import os, sys
import torch
from .models.embedding import FullyConnectedEmbed, SkipLSTM
from .models.contact import ContactCNN
from .models.interaction import ModelInteraction
def build_lm_1(state_dict_path):
"""
:meta private:
"""
model = SkipLSTM(21, 100, 1024, 3)
state_dict = torch.load(state_dict_path)
model.load_state_dict(state_dict)
model.eval()
return model
def build_human_1(state_dict_path):
"""
:meta private:
"""
embModel = FullyConnectedEmbed(6165, 100, 0.5)
conModel = ContactCNN(100, 50, 7)
model = ModelInteraction(embModel, conModel, use_W=True, pool_size=9)
state_dict = torch.load(state_dict_path)
model.load_state_dict(state_dict)
model.eval()
return model
VALID_MODELS = {
"lm_v1": build_lm_1,
"human_v1": build_human_1
}
[docs]def get_state_dict(version="human_v1", verbose=True):
"""
Download a pre-trained model if not already exists on local device.
:param version: Version of trained model to download [default: human_1]
:type version: str
:param verbose: Print model download status on stdout [default: True]
:type verbose: bool
:return: Path to state dictionary for pre-trained language model
:rtype: str
"""
state_dict_basename = f"dscript_{version}.pt"
state_dict_basedir = os.path.dirname(os.path.realpath(__file__))
state_dict_fullname = f"{state_dict_basedir}/{state_dict_basename}"
state_dict_url = f"http://cb.csail.mit.edu/cb/dscript/data/models/{state_dict_basename}"
if not os.path.exists(state_dict_fullname):
try:
import urllib.request
import shutil
if verbose: print(f"Downloading model {version} from {state_dict_url}...")
with urllib.request.urlopen(state_dict_url) as response, open(state_dict_fullname, 'wb') as out_file:
shutil.copyfileobj(response, out_file)
except Exception as e:
print("Unable to download model - {}".format(e))
sys.exit(1)
return state_dict_fullname
[docs]def get_pretrained(version="human_v1"):
"""
Get pre-trained model object.
Currently Available Models
==========================
See the `documentation <https://d-script.readthedocs.io/en/main/data.html#trained-models>`_ for most up-to-date list.
- ``lm_v1`` - Language model from `Bepler & Berger <https://github.com/tbepler/protein-sequence-embedding-iclr2019>`_.
- ``human_v1`` - Human trained model from D-SCRIPT manuscript.
Default: ``human_v1``
:param version: Version of pre-trained model to get
:type version: str
:return: Pre-trained model
:rtype: dscript.models.*
"""
if not version in VALID_MODELS:
raise ValueError("Model {} does not exist".format(version))
state_dict_path = get_state_dict(version)
return VALID_MODELS[version](state_dict_path)