Comments (27)
I was just reading the fast attention code, and I think it does exactly what we want. Typing is really the only reason the c++ code is torch-specific. Otherwise, all the logic should directly transfer to TF. My issue is that I don't know tf. I think we only need to change the #include
line at the top of the c++ file, the typing in the c++ file, and the wrapper code in __init__.py
, then it will be TF compatible
from gpt2-quickly.
close it, reopen if it needs.
from gpt2-quickly.
close it, reopen if it needs.
BERT & DistilBERT Performers now work with pretrained transformers (https://colab.research.google.com/drive/1A9reiUZbA7DELuJ8keTo73sIQ4dJJVoT#scrollTo=F5k6jxicGf3E)
Still working on autoregressive models like gpt .. but it seems the gains won't be that big for them; will keep you updated!
from gpt2-quickly.
Great work! @Muennighoff
After trying your code,BERT
&DistilBERT
Performers
does work better than the normal Transformer.
I tried GPT2 work with Performers a few days ago and got some clue, I think the reason ofPerformer
will perform well in non-causal models likeBERT
andT5
but will have a negative effect in causal models likeGPT
,
maybe is the matrix multiplication ofsoftmax(Q @ K) @ CausalMask @ V
is faster then the loop multiplication of_headwise_causal_numerator
.Yeah the problem seems to be the loop multiplication - We could try the causal multiplication from luciddrains - it shouldn't be too difficult to copy it over.
I'm also still running into shape errors when trying GPT2 Performer - did you get it to work?
It looks so cool! I'll try it and keep you updated!
from gpt2-quickly.
I don't think I'm the guy to do this (I don't use c++ or tensorflow), but I think this is a pretty easy problem for someone who at least knows tensorflow
I started an implementation here. @mymusise will you take a look?
from gpt2-quickly.
Thanks for the feedback!
- Yes, the installations of the
sentencepiece
is required. Trypip install -r requirements.txt
to update your environment. - It's ok to use
tokenizers.SentencePieceBPETokenizer
, but both of them require the input sentence should be pretokenized in no-spaces languages such as Chinese or Japanese. - I forgot to add
pretoken
inbuild_tokenizer.py
, I will update it later. fast attention
is still not available in this repo, hope norabelrose's work coming soon.
from gpt2-quickly.
Thanks for the feedback!
- Yes, the installations of the
sentencepiece
is required. Trypip install -r requirements.txt
to update your environment.- It's ok to use
tokenizers.SentencePieceBPETokenizer
, but both of them require the input sentence should be pretokenized in no-spaces languages such as Chinese or Japanese.- I forgot to add
pretoken
inbuild_tokenizer.py
, I will update it later.fast attention
is still not available in this repo, hope norabelrose's work coming soon.
1-3: Great, thanks for the info.
4: Yeah that would be awesome. I've also been working on it on a fork from transformers, but it's still unable to converge somehow, so I was thinking of using the original tensorflow implementation from g-research as you did. But I guess it didnt work out for you?
from gpt2-quickly.
so I was thinking of using the original tensorflow implementation from g-research as you did. But I guess it didnt work out for you?
I try both of them two months ago but still have some problems.
In my test:
transformer.PerformerAttention
works but it train and predict slower than the normal Transformer.- the
fast-attention
of google train and predict faster, but the result of prediction is bad. It's confusing that the accuracy will stop at 66% with training a small corpus like dataset/test/raw.txt. Perhaps I build the model in the wrong way.
Hope it helps :)
and wait for your good news 🚀
from gpt2-quickly.
Great work! @Muennighoff
After trying your code, BERT
& DistilBERT
Performers
does work better than the normal Transformer.
I tried GPT2 work with Performers a few days ago and got some clue, I think the reason of Performer
will perform well in non-causal models like BERT
and T5
but will have a negative effect in causal models like GPT
,
maybe is the matrix multiplication of softmax(Q @ K) @ CausalMask @ V
is faster then the loop multiplication of _headwise_causal_numerator
.
from gpt2-quickly.
Great work! @Muennighoff
After trying your code,BERT
&DistilBERT
Performers
does work better than the normal Transformer.
I tried GPT2 work with Performers a few days ago and got some clue, I think the reason ofPerformer
will perform well in non-causal models likeBERT
andT5
but will have a negative effect in causal models likeGPT
,
maybe is the matrix multiplication ofsoftmax(Q @ K) @ CausalMask @ V
is faster then the loop multiplication of_headwise_causal_numerator
.
Yeah the problem seems to be the loop multiplication - We could try the causal multiplication from lucidrains - it shouldn't be too difficult to copy it over.
I'm also still running into shape errors when trying GPT2 Performer - did you get it to work?
from gpt2-quickly.
@Muennighoff Hi, I got an OOM error after using the causal_linear_attention_noncuda
function from luciddrains
.
I think it's a bad idea to generate a content matrix with torch.einsum('...nd,...ne->...nde', k, v)
.
If the shape of Q, K, V
is [B, H, L, D], and the context will have a huge size [B, H, L, D, D]. In many applications, the value of D is in [64-128]. Maybe it has a little deviation from the design of performer
.
from gpt2-quickly.
@Muennighoff Hi, I got an OOM error after using the
causal_linear_attention_noncuda
function fromluciddrains
.
I think it's a bad idea to generate a content matrix withtorch.einsum('...nd,...ne->...nde', k, v)
.
If the shape ofQ, K, V
is [B, H, L, D], and the context will have a huge size [B, H, L, D, D]. In many applications, the value of D is in [64-128]. Maybe it has a little deviation from the design ofperformer
.
Yes i think that's why he also has the other version above (https://github.com/lucidrains/performer-pytorch/blob/3bff14e39284e7dc82952153099a63dcd3561dc0/performer_pytorch/performer_pytorch.py#L142) Did you try that one as well?
from gpt2-quickly.
Yes i think that's why he also has the other version above (https://github.com/lucidrains/performer-pytorch/blob/3bff14e39284e7dc82952153099a63dcd3561dc0/performer_pytorch/performer_pytorch.py#L142) Did you try that one as well?
I didn't try it, but I think maybe rewriting the loop multiplication with the basic API of CUDA, is a way to work out the performance. Some like fast_transformers.causal_product.CausalDotProduct
do.
from gpt2-quickly.
Hi all, I'm happy to join this conversation if there's anything to be done.
Is this too inefficient?
import numpy as np
q0 = np.array([[1, 2, 3], [4, 5, 6]])
k0 = np.array([[2, 3, 4], [5, 6, 7]])
v0 = np.array([[3, 4, 5, 6], [7, 8, 9, 10]])
q = np.expand_dims(q0, axis=1)
k = np.expand_dims(k0, axis=2)
v = np.expand_dims(v0, axis=2).transpose(0, 2, 1)
assert q.shape == (2, 1, 3)
assert k.shape == (2, 3, 1)
assert v.shape == (2, 1, 4)
kv = (k @ v)
assert kv.shape == (2, 3, 4)
kv_sum = np.cumsum(kv, axis=0)
assert kv_sum.shape == (2, 3, 4)
qkv = q @ kv_sum
assert qkv.shape == (2, 1, 4)
qkv = np.squeeze(qkv, axis=1)
assert (qkv == ((q0 @ k0.T) * np.array([[1, 0], [1, 1]])) @ v0).all()
from gpt2-quickly.
Welcome guy! @JamesDeAntonis
Is this too inefficient?
No, I think it can speed up the operation but will spend more memory.
And I'm not sure the dimension of q0
(2, 3) mean (L, D)? A sequence with two tokens and each token has 3 dimensions?
If so, I think the code above is equal to torch.einsum('...nd,...ne->...nde', k, v)
in a way.
As I comment above, it will spend a lot of memory.
If the shape of Q, K, V is [B, H, L, D], and the context will have a huge size [B, H, L, D, D]. In many applications, the value of D is in [64-128]. Maybe it has a little deviation from the design of performer.
from gpt2-quickly.
And I'm not sure the dimension of q0 (2, 3) mean (L, D)? A sequence with two tokens and each token has 3 dimensions?
Yes, that's what I meant.
If the shape of Q, K, V is [B, H, L, D], and the context will have a huge size [B, H, L, D, D]. In many applications, the value of D is in [64-128]. Maybe it has a little deviation from the design of performer.
Don't you iterate across heads in the code, to save memory? This would mean [B, 1, L, D, D]
, whereas regular transformer has [B, H, L, L]
. Generally speaking, I'm unsure what exactly you are trying to change.
from gpt2-quickly.
Does this mean CausalDotProduct
in fast-attention
is what we want?
from gpt2-quickly.
Don't you iterate across heads in the code, to save memory?
Great, that's a good idea.
This would mean [B, 1, L, D, D], whereas regular transformer has [B, H, L, L]
Generally speaking, the value of D is in [64-128], if we have L = 2048
, D = 128
, then L*D*D=2048*128*128
is much bigger than the regular transformer L*L=2048*2048
, if H=8
, then H*L*L
== L*D*D
, it means we can save more memory when L>2048
. But we still have to iterate each head if we do what you mention above.
I‘m not sure I've understood your mind。
I'm unsure what exactly you are trying to change.
Actually, I've no idea to speed up the causal multiplication while saving memory. 😷
from gpt2-quickly.
isn't this only solvable by implementing the for-loop directly in the lower-level language? I imagine this is effectively what fast attention does
from gpt2-quickly.
Does this mean
CausalDotProduct
infast-attention
is what we want?
Yes, but I'd not test yea, cause I can't find the tensorflow version and I've not mastered C++. 😂
from gpt2-quickly.
I imagine this is effectively what fast attention does
Yes, I have the same guess.
from gpt2-quickly.
Can we talk over a call? I just emailed you
from gpt2-quickly.
@JamesDeAntonis Cool!
I did not read the fast attention code seriously, sounds like it's worth trying.
from gpt2-quickly.
Cool! You are so efficient! @JamesDeAntonis
Actually, before converting to a tf version, I prefer to test fast-attention.CausalDotProduct
in PyTorch first.
At the same time, I will try to convert it base on your implementation.
Muennighoff seems to be solving the masking problem of the T5 decoder.
@Muennighoff Did you try fast-attention.CausalDotProduct
? Dose it work?
from gpt2-quickly.
@ice-americano (who I work with) ran if and it seemed to work to some degree. Compared to regular attention, he was getting significant improvements in memory usage, but a noticeable slowdown. We're not sure why the slowdown is occurring
from gpt2-quickly.
@mymusise I made some experiments with the fast-attention.CausalDotProduct on the SST notebook (https://colab.research.google.com/drive/1A9reiUZbA7DELuJ8keTo73sIQ4dJJVoT#scrollTo=f_R7D-mZyRXm) --- & the CausalDotProduct itself seems to work, however I get a shape mismatch in the denominator/normalization, since the sequence length of the key may not be the same as the query in a t5 enc-dec; As the keys come from the encoder, but the queries from the decoder input;
If i force the shapes to fit or just skip the normalization, it goes through but always predicts the same value giving 50% accuracy whereas the equivalent transformer converges (check it under t5/pytorch/encdec in the notebook);
this might tho be unrelated to the CausalDotProduct and could e.g. be caused by the attention mask or sth else --- @JamesDeAntonis did @ice-americano get it to converge?
from gpt2-quickly.
Yes, we got it to converge.
On our task, in relation to regular attention, it was (i) a step worse in terms of loss, (ii) noticeably lower memory utilization on long sequences, and (iii) slower (this is the most perplexing to us)
from gpt2-quickly.
Related Issues (8)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from gpt2-quickly.