Qwen2.5-3B — LoRA adapter for Marketplace Review Category Classification

Model Details

This repository contains a LoRA adapter trained to classify marketplace customer reviews into product categories. The adapter was fine-tuned on a downstream sequence-classification task and is intended to be used with the Qwen/Qwen2.5-3B base model using the PEFT library.

Model Description

  • Adapter type: LoRA (PEFT)

  • Base model: Qwen/Qwen2.5-3B

  • Task: Multiclass text classification (product category of marketplace reviews)

  • Labels (classes):

    • бытовая техника (home appliances)
    • обувь (shoes)
    • одежда (clothing)
    • посуда (tableware / kitchenware)
    • текстиль (textiles)
    • товары для детей (kids / baby products)
    • украшения и аксессуары (jewelry & accessories)
    • электроника (electronics)
    • нет товара (NO_PRODUCT)
  • Developed by: [More Information Needed]

  • Funded by [optional]: [More Information Needed]

  • Shared by [optional]: [More Information Needed]

  • Model type: [More Information Needed]

  • Language(s) (NLP): [More Information Needed]

  • License: [More Information Needed]

  • Finetuned from model [optional]: [More Information Needed]

Model Sources [optional]

  • Repository: [More Information Needed]
  • Paper [optional]: [More Information Needed]
  • Demo [optional]: [More Information Needed]

How to load (inference)

Install prerequisites:

pip install transformers peft safetensors huggingface_hub
# If you want 8-bit:
pip install bitsandbytes

Training Details

Training Data

LORA_R = 64

LORA_ALPHA = 128

LORA_DROPOUT = 0.10

NUM_EPOCHS = 5

PER_DEVICE_BATCH_SIZE = 1

GRAD_ACCUM_STEPS = 8

LEARNING_RATE = 2e-4

WEIGHT_DECAY = 0.01

WARMUP_RATIO = 0.03

MAX_SEQ_LENGTH = 256

SEED = 42

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

  • Hardware Type: [More Information Needed]
  • Hours used: [More Information Needed]
  • Cloud Provider: [More Information Needed]
  • Compute Region: [More Information Needed]
  • Carbon Emitted: [More Information Needed]

Framework versions

  • PEFT 0.15.2
Downloads last month
1
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for DikovAlexandr/Qwen2.5-3B-LoAR-Classifier

Base model

Qwen/Qwen2.5-3B
Adapter
(422)
this model

Paper for DikovAlexandr/Qwen2.5-3B-LoAR-Classifier