|
--- |
|
library_name: transformers |
|
tags: |
|
- indobert |
|
- indonlu |
|
- indobenchmark |
|
datasets: |
|
- fahrendrakhoirul/ecommerce-reviews-multilabel-dataset |
|
language: |
|
- id |
|
metrics: |
|
- f1 |
|
- precision |
|
- recall |
|
--- |
|
|
|
This model leverages IndoBERT for understanding language and a Long Short-Term Memory (LSTM) network to capture sequential information in customer reviews. It's designed for multi-label classification of e-commerce reviews, focusing on: |
|
|
|
- Produk (Product): Customer satisfaction with product quality, performance, and description accuracy. |
|
- Layanan Pelanggan (Customer Service): Interaction with sellers, their responsiveness, and complaint handling. |
|
- Pengiriman (Shipping/Delivery): Speed of delivery, item condition upon arrival, and timeliness. |
|
|
|
|
|
**How to import in PyTorch:** |
|
```python |
|
import torch.nn as nn |
|
from huggingface_hub import PyTorchModelHubMixin |
|
from transformers import AutoModelForSequenceClassification, AutoTokenizer |
|
|
|
class IndoBertLSTMEcommerceReview(nn.Module, PyTorchModelHubMixin): |
|
def __init__(self, bert): |
|
super().__init__() |
|
self.bert = bert |
|
self.lstm = nn.LSTM(bert.config.hidden_size, 128) |
|
self.linear = nn.Linear(128, 3) |
|
self.sigmoid = nn.Sigmoid() |
|
|
|
def forward(self, input_ids, attention_mask): |
|
outputs = self.bert(input_ids=input_ids, attention_mask=attention_mask) |
|
last_hidden_state = outputs.last_hidden_state |
|
lstm_out, _ = self.lstm(last_hidden_state) |
|
pooled = lstm_out[:, -1, :] |
|
logits = self.linear(pooled) |
|
probabilities = self.sigmoid(logits) |
|
return probabilities |
|
|
|
bert = AutoModelForSequenceClassification.from_pretrained("indobenchmark/indobert-base-p1", |
|
num_labels=3, |
|
problem_type="multi_label_classification") |
|
tokenizer = AutoTokenizer.from_pretrained("fahrendrakhoirul/indobert-lstm-finetuned-ecommerce-reviews") |
|
model = IndoBertLSTMEcommerceReview.from_pretrained("fahrendrakhoirul/indobert-lstm-finetuned-ecommerce-reviews", bert=bert) |
|
``` |
