prunedgpt2 - a question

[pepinu]

I've looked through the PrunedGPT2 class in the autoregressive.py side by side, and I did not see too much difference than in embedding creation. I wonder, do you have a pruned model somewhere available? If not what's the strategy for pruning?

[pepinu]

oh, I forgot to attach, I've updated all the kv_cache stuff from unpruned function and removed the syntax errors (? init function arg list looked like copy pasted by mistake), here it is:

class PrunedGPT2InferenceModel(GPT2PreTrainedModel):
    def __init__(self, config, gpt, text_pos_emb, embeddings, norm, linear, kv_cache):
        super().__init__(config)
        self.transformer = gpt
        self.text_pos_embedding = text_pos_emb
        self.embeddings = embeddings
        self.lm_head = nn.Sequential(norm, linear)
        self.kv_cache = kv_cache

    def store_mel_emb(self, mel_emb):
        self.cached_mel_emb = mel_emb

    def prepare_inputs_for_generation(self, input_ids, past=None, **kwargs):
        token_type_ids = kwargs.get("token_type_ids", None)
        if not self.kv_cache: past_key_values = None
        # only last token for inputs_ids if past is defined in kwargs
        if past:
            input_ids = input_ids[:, -1].unsqueeze(-1)
            if token_type_ids is not None:
                token_type_ids = token_type_ids[:, -1].unsqueeze(-1)

        attention_mask = kwargs.get("attention_mask", None)
        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:
                position_ids = position_ids[:, -1].unsqueeze(-1)
        else:
            position_ids = None
        return {
            "input_ids": input_ids,
            "past_key_values": past,
            "use_cache": kwargs.get("use_cache"),
            "position_ids": position_ids,
            "attention_mask": attention_mask,
            "token_type_ids": token_type_ids,
        }

    def forward(
        self,
        input_ids=None,
        past_key_values=None
        attention_mask=None,
        position_ids=None, **kwargs
        token_type_ids=None
        head_mask=None
        inputs_embeds=None
        encoder_hidden_states=None
        encoder_attention_mask=None
        labels=None
        use_cache=True
        output_attentions=False
        output_hidden_states=False
        return_dict=True
    ):
        assert self.cached_mel_emb is not None
        assert inputs_embeds is None  # Not supported by this inference model.
        assert labels is None  # Training not supported by this inference model.
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        # Create embedding
        mel_len = self.cached_mel_emb.shape[1]
        text_inputs = input_ids[:, mel_len:]
        text_emb = self.embeddings(text_inputs)
        text_emb = text_emb + self.text_pos_embedding(text_emb)
        mel_emb = self.cached_mel_emb.repeat_interleave(text_emb.shape[0]//self.cached_mel_emb.shape[0], 0)
        emb = torch.cat([mel_emb, text_emb], dim=1)

        transformer_outputs = self.transformer(
            inputs_embeds=emb,
            past_key_values=past_key_values,
            attention_mask=attention_mask,
            token_type_ids=token_type_ids,
            position_ids=position_ids,
            head_mask=head_mask,
            encoder_hidden_states=encoder_hidden_states,
            encoder_attention_mask=encoder_attention_mask,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )
        hidden_states = transformer_outputs[0]

        lm_logits = self.lm_head(hidden_states)

        if not return_dict:
            return (lm_logits,) + transformer_outputs[1:]
        return CausalLMOutputWithCrossAttentions(
            loss=None,
            logits=lm_logits,
            past_key_values=transformer_outputs.past_key_values,
            hidden_states=transformer_outputs.hidden_states,
            attentions=transformer_outputs.attentions,
            cross_attentions=transformer_outputs.cross_attentions,
        )

    @staticmethod
    def _reorder_cache(past, beam_idx):
        """
        This function is used to re-order the :obj:`past_key_values` cache if
        :meth:`~transformers.PreTrainedModel.beam_search` or :meth:`~transformers.PreTrainedModel.beam_sample` is
        called. This is required to match :obj:`past_key_values` with the correct beam_idx at every generation step.
        """
        return tuple(
            tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past)
            for layer_past in past
        )

However, it would be cool to rewrite this with inheritance

[152334H]

Committing the prunedgpt class code was a mistake on my part. It was a part of my attempt to create a proper huggingface class for the model that wasn't poorly inherited off of the base transformers.GPT2Model. Ultimately I made some false assumptions about what would be required; I do not have any actual pruned model atm.