README.md

---
license: apache-2.0
library_name: transformers
pipeline_tag: text-classification
language:
  - it
tags:
  - transformers
  - xlm-roberta
  - multilingual
  - social-media
  - text-classification
---
# it-no-bio-20251014-t14

**Slur reclamation binary classifier**  
Task: LGBTQ+ reclamation vs non-reclamation use of harmful words on social media text.

> Trial timestamp (UTC): 2025-10-14 10:43:41
>
> **Data case:** `it`

## Configuration (trial hyperparameters)

Model: Alibaba-NLP/gte-multilingual-base

| Hyperparameter | Value |
|---|---|
| LANGUAGES | it |
| LR | 3e-05 |
| EPOCHS | 3 |
| MAX_LENGTH | 256 |
| USE_BIO | False |
| USE_LANG_TOKEN | False |
| GATED_BIO | False |
| FOCAL_LOSS | True |
| FOCAL_GAMMA | 1.5 |
| USE_SAMPLER | True |
| R_DROP | True |
| R_KL_ALPHA | 1.0 |
| TEXT_NORMALIZE | True |

## Dev set results (summary)

| Metric | Value |
|---|---|
| f1_macro_dev_0.5 | 0.8676160051978992 |
| f1_weighted_dev_0.5 | 0.9129082823912861 |
| accuracy_dev_0.5 | 0.9079754601226994 |
| f1_macro_dev_best_global | 0.905011655011655 |
| f1_weighted_dev_best_global | 0.9400374676448295 |
| accuracy_dev_best_global | 0.9386503067484663 |
| f1_macro_dev_best_by_lang | 0.905011655011655 |
| f1_weighted_dev_best_by_lang | 0.9400374676448295 |
| accuracy_dev_best_by_lang | 0.9386503067484663 |
| default_threshold | 0.5 |
| best_threshold_global | 0.7000000000000001 |
| thresholds_by_lang | {"it": 0.7000000000000001} |

### Thresholds
- Default: `0.5`
- Best global: `0.7000000000000001`
- Best by language: `{
  "it": 0.7000000000000001
}`

## Detailed evaluation

### Classification report @ 0.5
```text
              precision    recall  f1-score   support

 no-recl (0)     0.9835    0.9015    0.9407       132
    recl (1)     0.6905    0.9355    0.7945        31

    accuracy                         0.9080       163
   macro avg     0.8370    0.9185    0.8676       163
weighted avg     0.9277    0.9080    0.9129       163
```

### Classification report @ best global threshold (t=0.70)
```text
              precision    recall  f1-score   support

 no-recl (0)     0.9766    0.9470    0.9615       132
    recl (1)     0.8000    0.9032    0.8485        31

    accuracy                         0.9387       163
   macro avg     0.8883    0.9251    0.9050       163
weighted avg     0.9430    0.9387    0.9400       163
```

### Classification report @ best per-language thresholds
```text
              precision    recall  f1-score   support

 no-recl (0)     0.9766    0.9470    0.9615       132
    recl (1)     0.8000    0.9032    0.8485        31

    accuracy                         0.9387       163
   macro avg     0.8883    0.9251    0.9050       163
weighted avg     0.9430    0.9387    0.9400       163
```


## Per-language metrics (at best-by-lang)

| lang | n | acc | f1_macro | f1_weighted | prec_macro | rec_macro | prec_weighted | rec_weighted |
|---|---:|---:|---:|---:|---:|---:|---:|---:|
| it | 163 | 0.9387 | 0.9050 | 0.9400 | 0.8883 | 0.9251 | 0.9430 | 0.9387 |


## Data
- Train/Dev: private multilingual splits with ~15% stratified Dev (by (lang,label)).
- Source: merged EN/IT/ES data with bios retained (ignored if unused by model).

## Usage
```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification, AutoConfig
import torch, numpy as np

repo = "SimoneAstarita/it-no-bio-20251014-t14"
tok = AutoTokenizer.from_pretrained(repo)
cfg = AutoConfig.from_pretrained(repo)
model = AutoModelForSequenceClassification.from_pretrained(repo)

texts = ["example text ..."]
langs = ["en"]

mode = "best_global"  # or "0.5", "by_lang"

enc = tok(texts, truncation=True, padding=True, max_length=256, return_tensors="pt")
with torch.no_grad():
    logits = model(**enc).logits
probs = torch.softmax(logits, dim=-1)[:, 1].cpu().numpy()

if mode == "0.5":
    th = 0.5
    preds = (probs >= th).astype(int)
elif mode == "best_global":
    th = getattr(cfg, "best_threshold_global", 0.5)
    preds = (probs >= th).astype(int)
elif mode == "by_lang":
    th_by_lang = getattr(cfg, "thresholds_by_lang", {})
    preds = np.zeros_like(probs, dtype=int)
    for lg in np.unique(langs):
        t = th_by_lang.get(lg, getattr(cfg, "best_threshold_global", 0.5))
        preds[np.array(langs) == lg] = (probs[np.array(langs) == lg] >= t).astype(int)
print(list(zip(texts, preds, probs)))
```

### Additional files
reports.json: all metrics (macro/weighted/accuracy) for @0.5, @best_global, and @best_by_lang.
config.json: stores thresholds: default_threshold, best_threshold_global, thresholds_by_lang.
postprocessing.json: duplicate threshold info for external tools.