Update README.md
Browse files
README.md
CHANGED
@@ -2,37 +2,72 @@
|
|
2 |
library_name: transformers
|
3 |
tags: []
|
4 |
---
|
|
|
5 |
# FastESM
|
|
|
6 |
|
7 |
-
|
8 |
|
9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
|
11 |
-
|
|
|
|
|
12 |
|
13 |
Outputting attention maps (or the contact prediction head) is not natively possible with SDPA. You can still pass ```output_attentions``` to have attention calculated manually and returned.
|
14 |
Various other optimizations also make the base implementation slightly different than the one in transformers.
|
15 |
|
|
|
|
|
|
|
|
|
|
|
16 |
## Use with 🤗 transformers
|
|
|
|
|
17 |
```python
|
18 |
import torch
|
19 |
-
from transformers import
|
20 |
|
21 |
model_path = 'Synthyra/FastESM2_650'
|
22 |
-
model =
|
23 |
-
tokenizer =
|
24 |
|
25 |
sequences = ['MPRTEIN', 'MSEQWENCE']
|
26 |
tokenized = tokenizer(sequences, padding=True, return_tensors='pt')
|
27 |
with torch.no_grad():
|
28 |
-
embeddings = model(**tokenized
|
29 |
|
30 |
print(embeddings.shape) # (1, 11, 1280)
|
31 |
```
|
32 |
|
33 |
-
|
34 |
-
|
35 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
|
37 |
## Embed entire datasets with no new code
|
38 |
To embed a list of protein sequences **fast**, just call embed_dataset. Sequences are sorted to reduce padding tokens, so the initial progress bar estimation is usually much longer than the actual time.
|
@@ -60,6 +95,7 @@ _ = model.embed_dataset(
|
|
60 |
sql_db_path='embeddings.db', # path to .db file of choice
|
61 |
)
|
62 |
```
|
|
|
63 |
## Model probes
|
64 |
We employ linear probing techniques on various PLMs and standard datasets, similar our previous [paper](https://www.biorxiv.org/content/10.1101/2024.07.30.605924v1), to assess the intrinsic correlation between pooled hidden states and valuable properties. FastESM performs very well.
|
65 |
|
|
|
2 |
library_name: transformers
|
3 |
tags: []
|
4 |
---
|
5 |
+
|
6 |
# FastESM
|
7 |
+
FastESM is a Huggingface compatible plug in version of ESM2 rewritten with a newer PyTorch attention implementation.
|
8 |
|
9 |
+
Load any ESM2 models into a FastEsm model to dramatically speed up training and inference without **ANY** cost in performance.
|
10 |
|
11 |
+
## Use with 🤗 transformers
|
12 |
+
```python
|
13 |
+
from transformers import AutoModel, AutoModelForMaskedLM, AutoModelForSequenceClassification, AutoModelForTokenClassification # any of these work
|
14 |
+
|
15 |
+
model_dict = {
|
16 |
+
'ESM2-8': 'facebook/esm2_t6_8M_UR50D',
|
17 |
+
'ESM2-35': 'facebook/esm2_t12_35M_UR50D',
|
18 |
+
'ESM2-150': 'facebook/esm2_t30_150M_UR50D',
|
19 |
+
'ESM2-650': 'facebook/esm2_t33_650M_UR50D',
|
20 |
+
'ESM2-3B': 'facebook/esm2_t36_3B_UR50D',
|
21 |
+
'ESM2-15B': 'facebook/esm2_t48_15B_UR50D',
|
22 |
+
}
|
23 |
|
24 |
+
model = AutoModelForMaskedLM.from_pretrained(model_dict['ESM2-8'], trust_remote_code=True)
|
25 |
+
tokenizer = model.tokenizer
|
26 |
+
```
|
27 |
|
28 |
Outputting attention maps (or the contact prediction head) is not natively possible with SDPA. You can still pass ```output_attentions``` to have attention calculated manually and returned.
|
29 |
Various other optimizations also make the base implementation slightly different than the one in transformers.
|
30 |
|
31 |
+
# FastESM2-650
|
32 |
+
|
33 |
+
## A faster half-precision version of ESM2-650 with FlashAttention2 and longer context
|
34 |
+
To enhance the weights with longer context and better fp16 support, we trained ESM2-650 50000 additional steps with a traditional MLM objective (20% masking) in fp16 mixed precision on [OMGprot50](https://huggingface.co/datasets/tattabio/OMG_prot50) up to sequence length of **2048**.
|
35 |
+
|
36 |
## Use with 🤗 transformers
|
37 |
+
|
38 |
+
### For working with embeddings
|
39 |
```python
|
40 |
import torch
|
41 |
+
from transformers import AutoModel, AutoTokenizer
|
42 |
|
43 |
model_path = 'Synthyra/FastESM2_650'
|
44 |
+
model = AutoModel.from_pretrained(model_path, torch_dtype=torch.float16, trust_remote_code=True).eval()
|
45 |
+
tokenizer = model.tokenizer
|
46 |
|
47 |
sequences = ['MPRTEIN', 'MSEQWENCE']
|
48 |
tokenized = tokenizer(sequences, padding=True, return_tensors='pt')
|
49 |
with torch.no_grad():
|
50 |
+
embeddings = model(**tokenized).last_hidden_state
|
51 |
|
52 |
print(embeddings.shape) # (1, 11, 1280)
|
53 |
```
|
54 |
|
55 |
+
### For working with sequence logits
|
56 |
+
```python
|
57 |
+
import torch
|
58 |
+
from transformers import AutoModelForMaskedLM, AutoTokenizer
|
59 |
+
|
60 |
+
model_path = 'Synthyra/FastESM2_650'
|
61 |
+
model = AutoModelForMaskedLM.from_pretrained(model_path, torch_dtype=torch.float16, trust_remote_code=True).eval()
|
62 |
+
tokenizer = model.tokenizer
|
63 |
+
|
64 |
+
sequences = ['MPRTEIN', 'MSEQWENCE']
|
65 |
+
tokenized = tokenizer(sequences, padding=True, return_tensors='pt')
|
66 |
+
with torch.no_grad():
|
67 |
+
logits = model(**tokenized).logits
|
68 |
+
|
69 |
+
print(logits.shape) # (1, 11, 33)
|
70 |
+
```
|
71 |
|
72 |
## Embed entire datasets with no new code
|
73 |
To embed a list of protein sequences **fast**, just call embed_dataset. Sequences are sorted to reduce padding tokens, so the initial progress bar estimation is usually much longer than the actual time.
|
|
|
95 |
sql_db_path='embeddings.db', # path to .db file of choice
|
96 |
)
|
97 |
```
|
98 |
+
|
99 |
## Model probes
|
100 |
We employ linear probing techniques on various PLMs and standard datasets, similar our previous [paper](https://www.biorxiv.org/content/10.1101/2024.07.30.605924v1), to assess the intrinsic correlation between pooled hidden states and valuable properties. FastESM performs very well.
|
101 |
|