SimoneAstarita/Pride-large-try-sweep-20251010-105542-t00
08
Pride-large-try-sweep-20251010-105542-t00
Multilingual XLM-T (EN/IT/ES) binary classifier Task: LGBTQ+ reclamation vs non-reclamation on social media text.
Trial timestamp (UTC): 2025-10-10 10:55:42 Data case: es-itConfiguration (trial hyperparameters)
Model: Alibaba-NLP/gte-multilingual-base
Dev set results (summary)
Thresholds
- Default:
0.5 - Best global:
0.7000000000000001 - Best by language:
{ "it": 0.75, "es": 0.7000000000000001 }
Detailed evaluation
Classification report @ 0.5
precision recall f1-score support
no-recl (0) 0.9552 0.8730 0.9122 244
recl (1) 0.5694 0.8039 0.6667 51
accuracy 0.8610 295
macro avg 0.7623 0.8384 0.7894 295
weighted avg 0.8885 0.8610 0.8698 295Classification report @ best global threshold (t=0.70)
precision recall f1-score support
no-recl (0) 0.9545 0.9467 0.9506 244
recl (1) 0.7547 0.7843 0.7692 51
accuracy 0.9186 295
macro avg 0.8546 0.8655 0.8599 295
weighted avg 0.9200 0.9186 0.9193 295Classification report @ best per-language thresholds
precision recall f1-score support
no-recl (0) 0.9547 0.9508 0.9528 244
recl (1) 0.7692 0.7843 0.7767 51
accuracy 0.9220 295
macro avg 0.8620 0.8676 0.8647 295
weighted avg 0.9227 0.9220 0.9223 295Per-language metrics (at best-by-lang)
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
from transformers import AutoTokenizer, AutoModelForSequenceClassification, AutoConfig
import torch, numpy as np
repo = "SimoneAstarita/Pride-large-try-sweep-20251010-105542-t00"
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)))
### 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.
report_0.5.txt, report_best.txt — readable classification reports.
postprocessing.json — duplicate threshold info for external tools.