metadata
library_name: setfit
tags:
- setfit
- absa
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
metrics:
- accuracy
widget:
- text: >-
kekurangan nya yaitu jaringan saya udah bagus:Terima kasih untuk supercell
telah membuat game baru. Tapi ada kekurangan nya yaitu jaringan saya udah
bagus tapi ketika saya main dia keluar padahal kalau saya login coc sm
clash Royale masih bagus. Di harapkan kedepannya server nya makin bagus.
- text: >-
sinyal selalu bermasalah,:sinyal selalu bermasalah, login kadang macet dan
saat masuk game sering eror
- text: >-
keren dan lucu² karakter nya😍.:Mekanik game nya simple, gameplay nya juga
asik, seru dan lumayan gampang. Cuma perlu internet yg sinyalnya kuat dan
cepet kalo MW lancar mainin nya, entah sinyal dari provider ku yg lemah
atau karena gamenya yg masi baru dan masi ada bug, itu gak terlalu menarik
buat dibahas. Overall gamenya bagus, keren dan lucu² karakter nya😍.
Thanks supercell.
- text: >-
tapi entah kenapa layar gameku kayak ngefrezee ketika:Overall gamenya
bagus tapi entah kenapa layar gameku kayak ngefrezee ketika mau coba
sambung dengan supercell id. Tolong di perbaiki ya tim supercell
- text: >-
banyak yang mengeluh game ini sering frame:Game yang seru!! Tapi, untuk
persediaan peti tolong diperbanyak lagi. Oh iya, banyak yang mengeluh game
ini sering frame drop atau server jelek. Entahlah, mungkin ada yang salah
dengan hp kalian, padahal di hp kentang saya yang sudah berusia 5 tahun
masih berjalan lancar-lancar saja. Dengan RAM 3 GB dan Snapdragon 636,
masih kuat buat jalanin game ini. Server juga tidak ada masalah, main 1
jam bahkan lebih tetap aman-aman saja.
pipeline_tag: text-classification
inference: false
SetFit Polarity Model
This is a SetFit model that can be used for Aspect Based Sentiment Analysis (ABSA). A LogisticRegression instance is used for classification. In particular, this model is in charge of classifying aspect polarities.
The model has been trained using an efficient few-shot learning technique that involves:
- Fine-tuning a Sentence Transformer with contrastive learning.
- Training a classification head with features from the fine-tuned Sentence Transformer.
This model was trained within the context of a larger system for ABSA, which looks like so:
- Use a spaCy model to select possible aspect span candidates.
- Use a SetFit model to filter these possible aspect span candidates.
- Use this SetFit model to classify the filtered aspect span candidates.
Model Details
Model Description
- Model Type: SetFit
- Classification head: a LogisticRegression instance
- spaCy Model: id_core_news_trf
- SetFitABSA Aspect Model: Funnyworld1412/ABSA_game_squad_busters-aspect
- SetFitABSA Polarity Model: Funnyworld1412/ABSA_game_squad_busters-polarity
- Maximum Sequence Length: 8192 tokens
- Number of Classes: 2 classes
Model Sources
- Repository: SetFit on GitHub
- Paper: Efficient Few-Shot Learning Without Prompts
- Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts
Model Labels
Label | Examples |
---|---|
Negative |
|
Positive |
|
Uses
Direct Use for Inference
First install the SetFit library:
pip install setfit
Then you can load this model and run inference.
from setfit import AbsaModel
# Download from the 🤗 Hub
model = AbsaModel.from_pretrained(
"Funnyworld1412/ABSA_game_squad_busters-aspect",
"Funnyworld1412/ABSA_game_squad_busters-polarity",
)
# Run inference
preds = model("The food was great, but the venue is just way too busy.")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 11 | 39.7490 | 94 |
Label | Training Sample Count |
---|---|
konflik | 0 |
negatif | 0 |
netral | 0 |
positif | 0 |
Training Hyperparameters
- batch_size: (4, 4)
- num_epochs: (1, 1)
- max_steps: -1
- sampling_strategy: oversampling
- num_iterations: 10
- body_learning_rate: (2e-05, 1e-05)
- head_learning_rate: 0.01
- loss: CosineSimilarityLoss
- distance_metric: cosine_distance
- margin: 0.25
- end_to_end: False
- use_amp: False
- warmup_proportion: 0.1
- seed: 42
- eval_max_steps: -1
- load_best_model_at_end: False
Training Results
Epoch | Step | Training Loss | Validation Loss |
---|---|---|---|
0.0004 | 1 | 0.1468 | - |
0.0198 | 50 | 0.2275 | - |
0.0395 | 100 | 0.1824 | - |
0.0593 | 150 | 0.1943 | - |
0.0791 | 200 | 0.0063 | - |
0.0988 | 250 | 0.2251 | - |
0.1186 | 300 | 0.0068 | - |
0.1383 | 350 | 0.0046 | - |
0.1581 | 400 | 0.0015 | - |
0.1779 | 450 | 0.0014 | - |
0.1976 | 500 | 0.0018 | - |
0.2174 | 550 | 0.2301 | - |
0.2372 | 600 | 0.0011 | - |
0.2569 | 650 | 0.0051 | - |
0.2767 | 700 | 0.0015 | - |
0.2964 | 750 | 0.0016 | - |
0.3162 | 800 | 0.0007 | - |
0.3360 | 850 | 0.0027 | - |
0.3557 | 900 | 0.0014 | - |
0.3755 | 950 | 0.0077 | - |
0.3953 | 1000 | 0.001 | - |
0.4150 | 1050 | 0.0006 | - |
0.4348 | 1100 | 0.0009 | - |
0.4545 | 1150 | 0.1986 | - |
0.4743 | 1200 | 0.0004 | - |
0.4941 | 1250 | 0.0008 | - |
0.5138 | 1300 | 0.0008 | - |
0.5336 | 1350 | 0.0011 | - |
0.5534 | 1400 | 0.0088 | - |
0.5731 | 1450 | 0.001 | - |
0.5929 | 1500 | 0.0025 | - |
0.6126 | 1550 | 0.0006 | - |
0.6324 | 1600 | 0.0005 | - |
0.6522 | 1650 | 0.0006 | - |
0.6719 | 1700 | 0.0024 | - |
0.6917 | 1750 | 0.0725 | - |
0.7115 | 1800 | 0.1236 | - |
0.7312 | 1850 | 0.0006 | - |
0.7510 | 1900 | 0.001 | - |
0.7708 | 1950 | 0.0003 | - |
0.7905 | 2000 | 0.0003 | - |
0.8103 | 2050 | 0.0004 | - |
0.8300 | 2100 | 0.0004 | - |
0.8498 | 2150 | 0.0005 | - |
0.8696 | 2200 | 0.0003 | - |
0.8893 | 2250 | 0.0005 | - |
0.9091 | 2300 | 0.0003 | - |
0.9289 | 2350 | 0.0004 | - |
0.9486 | 2400 | 0.0005 | - |
0.9684 | 2450 | 0.0006 | - |
0.9881 | 2500 | 0.0007 | - |
Framework Versions
- Python: 3.10.13
- SetFit: 1.0.3
- Sentence Transformers: 3.0.1
- spaCy: 3.7.5
- Transformers: 4.36.2
- PyTorch: 2.1.2
- Datasets: 2.19.2
- Tokenizers: 0.15.2
Citation
BibTeX
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}