lockR/sravni_ru_gazprombank_classification
Viewer • Updated • 703 • 13
How to use lockR/xlm-roberta-finance-multi-label-classification with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-classification", model="lockR/xlm-roberta-finance-multi-label-classification") # Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("lockR/xlm-roberta-finance-multi-label-classification")
model = AutoModelForSequenceClassification.from_pretrained("lockR/xlm-roberta-finance-multi-label-classification", device_map="auto")Used in project https://github.com/L0ckR/LCT_hackaton
For ЛЦТ Хакатон
This model trained by transfomers Trainer using this params:
learning_rate=2e-5,
per_device_train_batch_size=3,
per_device_eval_batch_size=3,
num_train_epochs=50,
weight_decay=0.01,
eval_strategy="epoch",
save_strategy="epoch",
load_best_model_at_end=True,
lr_scheduler_type="linear",
warmup_ratio=0.1,
metric_for_best_model="f1",
greater_is_better=True
Base model
FacebookAI/xlm-roberta-base