mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-02 04:10:06 +06:00
508a704055
* No more Tuple, List, Dict * make fixup * More style fixes * Docstring fixes with regex replacement * Trigger tests * Redo fixes after rebase * Fix copies * [test all] * update * [test all] * update * [test all] * make style after rebase * Patch the hf_argparser test * Patch the hf_argparser test * style fixes * style fixes * style fixes * Fix docstrings in Cohere test * [test all] --------- Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
67 lines
3.7 KiB
Python
67 lines
3.7 KiB
Python
# 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
|
|
# This file was automatically generated from examples/modular-transformers/modular_add_function.py.
|
|
# Do NOT edit this file manually as any edits will be overwritten by the generation of
|
|
# the file from the modular. If any change should be done, please apply the change to the
|
|
# modular_add_function.py file directly. One of our CI enforces this.
|
|
# 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
|
|
# Note that zamba does not have the `apply_rotary_pos_emb` function!
|
|
from typing import Optional
|
|
|
|
import torch
|
|
from torch import nn
|
|
|
|
|
|
def rotate_half(x):
|
|
"""Rotates half the hidden dims of the input."""
|
|
x1 = x[..., : x.shape[-1] // 2]
|
|
x2 = x[..., x.shape[-1] // 2 :]
|
|
return torch.cat((-x2, x1), dim=-1)
|
|
|
|
|
|
def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1):
|
|
"""Applies Rotary Position Embedding to the query and key tensors.
|
|
|
|
Args:
|
|
q (`torch.Tensor`): The query tensor.
|
|
k (`torch.Tensor`): The key tensor.
|
|
cos (`torch.Tensor`): The cosine part of the rotary embedding.
|
|
sin (`torch.Tensor`): The sine part of the rotary embedding.
|
|
position_ids (`torch.Tensor`, *optional*):
|
|
Deprecated and unused.
|
|
unsqueeze_dim (`int`, *optional*, defaults to 1):
|
|
The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
|
|
sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
|
|
that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
|
|
k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
|
|
cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
|
|
the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
|
|
Returns:
|
|
`tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
|
|
"""
|
|
cos = cos.unsqueeze(unsqueeze_dim)
|
|
sin = sin.unsqueeze(unsqueeze_dim)
|
|
q_embed = (q * cos) + (rotate_half(q) * sin)
|
|
k_embed = (k * cos) + (rotate_half(k) * sin)
|
|
return q_embed, k_embed
|
|
|
|
|
|
class TestAttention(nn.Module):
|
|
"""
|
|
Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer
|
|
and "Generating Long Sequences with Sparse Transformers".
|
|
|
|
Adapted from transformers.models.mistral.modeling_mistral.MistralAttention:
|
|
The input dimension here is attention_hidden_size = 2 * hidden_size, and head_dim = attention_hidden_size // num_heads.
|
|
The extra factor of 2 comes from the input being the concatenation of original_hidden_states with the output of the previous (mamba) layer
|
|
(see fig. 2 in https://huggingface.co/papers/2405.16712).
|
|
Additionally, replaced
|
|
attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim) with
|
|
attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim/2)
|
|
"""
|
|
|
|
def __init__(self):
|
|
pass
|
|
|
|
def forward(self) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]:
|
|
_ = apply_rotary_pos_emb(1, 1, 1, 1)
|