Text Classification
Transformers
Safetensors
English
Korean
Japanese
qev_decision_merged
feature-extraction
qev
dlp
typed-decision
japanese
bf16
custom_code
Instructions to use querypieai/QEV2.1-12B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use querypieai/QEV2.1-12B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="querypieai/QEV2.1-12B", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("querypieai/QEV2.1-12B", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download code/train_launcher.template.sh from querypieai/QEV2.1-12B: direct link, hf CLI and curl.
- Browser
- Download file 1.05 kB
-
https://huggingface.co/querypieai/QEV2.1-12B/resolve/main/code/train_launcher.template.sh
- Command line
-
hf download hf://querypieai/QEV2.1-12B/code/train_launcher.template.sh
-
curl -L -o train_launcher.template.sh https://huggingface.co/querypieai/QEV2.1-12B/resolve/main/code/train_launcher.template.sh
1.05 kB
| set -euo pipefail | |
| # Fill these paths on the training host. This template intentionally contains | |
| # no cluster path, GPU assignment, token or dataset record. | |
| PYTHON_BIN="${PYTHON_BIN:-python}" | |
| QEV_SOURCE="${QEV_SOURCE:?set QEV_SOURCE to the checked-out QEV training package}" | |
| TRAIN_JSONL="${TRAIN_JSONL:?set TRAIN_JSONL to a private multitask split}" | |
| DEV_JSONL="${DEV_JSONL:?set DEV_JSONL to a private development split}" | |
| OUT_DIR="${OUT_DIR:?set OUT_DIR to a new run directory}" | |
| BASE_MODEL="${BASE_MODEL:?set BASE_MODEL to the approved base lineage}" | |
| BASE_REVISION="${BASE_REVISION:?set BASE_REVISION to the recorded commit}" | |
| INIT_FROM="${INIT_FROM:?set INIT_FROM to the existing general-stage checkpoint}" | |
| export PYTHONPATH="$QEV_SOURCE/src${PYTHONPATH:+:$PYTHONPATH}" | |
| exec "$PYTHON_BIN" -u -m qev.train \ | |
| --data "$TRAIN_JSONL" --out "$OUT_DIR" \ | |
| --base "$BASE_MODEL" --base_revision "$BASE_REVISION" \ | |
| --init_from "$INIT_FROM" --eval_data "$DEV_JSONL" \ | |
| --epochs 1 --accum 8 --lr 5e-6 \ | |
| --dtype bf16 --checkpointing 1 | |