pairwise_model.py

import numpy as np
import torch
import torch.nn as nn
from torch.utils.data import Dataset, DataLoader
from transformers import AutoModel, AutoConfig, AutoTokenizer
import pandas as pd
from optimum.intel import OVModelForQuestionAnswering
import openvino.inference_engine as ie
import os
import gradio as gr

AUTH_TOKEN = "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"

tokenizer = AutoTokenizer.from_pretrained('nguyenvulebinh/vi-mrc-base',
                                          use_auth_token=AUTH_TOKEN)
pad_token_id = tokenizer.pad_token_id

# Load the model
model_xml = "openvino_stage1/stage1.xml"
model_bin = "openvino_stage1/stage1.bin"
# Create an Inference Engine object
ie_core = ie.IECore()
# Read the IR files"
net = ie_core.read_network(model=model_xml, weights=model_bin)

class PairwiseModel_modify(nn.Module):
    def __init__(self, model_name, max_length=384, batch_size=16, device="cpu"):
        super(PairwiseModel_modify, self).__init__()
        self.max_length = max_length
        self.batch_size = batch_size
        self.device = device
        # self.model = AutoModel.from_pretrained(model_name , use_auth_token=AUTH_TOKEN)
        self.config = AutoConfig.from_pretrained(model_name, use_auth_token=AUTH_TOKEN)
        self.fc = nn.Linear(768, 1).to(self.device)

    def forward(self, ids, masks):
        # Export the model to ONNX format
        input_feed = {"input_ids": ids.cpu().numpy().astype(np.int64), "attention_mask": masks.cpu().numpy().astype(np.int64)}
        # Specify the input shapes (batch_size, max_sequence_length)
        input_shapes = {"input_ids": ids.shape, "attention_mask": masks.shape}

        # Set the input shapes in the network
        net.reshape(input_shapes)

        # Load the network with the specified input shapes
        exec_net = ie_core.load_network(network=net, device_name="CPU")
        outputs = exec_net.infer(input_feed)

        # Get the output tensor and apply the linear layer
        out = torch.from_numpy(outputs["output"]).to(self.device)
        out = out[:, 0]
        return out

    def stage1_ranking(self, question, texts):
        tmp = pd.DataFrame()
        tmp["text"] = [" ".join(x.split()) for x in texts]
        tmp["question"] = question
        valid_dataset = SiameseDatasetStage1(tmp, tokenizer, self.max_length, is_test=True)
        valid_loader = DataLoader(valid_dataset, batch_size=self.batch_size, collate_fn=collate_fn,
                                  num_workers=0, shuffle=False, pin_memory=True)
        preds = []
        with torch.no_grad():
            bar = enumerate(valid_loader)
            for step, data in bar:
                ids = data["ids"].to(self.device)
                masks = data["masks"].to(self.device)
                preds.append(torch.sigmoid(self(ids, masks)).view(-1))
            preds = torch.concat(preds)
        return preds.cpu().numpy()
    

class SiameseDatasetStage1(Dataset):

    def __init__(self, df, tokenizer, max_length, is_test=False):
        self.df = df
        self.max_length = max_length
        self.tokenizer = tokenizer
        self.is_test = is_test
        self.content1 = tokenizer.batch_encode_plus(list(df.question.values), max_length=max_length, truncation=True)[
            "input_ids"]
        self.content2 = tokenizer.batch_encode_plus(list(df.text.values), max_length=max_length, truncation=True)[
            "input_ids"]
        if not self.is_test:
            self.targets = self.df.label

    def __len__(self):
        return len(self.df)

    def __getitem__(self, index):
        return {
            'ids1': torch.tensor(self.content1[index], dtype=torch.long),
            'ids2': torch.tensor(self.content2[index][1:], dtype=torch.long),
            'target': torch.tensor(0) if self.is_test else torch.tensor(self.targets[index], dtype=torch.float)
        }


class SiameseDatasetStage2(Dataset):

    def __init__(self, df, tokenizer, max_length, is_test=False):
        self.df = df
        self.max_length = max_length
        self.tokenizer = tokenizer
        self.is_test = is_test
        self.df["content1"] = self.df.apply(lambda row: row.question + f" {tokenizer.sep_token} " + row.answer, axis=1)
        self.df["content2"] = self.df.apply(lambda row: row.title + f" {tokenizer.sep_token} " + row.candidate, axis=1)
        self.content1 = tokenizer.batch_encode_plus(list(df.content1.values), max_length=max_length, truncation=True)[
            "input_ids"]
        self.content2 = tokenizer.batch_encode_plus(list(df.content2.values), max_length=max_length, truncation=True)[
            "input_ids"]
        if not self.is_test:
            self.targets = self.df.label

    def __len__(self):
        return len(self.df)

    def __getitem__(self, index):
        return {
            'ids1': torch.tensor(self.content1[index], dtype=torch.long),
            'ids2': torch.tensor(self.content2[index][1:], dtype=torch.long),
            'target': torch.tensor(0) if self.is_test else torch.tensor(self.targets[index], dtype=torch.float)
        }


def collate_fn(batch):
    ids = [torch.cat([x["ids1"], x["ids2"]]) for x in batch]
    targets = [x["target"] for x in batch]
    max_len = np.max([len(x) for x in ids])
    masks = []
    for i in range(len(ids)):
        if len(ids[i]) < max_len:
            ids[i] = torch.cat((ids[i], torch.tensor([pad_token_id, ] * (max_len - len(ids[i])), dtype=torch.long)))
        masks.append(ids[i] != pad_token_id)
    # print(tokenizer.decode(ids[0]))
    outputs = {
        "ids": torch.vstack(ids),
        "masks": torch.vstack(masks),
        "target": torch.vstack(targets).view(-1)
    }
    return outputs