None else output + + return SequenceClassifierOutputWithPast( + loss=loss, + logits=pooled_logits, + past_key_values=transformer_outputs.past_key_values, + hidden_states=transformer_outputs.hidden_states, + attentions=transformer_outputs.attentions, + ) diff --git a/src/billm/modeling_mistral.py b/src/billm/modeling_mistral.py index 2191b4d..ce13833 100644 --- a/src/billm/modeling_mistral.py +++ b/src/billm/modeling_mistral.py @@ -7,6 +7,9 @@ from .config import BiLLM_START_INDEX, logger +_CONFIG_FOR_DOC = "MistralConfig" + + @add_start_docstrings( "The bare Mistral Model outputting raw hidden-states without any specific head on top.", MISTRAL_START_DOCSTRING, @@ -28,7 +31,8 @@ def __init__(self, config: MistralConfig): self.bidirectionas = [BiLLM_START_INDEX > -1 and layer_idx >= BiLLM_START_INDEX for layer_idx in range(config.num_hidden_layers)] self.layers = nn.ModuleList( - [MistralDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]) + [MistralDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)] + ) self._attn_implementation = config._attn_implementation self.norm = MistralRMSNorm(config.hidden_size, eps=config.rms_norm_eps) @@ -195,6 +199,192 @@ def forward( ) +class MistralForCausalLM(MistralPreTrainedModel): + _tied_weights_keys = ["lm_head.weight"] + + def __init__(self, config): + super().__init__(config) + self.model = MistralModel(config) + self.vocab_size = config.vocab_size + self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + def get_output_embeddings(self): + return self.lm_head + + def set_output_embeddings(self, new_embeddings): + self.lm_head = new_embeddings + + def set_decoder(self, decoder): + self.model = decoder + + def get_decoder(self): + return self.model + + @add_start_docstrings_to_model_forward(MISTRAL_INPUTS_DOCSTRING) + @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, CausalLMOutputWithPast]: + r""" + Args: + labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): + Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., + config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored + (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`. + + Returns: + + Example: + + ```python + >>> from transformers import AutoTokenizer, MistralForCausalLM + + >>> model = MistralForCausalLM.from_pretrained("mistralai/Mistral-7B-v0.1") + >>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1") + + >>> prompt = "Hey, are you conscious? Can you talk to me?" + >>> inputs = tokenizer(prompt, return_tensors="pt") + + >>> # Generate + >>> generate_ids = model.generate(inputs.input_ids, max_length=30) + >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] + "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you." + ```""" + + output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions + output_hidden_states = ( + output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states + ) + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn) + outputs = self.model( + input_ids=input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + + hidden_states = outputs[0] + logits = self.lm_head(hidden_states) + logits = logits.float() + + loss = None + if labels is not None: + # Shift so that tokens < n predict n + shift_logits = logits[..., :-1, :].contiguous() + shift_labels = labels[..., 1:].contiguous() + # Flatten the tokens + shift_logits = shift_logits.view(-1, self.config.vocab_size) + shift_labels = shift_labels.view(-1) + # Ensure tensors are on the same device + shift_labels = shift_labels.to(shift_logits.device) + loss_fct = CrossEntropyLoss() + loss = loss_fct(shift_logits, shift_labels) + + if not return_dict: + output = (logits,) + outputs[1:] + return (loss,) + output if loss is not None else output + + return CausalLMOutputWithPast( + loss=loss, + logits=logits, + past_key_values=outputs.past_key_values, + hidden_states=outputs.hidden_states, + attentions=outputs.attentions, + ) + + def prepare_inputs_for_generation( + self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs + ): + # Omit tokens covered by past_key_values + if past_key_values is not None: + if isinstance(past_key_values, Cache): + cache_length = past_key_values.get_seq_length() + past_length = past_key_values.seen_tokens + max_cache_length = past_key_values.get_max_length() + else: + cache_length = past_length = past_key_values[0][0].shape[2] + max_cache_length = None + + # Keep only the unprocessed tokens: + # 1 - If the length of the attention_mask exceeds the length of input_ids, then we are in a setting where + # some of the inputs are exclusively passed as part of the cache (e.g. when passing input_embeds as + # input) + if attention_mask is not None and attention_mask.shape[1] > input_ids.shape[1]: + input_ids = input_ids[:, -(attention_mask.shape[1] - past_length) :] + # 2 - If the past_length is smaller than input_ids', then input_ids holds all input tokens. We can discard + # input_ids based on the past_length. + elif past_length < input_ids.shape[1]: + input_ids = input_ids[:, past_length:] + # 3 - Otherwise (past_length >= input_ids.shape[1]), let's assume input_ids only has unprocessed tokens. + + # If we are about to go beyond the maximum cache length, we need to crop the input attention mask. + if ( + max_cache_length is not None + and attention_mask is not None + and cache_length + input_ids.shape[1] > max_cache_length + ): + attention_mask = attention_mask[:, -max_cache_length:] + + position_ids = kwargs.get("position_ids", None) + if attention_mask is not None and position_ids is None: + # create position_ids on the fly for batch generation + position_ids = attention_mask.long().cumsum(-1) - 1 + position_ids.masked_fill_(attention_mask == 0, 1) + if past_key_values: + position_ids = position_ids[:, -input_ids.shape[1] :] + + # if `inputs_embeds` are passed, we only want to use them in the 1st generation step + if inputs_embeds is not None and past_key_values is None: + model_inputs = {"inputs_embeds": inputs_embeds} + else: + model_inputs = {"input_ids": input_ids} + + model_inputs.update( + { + "position_ids": position_ids, + "past_key_values": past_key_values, + "use_cache": kwargs.get("use_cache"), + "attention_mask": attention_mask, + } + ) + return model_inputs + + @staticmethod + def _reorder_cache(past_key_values, beam_idx): + reordered_past = () + for layer_past in past_key_values: + reordered_past += ( + tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past), + ) + return reordered_past + + @add_start_docstrings( """ The Bi-Mistral Model transformer with a token classification head on top (linear layer). @@ -277,3 +467,127 @@ def forward( hidden_states=outputs.hidden_states, attentions=outputs.attentions, ) + + +@add_start_docstrings( + """ + The Mistral Model transformer with a sequence classification head on top (linear layer). + + [`MistralForSequenceClassification`] uses the last token in order to do the classification, as other causal models + (e.g. GPT-2) do. + + Since it does classification on the last token, it requires to know the position of the last token. If a + `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If + no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the + padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in + each row of the batch). + """, + MISTRAL_START_DOCSTRING, +) +# Copied from transformers.models.llama.modeling_llama.LlamaForSequenceClassification with Llama->Mistral, LLAMA->MISTRAL +class MistralForSequenceClassification(MistralPreTrainedModel): + def __init__(self, config): + super().__init__(config) + self.num_labels = config.num_labels + self.model = MistralModel(config) + self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + @add_start_docstrings_to_model_forward(MISTRAL_INPUTS_DOCSTRING) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, SequenceClassifierOutputWithPast]: + r""" + labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): + Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., + config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If + `config.num_labels > 1` a classification loss is computed (Cross-Entropy). + """ + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + transformer_outputs = self.model( + input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + hidden_states = transformer_outputs[0] + logits = self.score(hidden_states) + + if input_ids is not None: + batch_size = input_ids.shape[0] + else: + batch_size = inputs_embeds.shape[0] + + if self.config.pad_token_id is None and batch_size != 1: + raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.") + if self.config.pad_token_id is None: + sequence_lengths = -1 + else: + if input_ids is not None: + # if no pad token found, use modulo instead of reverse indexing for ONNX compatibility + sequence_lengths = torch.eq(input_ids, self.config.pad_token_id).int().argmax(-1) - 1 + sequence_lengths = sequence_lengths % input_ids.shape[-1] + sequence_lengths = sequence_lengths.to(logits.device) + else: + sequence_lengths = -1 + + pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths] + + loss = None + if labels is not None: + labels = labels.to(logits.device) + if self.config.problem_type is None: + if self.num_labels == 1: + self.config.problem_type = "regression" + elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int): + self.config.problem_type = "single_label_classification" + else: + self.config.problem_type = "multi_label_classification" + + if self.config.problem_type == "regression": + loss_fct = MSELoss() + if self.num_labels == 1: + loss = loss_fct(pooled_logits.squeeze(), labels.squeeze()) + else: + loss = loss_fct(pooled_logits, labels) + elif self.config.problem_type == "single_label_classification": + loss_fct = CrossEntropyLoss() + loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1)) + elif self.config.problem_type == "multi_label_classification": + loss_fct = BCEWithLogitsLoss() + loss = loss_fct(pooled_logits, labels) + if not return_dict: + output = (pooled_logits,) + transformer_outputs[1:] + return ((loss,) + output) if loss is not None else output + + return SequenceClassifierOutputWithPast( + loss=loss, + logits=pooled_logits, + past_key_values=transformer_outputs.past_key_values, + hidden_states=transformer_outputs.hidden_states, + attentions=transformer_outputs.attentions, + ) diff --git a/src/billm/modeling_qwen2.py b/src/billm/modeling_qwen2.py new file mode 100644 index 0000000..1cdb77e --- /dev/null +++ b/src/billm/modeling_qwen2.py @@ -0,0 +1,592 @@ +# -*- coding: utf-8 -*- + +from transformers.models.qwen2.modeling_qwen2 import * +from transformers.modeling_attn_mask_utils import _prepare_4d_causal_attention_mask, _prepare_4d_causal_attention_mask_for_sdpa +from transformers.modeling_outputs import TokenClassifierOutput + +from .config import BiLLM_START_INDEX, logger + + +_CONFIG_FOR_DOC = "Qwen2Config" + + +@add_start_docstrings( + "The bare Qwen2 Model outputting raw hidden-states without any specific head on top.", + QWEN2_START_DOCSTRING, +) +class Qwen2Model(Qwen2PreTrainedModel): + """ + Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`Qwen2DecoderLayer`] + + Args: + config: Qwen2Config + """ + + def __init__(self, config: Qwen2Config): + super().__init__(config) + self.padding_idx = config.pad_token_id + self.vocab_size = config.vocab_size + + self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx) + self.bidirectionas = [BiLLM_START_INDEX > -1 and layer_idx >= BiLLM_START_INDEX + for layer_idx in range(config.num_hidden_layers)] + self.layers = nn.ModuleList( + [Qwen2DecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)] + ) + self._attn_implementation = config._attn_implementation + self.norm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) + + self.gradient_checkpointing = False + # Initialize weights and apply final processing + self.post_init() + if BiLLM_START_INDEX > -1: + logger.info(f'Here is the Bi-Qwen2 model! BiLLM_START_INDEX={BiLLM_START_INDEX}') + else: + logger.info(f'BiLLM_START_INDEX={BiLLM_START_INDEX}, BiLLM is disabled.') + + def get_input_embeddings(self): + return self.embed_tokens + + def set_input_embeddings(self, value): + self.embed_tokens = value + + @add_start_docstrings_to_model_forward(QWEN2_INPUTS_DOCSTRING) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, BaseModelOutputWithPast]: + output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions + output_hidden_states = ( + output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states + ) + use_cache = use_cache if use_cache is not None else self.config.use_cache + + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + # retrieve input_ids and inputs_embeds + if input_ids is not None and inputs_embeds is not None: + raise ValueError("You cannot specify both decoder_input_ids and decoder_inputs_embeds at the same time") + elif input_ids is not None: + batch_size, seq_length = input_ids.shape + elif inputs_embeds is not None: + batch_size, seq_length, _ = inputs_embeds.shape + else: + raise ValueError("You have to specify either decoder_input_ids or decoder_inputs_embeds") + + if self.gradient_checkpointing and self.training: + if use_cache: + logger.warning_once( + "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..." + ) + use_cache = False + + past_key_values_length = 0 + + if use_cache: + use_legacy_cache = not isinstance(past_key_values, Cache) + if use_legacy_cache: + past_key_values = DynamicCache.from_legacy_cache(past_key_values) + past_key_values_length = past_key_values.get_usable_length(seq_length) + + if position_ids is None: + device = input_ids.device if input_ids is not None else inputs_embeds.device + position_ids = torch.arange( + past_key_values_length, seq_length + past_key_values_length, dtype=torch.long, device=device + ) + position_ids = position_ids.unsqueeze(0).view(-1, seq_length) + else: + position_ids = position_ids.view(-1, seq_length).long() + + if inputs_embeds is None: + inputs_embeds = self.embed_tokens(input_ids) + + if attention_mask is not None and self._attn_implementation == "flash_attention_2" and use_cache: + is_padding_right = attention_mask[:, -1].sum().item() != batch_size + if is_padding_right: + raise ValueError( + "You are attempting to perform batched generation with padding_side='right'" + " this may lead to unexpected behaviour for Flash Attention version of Qwen2. Make sure to " + " call `tokenizer.padding_side = 'left'` before tokenizing the input. " + ) + + if self._attn_implementation == "flash_attention_2": + # 2d mask is passed through the layers + attention_mask = attention_mask if (attention_mask is not None and 0 in attention_mask) else None + elif self._attn_implementation == "sdpa" and not output_attentions: + # output_attentions=True can not be supported when using SDPA, and we fall back on + # the manual implementation that requires a 4D causal mask in all cases. + attention_mask = _prepare_4d_causal_attention_mask_for_sdpa( + attention_mask, + (batch_size, seq_length), + inputs_embeds, + past_key_values_length, + ) + else: + # 4d mask is passed through the layers + attention_mask = _prepare_4d_causal_attention_mask( + attention_mask, + (batch_size, seq_length), + inputs_embeds, + past_key_values_length, + sliding_window=self.config.sliding_window, + ) + bi_attention_mask = torch.zeros_like(attention_mask) + + hidden_states = inputs_embeds + + # decoder layers + all_hidden_states = () if output_hidden_states else None + all_self_attns = () if output_attentions else None + next_decoder_cache = None + + for is_bidirectional, decoder_layer in zip(self.bidirectionas, self.layers): + if output_hidden_states: + all_hidden_states += (hidden_states,) + + if self.gradient_checkpointing and self.training: + layer_outputs = self._gradient_checkpointing_func( + decoder_layer.__call__, + hidden_states, + bi_attention_mask if is_bidirectional else attention_mask, + position_ids, + past_key_values, + output_attentions, + use_cache, + ) + else: + layer_outputs = decoder_layer( + hidden_states, + attention_mask=bi_attention_mask if is_bidirectional else attention_mask, + position_ids=position_ids, + past_key_value=past_key_values, + output_attentions=output_attentions, + use_cache=use_cache, + ) + + hidden_states = layer_outputs[0] + + if use_cache: + next_decoder_cache = layer_outputs[2 if output_attentions else 1] + + if output_attentions: + all_self_attns += (layer_outputs[1],) + + hidden_states = self.norm(hidden_states) + + # add hidden states from the last decoder layer + if output_hidden_states: + all_hidden_states += (hidden_states,) + + next_cache = None + if use_cache: + next_cache = next_decoder_cache.to_legacy_cache() if use_legacy_cache else next_decoder_cache + + if not return_dict: + return tuple(v for v in [hidden_states, next_cache, all_hidden_states, all_self_attns] if v is not None) + return BaseModelOutputWithPast( + last_hidden_state=hidden_states, + past_key_values=next_cache, + hidden_states=all_hidden_states, + attentions=all_self_attns, + ) + + +class Qwen2ForCausalLM(Qwen2PreTrainedModel): + _tied_weights_keys = ["lm_head.weight"] + + def __init__(self, config): + super().__init__(config) + self.model = Qwen2Model(config) + self.vocab_size = config.vocab_size + self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + def get_output_embeddings(self): + return self.lm_head + + def set_output_embeddings(self, new_embeddings): + self.lm_head = new_embeddings + + def set_decoder(self, decoder): + self.model = decoder + + def get_decoder(self): + return self.model + + @add_start_docstrings_to_model_forward(QWEN2_INPUTS_DOCSTRING) + @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, CausalLMOutputWithPast]: + r""" + Args: + labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): + Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., + config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored + (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`. + + Returns: + + Example: + + ```python + >>> from transformers import AutoTokenizer, Qwen2ForCausalLM + + >>> model = Qwen2ForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS) + >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER) + + >>> prompt = "Hey, are you conscious? Can you talk to me?" + >>> inputs = tokenizer(prompt, return_tensors="pt") + + >>> # Generate + >>> generate_ids = model.generate(inputs.input_ids, max_length=30) + >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] + "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you." + ```""" + + output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions + output_hidden_states = ( + output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states + ) + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn) + outputs = self.model( + input_ids=input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + + hidden_states = outputs[0] + logits = self.lm_head(hidden_states) + logits = logits.float() + + loss = None + if labels is not None: + # Shift so that tokens < n predict n + shift_logits = logits[..., :-1, :].contiguous() + shift_labels = labels[..., 1:].contiguous() + # Flatten the tokens + loss_fct = CrossEntropyLoss() + shift_logits = shift_logits.view(-1, self.config.vocab_size) + shift_labels = shift_labels.view(-1) + # Enable model parallelism + shift_labels = shift_labels.to(shift_logits.device) + loss = loss_fct(shift_logits, shift_labels) + + if not return_dict: + output = (logits,) + outputs[1:] + return (loss,) + output if loss is not None else output + + return CausalLMOutputWithPast( + loss=loss, + logits=logits, + past_key_values=outputs.past_key_values, + hidden_states=outputs.hidden_states, + attentions=outputs.attentions, + ) + + def prepare_inputs_for_generation( + self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs + ): + # Omit tokens covered by past_key_values + if past_key_values is not None: + if isinstance(past_key_values, Cache): + cache_length = past_key_values.get_seq_length() + past_length = past_key_values.seen_tokens + max_cache_length = past_key_values.get_max_length() + else: + cache_length = past_length = past_key_values[0][0].shape[2] + max_cache_length = None + + # Keep only the unprocessed tokens: + # 1 - If the length of the attention_mask exceeds the length of input_ids, then we are in a setting where + # some of the inputs are exclusively passed as part of the cache (e.g. when passing input_embeds as + # input) + if attention_mask is not None and attention_mask.shape[1] > input_ids.shape[1]: + input_ids = input_ids[:, -(attention_mask.shape[1] - past_length) :] + # 2 - If the past_length is smaller than input_ids', then input_ids holds all input tokens. We can discard + # input_ids based on the past_length. + elif past_length < input_ids.shape[1]: + input_ids = input_ids[:, past_length:] + # 3 - Otherwise (past_length >= input_ids.shape[1]), let's assume input_ids only has unprocessed tokens. + + # If we are about to go beyond the maximum cache length, we need to crop the input attention mask. + if ( + max_cache_length is not None + and attention_mask is not None + and cache_length + input_ids.shape[1] > max_cache_length + ): + attention_mask = attention_mask[:, -max_cache_length:] + + position_ids = kwargs.get("position_ids", None) + if attention_mask is not None and position_ids is None: + # create position_ids on the fly for batch generation + position_ids = attention_mask.long().cumsum(-1) - 1 + position_ids.masked_fill_(attention_mask == 0, 1) + if past_key_values: + position_ids = position_ids[:, -input_ids.shape[1] :] + + # if `inputs_embeds` are passed, we only want to use them in the 1st generation step + if inputs_embeds is not None and past_key_values is None: + model_inputs = {"inputs_embeds": inputs_embeds} + else: + model_inputs = {"input_ids": input_ids} + + model_inputs.update( + { + "position_ids": position_ids, + "past_key_values": past_key_values, + "use_cache": kwargs.get("use_cache"), + "attention_mask": attention_mask, + } + ) + return model_inputs + + @staticmethod + def _reorder_cache(past_key_values, beam_idx): + reordered_past = () + for layer_past in past_key_values: + reordered_past += ( + tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past), + ) + return reordered_past + + +@add_start_docstrings( + """ + The Qwen2 Model transformer with a sequence classification head on top (linear layer). + + [`Qwen2ForSequenceClassification`] uses the last token in order to do the classification, as other causal models + (e.g. GPT-2) do. + + Since it does classification on the last token, it requires to know the position of the last token. If a + `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If + no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the + padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in + each row of the batch). + """, + QWEN2_START_DOCSTRING, +) +class Qwen2ForSequenceClassification(Qwen2PreTrainedModel): + def __init__(self, config): + super().__init__(config) + self.num_labels = config.num_labels + self.model = Qwen2Model(config) + self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + @add_start_docstrings_to_model_forward(QWEN2_INPUTS_DOCSTRING) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, SequenceClassifierOutputWithPast]: + r""" + labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): + Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., + config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If + `config.num_labels > 1` a classification loss is computed (Cross-Entropy). + """ + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + transformer_outputs = self.model( + input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + hidden_states = transformer_outputs[0] + logits = self.score(hidden_states) + + if input_ids is not None: + batch_size = input_ids.shape[0] + else: + batch_size = inputs_embeds.shape[0] + + if self.config.pad_token_id is None and batch_size != 1: + raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.") + if self.config.pad_token_id is None: + sequence_lengths = -1 + else: + if input_ids is not None: + # if no pad token found, use modulo instead of reverse indexing for ONNX compatibility + sequence_lengths = torch.eq(input_ids, self.config.pad_token_id).int().argmax(-1) - 1 + sequence_lengths = sequence_lengths % input_ids.shape[-1] + sequence_lengths = sequence_lengths.to(logits.device) + else: + sequence_lengths = -1 + + pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths] + + loss = None + if labels is not None: + labels = labels.to(logits.device) + if self.config.problem_type is None: + if self.num_labels == 1: + self.config.problem_type = "regression" + elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int): + self.config.problem_type = "single_label_classification" + else: + self.config.problem_type = "multi_label_classification" + + if self.config.problem_type == "regression": + loss_fct = MSELoss() + if self.num_labels == 1: + loss = loss_fct(pooled_logits.squeeze(), labels.squeeze()) + else: + loss = loss_fct(pooled_logits, labels) + elif self.config.problem_type == "single_label_classification": + loss_fct = CrossEntropyLoss() + loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1)) + elif self.config.problem_type == "multi_label_classification": + loss_fct = BCEWithLogitsLoss() + loss = loss_fct(pooled_logits, labels) + if not return_dict: + output = (pooled_logits,) + transformer_outputs[1:] + return ((loss,) + output) if loss is not None else output + + return SequenceClassifierOutputWithPast( + loss=loss, + logits=pooled_logits, + past_key_values=transformer_outputs.past_key_values, + hidden_states=transformer_outputs.hidden_states, + attentions=transformer_outputs.attentions, + ) + + +@add_start_docstrings( + """ + The Bi-Qwen2 Model transformer with a token classification head on top (linear layer). + """, + QWEN2_START_DOCSTRING, +) +class Qwen2ForTokenClassification(Qwen2PreTrainedModel): + def __init__(self, config): + super().__init__(config) + self.num_labels = config.num_labels + self.model = Qwen2Model(config) + if hasattr(config, "classifier_dropout") and config.classifier_dropout is not None: + classifier_dropout = config.classifier_dropout + elif hasattr(config, "hidden_dropout") and config.hidden_dropout is not None: + classifier_dropout = config.hidden_dropout + else: + classifier_dropout = 0.1 + self.dropout = nn.Dropout(classifier_dropout) + self.classifier = nn.Linear(config.hidden_size, config.num_labels) + + # Initialize weights and apply final processing + self.post_init() + + def get_input_embeddings(self): + return self.model.embed_tokens + + def set_input_embeddings(self, value): + self.model.embed_tokens = value + + @add_start_docstrings_to_model_forward(QWEN2_INPUTS_DOCSTRING) + def forward( + self, + input_ids: torch.LongTensor = None, + attention_mask: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + past_key_values: Optional[List[torch.FloatTensor]] = None, + inputs_embeds: Optional[torch.FloatTensor] = None, + labels: Optional[torch.LongTensor] = None, + use_cache: Optional[bool] = None, + output_attentions: Optional[bool] = None, + output_hidden_states: Optional[bool] = None, + return_dict: Optional[bool] = None, + ) -> Union[Tuple, SequenceClassifierOutputWithPast]: + r""" + labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): + Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., + config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If + `config.num_labels > 1` a classification loss is computed (Cross-Entropy). + """ + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + + outputs = self.model( + input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + past_key_values=past_key_values, + inputs_embeds=inputs_embeds, + use_cache=use_cache, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + sequence_output = outputs[0] + + sequence_output = self.dropout(sequence_output) + logits = self.classifier(sequence_output) + + loss = None + if labels is not None: + loss_fct = CrossEntropyLoss() + loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1)) + + if not return_dict: + output = (logits,) + outputs[2:] + return ((loss,) + output) if loss is not None else output + + return TokenClassifierOutput( + loss=loss, + logits=logits, + hidden_states=outputs.hidden_states, + attentions=outputs.attentions, + ) diff --git a/tests/test_modeling_llama.py b/tests/test_modeling_llama.py index bf3b7e2..e4bdd23 100644 --- a/tests/test_modeling_llama.py +++ b/tests/test_modeling_llama.py @@ -13,3 +13,63 @@ def test_llama_model(): num_hidden_layers=2, num_attention_heads=2)) assert model is not None + + +def test_billama_model(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import LlamaModel, LlamaConfig + + model = LlamaModel(LlamaConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + + + +def test_billama_lm(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import LlamaForCausalLM, LlamaConfig + + model = LlamaForCausalLM(LlamaConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 + + +def test_billama_seq_clf(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import LlamaForSequenceClassification, LlamaConfig + + model = LlamaForSequenceClassification(LlamaConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 + + +def test_billama_token_clf(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import LlamaForTokenClassification, LlamaConfig + + model = LlamaForTokenClassification(LlamaConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 diff --git a/tests/test_modeling_mistral.py b/tests/test_modeling_mistral.py index 0ba8ed1..77c4805 100644 --- a/tests/test_modeling_mistral.py +++ b/tests/test_modeling_mistral.py @@ -2,6 +2,20 @@ def test_mistral_model(): + import os + os.environ['BiLLM_START_INDEX'] = '-1' + + from billm import MistralModel, MistralConfig + + model = MistralModel(MistralConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + + +def test_bimistral_model(): import os os.environ['BiLLM_START_INDEX'] = '1' @@ -13,3 +27,49 @@ def test_mistral_model(): num_hidden_layers=2, num_attention_heads=2)) assert model is not None + + + +def test_bimistral_lm(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import MistralForCausalLM, MistralConfig + + model = MistralForCausalLM(MistralConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 + + +def test_bimistral_seq_clf(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import MistralForSequenceClassification, MistralConfig + + model = MistralForSequenceClassification(MistralConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 + + +def test_bimistral_token_clf(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import MistralForTokenClassification, MistralConfig + + model = MistralForTokenClassification(MistralConfig(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 diff --git a/tests/test_modeling_qwen2.py b/tests/test_modeling_qwen2.py new file mode 100644 index 0000000..a69eeb9 --- /dev/null +++ b/tests/test_modeling_qwen2.py @@ -0,0 +1,75 @@ +# -*- coding: utf-8 -*- + + +def test_qwen2_model(): + import os + os.environ['BiLLM_START_INDEX'] = '-1' + + from billm import Qwen2Model, Qwen2Config + + model = Qwen2Model(Qwen2Config(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + + +def test_biqwen2_model(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import Qwen2Model, Qwen2Config + + model = Qwen2Model(Qwen2Config(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + + + +def test_biqwen2_lm(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import Qwen2ForCausalLM, Qwen2Config + + model = Qwen2ForCausalLM(Qwen2Config(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 + + +def test_biqwen2_seq_clf(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import Qwen2ForSequenceClassification, Qwen2Config + + model = Qwen2ForSequenceClassification(Qwen2Config(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0 + + +def test_biqwen2_token_clf(): + import os + os.environ['BiLLM_START_INDEX'] = '1' + + from billm import Qwen2ForTokenClassification, Qwen2Config + + model = Qwen2ForTokenClassification(Qwen2Config(vocab_size=128, + hidden_size=32, + intermediate_size=64, + num_hidden_layers=2, + num_attention_heads=2)) + assert model is not None + assert len(model.model.bidirectionas) > 0