Text Classification
Transformers
Safetensors
Russian
English
bert
tiny-bert
rubert-tiny2
binary-classification
jobs
vacancy-classification
it-classification
non-it-classification
Eval Results (legacy)
text-embeddings-inference
Instructions to use AndreiTolmachev/it-vs-nonit-roles-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AndreiTolmachev/it-vs-nonit-roles-tiny with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="AndreiTolmachev/it-vs-nonit-roles-tiny")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("AndreiTolmachev/it-vs-nonit-roles-tiny") model = AutoModelForSequenceClassification.from_pretrained("AndreiTolmachev/it-vs-nonit-roles-tiny") - Notebooks
- Google Colab
- Kaggle
| { | |
| "task": "it_vs_nonit_binary", | |
| "positive_definition": "is_it == 1 / role_category != 'Не IT'", | |
| "labels": [ | |
| "NonIT", | |
| "IT" | |
| ], | |
| "max_len": 384, | |
| "description_chars": 2000, | |
| "title_only": false, | |
| "base_model": "cointegrated/rubert-tiny2", | |
| "trained_on": "data/labeled/it_nonit_train_28_05_v2.csv", | |
| "rows": 27147, | |
| "positive_rows": 9836, | |
| "negative_rows": 17311, | |
| "best_epoch": 3, | |
| "best_threshold": 0.2934192717075348, | |
| "best_precision_at_threshold": 0.9089195979899497, | |
| "best_recall_at_threshold": 0.9803523035230353, | |
| "best_f1_at_threshold": 0.9432855280312906, | |
| "best_roc_auc": 0.9952188422538826, | |
| "target_recall": 0.98, | |
| "pos_weight": 1.760047846889952 | |
| } |