File size: 1,575 Bytes
9a6d76f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
from typing import Any, List, Dict
from pathlib import Path
import torch
from transformers import AutoModelForMaskedLM, AutoTokenizer
class EndpointHandler():
def __init__(self, path="."):
self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
self.tokenizer = AutoTokenizer.from_pretrained(path)
self.model = AutoModelForMaskedLM.from_pretrained(path).to(self.device)
def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
"""
Args:
data (:obj:):
includes the input data and the parameters for the inference.
Return:
A :obj:`list`:. The list contains the embeddings of the inference inputs
"""
inputs = data.get("inputs", data)
with torch.no_grad():
tokens = self.tokenizer(
inputs, padding=True, truncation=True, return_tensors='pt'
).to(self.device)
outputs = self.model(**tokens)
vecs = torch.max(
torch.log(
1 + torch.relu(outputs.logits)
) * tokens.attention_mask.unsqueeze(-1),
dim=1
)[0]
embeds = []
for vec in vecs:
# extract non-zero positions
cols = vec.nonzero().squeeze().cpu().tolist()
# extract the non-zero values
weights = vec[cols].cpu().tolist()
sparse = {
"indices": cols,
"values": weights,
}
embeds.append(sparse)
return embeds |