Text Classification
Transformers
ONNX
Safetensors
llama
safety
content-moderation
multi-label
multilingual
tiny-language-model
from-scratch
distilled
text-embeddings-inference
Instructions to use KucLab/Censer.8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use KucLab/Censer.8 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="KucLab/Censer.8")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("KucLab/Censer.8") model = AutoModelForSequenceClassification.from_pretrained("KucLab/Censer.8", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download reports/comparison.json from KucLab/Censer.8: direct link, hf CLI and curl.
- Browser
- Download file 6.08 kB
-
https://huggingface.co/KucLab/Censer.8/resolve/main/reports/comparison.json
- Command line
-
hf download hf://KucLab/Censer.8/reports/comparison.json
-
curl -L -o comparison.json https://huggingface.co/KucLab/Censer.8/resolve/main/reports/comparison.json
6.08 kB
| { | |
| "device": "cuda", | |
| "gpu": "AMD Radeon Graphics", | |
| "n_test": 6496, | |
| "models": [ | |
| { | |
| "repo": "KucLab/Censer.8 (this work)", | |
| "params": 903552, | |
| "checkpoint_mb": 3.62, | |
| "latency_ms": 1.3683213100011926, | |
| "throughput_s": 1843.306733690932, | |
| "vram_mb": 105.291776, | |
| "metrics": { | |
| "accuracy": 0.8568349753694581, | |
| "precision": 0.8622095393395842, | |
| "recall": 0.7813077207240487, | |
| "f1": 0.8197674418604651, | |
| "tp": 2115, | |
| "fp": 338, | |
| "tn": 3451, | |
| "fn": 592, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "KucLab/Censer.5 (prev)", | |
| "params": 902272, | |
| "checkpoint_mb": 3.62, | |
| "latency_ms": 1.2949280750763137, | |
| "throughput_s": 1874.5434140690643, | |
| "vram_mb": 96.704, | |
| "metrics": { | |
| "accuracy": 0.8763854679802956, | |
| "precision": 0.8644716692189893, | |
| "recall": 0.8341337273734761, | |
| "f1": 0.8490317728896409, | |
| "tp": 2258, | |
| "fp": 354, | |
| "tn": 3435, | |
| "fn": 449, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "KucLab/Censer.4 (old)", | |
| "params": 902272, | |
| "checkpoint_mb": 3.62, | |
| "latency_ms": 1.3153410800441634, | |
| "throughput_s": 1817.5694664505695, | |
| "vram_mb": 95.030784, | |
| "metrics": { | |
| "accuracy": 0.874384236453202, | |
| "precision": 0.868042039704165, | |
| "recall": 0.8237901736239379, | |
| "f1": 0.8453373768006066, | |
| "tp": 2230, | |
| "fp": 339, | |
| "tn": 3450, | |
| "fn": 477, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "KucLab/Censer.3 (old)", | |
| "params": 14424000, | |
| "checkpoint_mb": 57.71, | |
| "latency_ms": 4.491717959899688, | |
| "throughput_s": 272.4166469119276, | |
| "vram_mb": 239.994368, | |
| "metrics": { | |
| "accuracy": 0.8574507389162561, | |
| "precision": 0.8552054248105305, | |
| "recall": 0.7920206871074991, | |
| "f1": 0.8224012274645186, | |
| "tp": 2144, | |
| "fp": 363, | |
| "tn": 3426, | |
| "fn": 563, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "KucLab/Censer.2 (old)", | |
| "params": 7343872, | |
| "checkpoint_mb": 29.39, | |
| "latency_ms": 2.7238355050212704, | |
| "throughput_s": 548.6086914497098, | |
| "vram_mb": 170.506752, | |
| "metrics": { | |
| "accuracy": 0.8625307881773399, | |
| "precision": 0.8610668789808917, | |
| "recall": 0.7990395271518286, | |
| "f1": 0.8288944242191991, | |
| "tp": 2163, | |
| "fp": 349, | |
| "tn": 3440, | |
| "fn": 544, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "KucLab/Censer.1 (baseline)", | |
| "params": 1115520, | |
| "checkpoint_mb": 4.47, | |
| "latency_ms": 1.722235180059215, | |
| "throughput_s": 1681.7878610892315, | |
| "vram_mb": 96.736768, | |
| "metrics": { | |
| "accuracy": 0.8648399014778325, | |
| "precision": 0.8478508938759984, | |
| "recall": 0.8234207609900258, | |
| "f1": 0.8354572713643178, | |
| "tp": 2229, | |
| "fp": 400, | |
| "tn": 3389, | |
| "fn": 478, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "keyword baseline", | |
| "params": 0, | |
| "checkpoint_mb": 0.0, | |
| "latency_ms": 0.03348291794806168, | |
| "throughput_s": 29865.975287792673, | |
| "vram_mb": 0.0, | |
| "metrics": { | |
| "accuracy": 0.5681958128078818, | |
| "precision": 0.37371134020618557, | |
| "recall": 0.05356483191725157, | |
| "f1": 0.09369951534733441, | |
| "tp": 145, | |
| "fp": 243, | |
| "tn": 3546, | |
| "fn": 2562, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "martin-ha/toxic-comment-model", | |
| "params": 66955010, | |
| "checkpoint_mb": 0.0, | |
| "latency_ms": 7.7855301699310075, | |
| "throughput_s": 129.49495130867692, | |
| "vram_mb": 539.696128, | |
| "metrics": { | |
| "accuracy": 0.7438423645320197, | |
| "precision": 0.9610963748894783, | |
| "recall": 0.40155153306243074, | |
| "f1": 0.566440854611777, | |
| "tp": 1087, | |
| "fp": 44, | |
| "tn": 3745, | |
| "fn": 1620, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "unitary/toxic-bert", | |
| "params": 109486854, | |
| "checkpoint_mb": 0.0, | |
| "latency_ms": 14.037243769998895, | |
| "throughput_s": 67.84813601324072, | |
| "vram_mb": 708.03712, | |
| "metrics": { | |
| "accuracy": 0.7944889162561576, | |
| "precision": 0.9654002713704206, | |
| "recall": 0.5256741780568895, | |
| "f1": 0.6806983975125568, | |
| "tp": 1423, | |
| "fp": 51, | |
| "tn": 3738, | |
| "fn": 1284, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "unitary/unbiased-toxic-roberta", | |
| "params": 124657936, | |
| "checkpoint_mb": 0.0, | |
| "latency_ms": 14.808327034988906, | |
| "throughput_s": 65.45585524338709, | |
| "vram_mb": 770.71872, | |
| "metrics": { | |
| "accuracy": 0.8218903940886699, | |
| "precision": 0.9398410896708286, | |
| "recall": 0.6117473217584042, | |
| "f1": 0.7411053927053033, | |
| "tp": 1656, | |
| "fp": 106, | |
| "tn": 3683, | |
| "fn": 1051, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "Hate-speech-CNERG/dehatebert-mono-english", | |
| "params": 167357954, | |
| "checkpoint_mb": 0.0, | |
| "latency_ms": 14.519948939996539, | |
| "throughput_s": 66.2599401828871, | |
| "vram_mb": 923.411968, | |
| "metrics": { | |
| "accuracy": 0.635929802955665, | |
| "precision": 0.875, | |
| "recall": 0.14739564093091984, | |
| "f1": 0.2522921277268416, | |
| "tp": 399, | |
| "fp": 57, | |
| "tn": 3732, | |
| "fn": 2308, | |
| "total": 6496 | |
| } | |
| }, | |
| { | |
| "repo": "Hate-speech-CNERG/bert-base-uncased-hatexplain", | |
| "params": 109484547, | |
| "checkpoint_mb": 0.0, | |
| "latency_ms": 14.21610706995125, | |
| "throughput_s": 67.52212637769556, | |
| "vram_mb": 691.767296, | |
| "metrics": { | |
| "accuracy": 0.7039716748768473, | |
| "precision": 0.7606382978723404, | |
| "recall": 0.4226080531954193, | |
| "f1": 0.5433388743766325, | |
| "tp": 1144, | |
| "fp": 360, | |
| "tn": 3429, | |
| "fn": 1563, | |
| "total": 6496 | |
| } | |
| } | |
| ] | |
| } |