def _load_model_args(self, input_dir):
args = ModelArgs()
args.load(input_dir)
return args
import os
import sys
import torch
import torch.nn as nn
from torch.utils.data import DataLoader, Dataset, Subset
import torch.nn.functional as F
from src import deepSVDD
from src.deepSVDD import *
from src.utils.config import Config
from src.base.torchvision_dataset import TorchvisionDataset
from src.utils.config import Config
from src.base.base_net import BaseNet
from simpletransformers.language_representation import RepresentationModel
from simpletransformers.config.model_args import ModelArgs
model_args = ModelArgs(max_seq_length=156)
model = RepresentationModel(model_type="roberta",
model_name="seyonec/PubChem10M_SMILES_BPE_396_250",
use_cuda=False)
def mean_pool(model, sentences):
attn_mask = model._tokenize(sentences)['attention_mask'].numpy()
word_vectors = model.encode_sentences(sentences, combine_strategy=0)
return word_vectors
def smiles2txt(dataset):
''' reading the smiles from the csv file and saves them in txt file in order to get the
graph embendings from each smile'''