Text Classification
Transformers
Safetensors
English
deberta-v2
sentiment-analysis
imdb
deberta-v3
binary-classification
Eval Results (legacy)
text-embeddings-inference
Instructions to use retaj249/imdb-deberta-v3-sentiment with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use retaj249/imdb-deberta-v3-sentiment with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="retaj249/imdb-deberta-v3-sentiment")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("retaj249/imdb-deberta-v3-sentiment") model = AutoModelForSequenceClassification.from_pretrained("retaj249/imdb-deberta-v3-sentiment", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "schema_version": 1, | |
| "task": "sentiment-analysis", | |
| "dataset": "IMDB", | |
| "base_model": "microsoft/deberta-v3-base", | |
| "model_selection": { | |
| "candidates_screened": 5, | |
| "selected_run": "single_seed_42", | |
| "seed": 42, | |
| "ensemble_weights": [1.0], | |
| "use_tta": false | |
| }, | |
| "train_config": { | |
| "learning_rate": 0.000016913198431430903, | |
| "num_train_epochs": 3, | |
| "effective_batch_size": 32, | |
| "weight_decay": 0.01, | |
| "warmup_ratio": 0.05, | |
| "lr_scheduler_type": "cosine", | |
| "classifier_dropout": 0.2, | |
| "max_grad_norm": 1.0, | |
| "use_llrd": false | |
| }, | |
| "tokenizer": { | |
| "source": "exported fine-tuned checkpoint", | |
| "use_fast": false | |
| }, | |
| "num_labels": 2, | |
| "id2label": { | |
| "0": "negative", | |
| "1": "positive" | |
| }, | |
| "label2id": { | |
| "negative": 0, | |
| "positive": 1 | |
| }, | |
| "positive_class_id": 1, | |
| "preprocessing": "minimal_normalize_v1", | |
| "truncation_strategy": "head_tail", | |
| "max_length": 384, | |
| "padding": "dynamic", | |
| "pad_to_multiple_of": 8, | |
| "decision_threshold": 0.477, | |
| "inference_batch_size": 8, | |
| "probability_function": "softmax_float32", | |
| "validation_metrics": { | |
| "accuracy": 0.9556630390971382, | |
| "f1": 0.9562101910828026, | |
| "threshold": 0.477 | |
| }, | |
| "final_seed_validation_metrics": { | |
| "accuracy": 0.954453849254333, | |
| "precision": 0.948535233570863, | |
| "recall": 0.9614767255216693, | |
| "f1": 0.9549621363092866, | |
| "macro_f1": 0.954448047363775, | |
| "log_loss": 0.16217701312185426, | |
| "roc_auc": 0.990061801002073, | |
| "best_epoch": 2.0 | |
| }, | |
| "test_metrics": { | |
| "accuracy": 0.96088, | |
| "precision": 0.9507119386637459, | |
| "recall": 0.97216, | |
| "f1": 0.9613163515544656, | |
| "macro_f1": 0.9608750218003739, | |
| "roc_auc": 0.992435184, | |
| "log_loss": 0.1306040386149452, | |
| "confusion_matrix": [ | |
| [11870, 630], | |
| [348, 12152] | |
| ], | |
| "examples": 25000 | |
| }, | |
| "frozen_config_fingerprint": "db302e66a1def80c7493849f7bc10abe64db0a2073d8dc9fa37efc6f4ae0cabf", | |
| "artifact_provenance": { | |
| "run_fingerprint": "aa9b5ed91723e614419119d81578ff5f4c837322f7bfe9a5ddc8b37a7c6dbe8e", | |
| "config_sha256": "2cb2f01576072d94ecc9445c48b1de53cd2fc556a444c1a0f3e5086318b95d62", | |
| "model_sha256": "e84d203742335db3bef1e7ff1e4b04aeb5cf4811b06e132a10fd96a1ab747ddd", | |
| "tokenizer_spm_sha256": "c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd", | |
| "validation_predictions_sha256": "c0d78a423c518b84cdb2e0e2a54350cab8e5aa2b85374150d1044a150f4d68ef", | |
| "training_runtime": { | |
| "torch": "2.11.0+cu128", | |
| "transformers": "4.57.3", | |
| "sentencepiece": "0.2.2" | |
| } | |
| } | |
| } | |