Instructions to use rcds/MiniLM-swiss_citation_extraction-de-fr-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use rcds/MiniLM-swiss_citation_extraction-de-fr-it with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="rcds/MiniLM-swiss_citation_extraction-de-fr-it")# Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("rcds/MiniLM-swiss_citation_extraction-de-fr-it") model = AutoModelForTokenClassification.from_pretrained("rcds/MiniLM-swiss_citation_extraction-de-fr-it", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "de_predict/_CITATION_f1": 0.9793131792857794, | |
| "de_predict/_CITATION_precision": 0.9852522282458881, | |
| "de_predict/_CITATION_recall": 0.9734453018610985, | |
| "de_predict/_LAW_f1": 0.9207842961099632, | |
| "de_predict/_LAW_precision": 0.8598544432559407, | |
| "de_predict/_LAW_recall": 0.9910077594333921, | |
| "de_predict/_accuracy_normalized": 0.9880353522464387, | |
| "de_predict/_loss": 0.035749662667512894, | |
| "de_predict/_macro-f1": 0.9504272924171073, | |
| "de_predict/_macro-precision": 0.9822265306472453, | |
| "de_predict/_macro-recall": 0.9232171398568052, | |
| "de_predict/_micro-f1": 0.9405898834091524, | |
| "de_predict/_micro-precision": 0.9849051246865093, | |
| "de_predict/_micro-recall": 0.9000908134288556, | |
| "de_predict/_runtime": 265.8036, | |
| "de_predict/_samples_per_second": 70.277, | |
| "de_predict/_steps_per_second": 0.549, | |
| "de_predict/_weighted-f1": 0.939658320951984, | |
| "de_predict/_weighted-precision": 0.9854977355183103, | |
| "de_predict/_weighted-recall": 0.9000908134288556, | |
| "epoch": 23.0, | |
| "fr_predict/_CITATION_f1": 0.9554686901203342, | |
| "fr_predict/_CITATION_precision": 0.9684586699813549, | |
| "fr_predict/_CITATION_recall": 0.9428225684465286, | |
| "fr_predict/_LAW_f1": 0.910095519316377, | |
| "fr_predict/_LAW_precision": 0.8366717393986756, | |
| "fr_predict/_LAW_recall": 0.9976459048553212, | |
| "fr_predict/_accuracy_normalized": 0.9830767480044869, | |
| "fr_predict/_loss": 0.04828711599111557, | |
| "fr_predict/_macro-f1": 0.9330080903677362, | |
| "fr_predict/_macro-precision": 0.9702342366509249, | |
| "fr_predict/_macro-recall": 0.9029739799827206, | |
| "fr_predict/_micro-f1": 0.920617324580396, | |
| "fr_predict/_micro-precision": 0.9842228065627199, | |
| "fr_predict/_micro-recall": 0.8647338279317974, | |
| "fr_predict/_runtime": 102.8006, | |
| "fr_predict/_samples_per_second": 74.863, | |
| "fr_predict/_steps_per_second": 0.593, | |
| "fr_predict/_weighted-f1": 0.9198669665372888, | |
| "fr_predict/_weighted-precision": 0.9861681830521788, | |
| "fr_predict/_weighted-recall": 0.8647338279317974, | |
| "it_predict/_CITATION_f1": 0.9703896103896105, | |
| "it_predict/_CITATION_precision": 0.9769874476987448, | |
| "it_predict/_CITATION_recall": 0.9638802889576883, | |
| "it_predict/_LAW_f1": 0.9099276791584483, | |
| "it_predict/_LAW_precision": 0.8422590068159689, | |
| "it_predict/_LAW_recall": 0.9894195024306548, | |
| "it_predict/_accuracy_normalized": 0.9892137683075134, | |
| "it_predict/_loss": 0.031956374645233154, | |
| "it_predict/_macro-f1": 0.9413484848298093, | |
| "it_predict/_macro-precision": 0.9766498956941716, | |
| "it_predict/_macro-recall": 0.9119834901073706, | |
| "it_predict/_micro-f1": 0.9311429570080392, | |
| "it_predict/_micro-precision": 0.9803127874885005, | |
| "it_predict/_micro-recall": 0.8866699950074888, | |
| "it_predict/_runtime": 26.6394, | |
| "it_predict/_samples_per_second": 69.146, | |
| "it_predict/_steps_per_second": 0.563, | |
| "it_predict/_weighted-f1": 0.929971077318579, | |
| "it_predict/_weighted-precision": 0.9813271971464931, | |
| "it_predict/_weighted-recall": 0.8866699950074888, | |
| "predict/_CITATION_f1": 0.973621340187501, | |
| "predict/_CITATION_precision": 0.981138340970977, | |
| "predict/_CITATION_recall": 0.9662186467837405, | |
| "predict/_LAW_f1": 0.9168199439712499, | |
| "predict/_LAW_precision": 0.8514980289093298, | |
| "predict/_LAW_recall": 0.9929968125536349, | |
| "predict/_accuracy_normalized": 0.986841752305624, | |
| "predict/_loss": 0.03871648758649826, | |
| "predict/_macro-f1": 0.9455976917351873, | |
| "predict/_macro-precision": 0.9796077296686877, | |
| "predict/_macro-recall": 0.9169959471957758, | |
| "predict/_micro-f1": 0.934344809828224, | |
| "predict/_micro-precision": 0.9844524443053164, | |
| "predict/_micro-recall": 0.8890909776278342, | |
| "predict/_runtime": 396.5316, | |
| "predict/_samples_per_second": 71.162, | |
| "predict/_steps_per_second": 0.557, | |
| "predict/_weighted-f1": 0.9333974918752409, | |
| "predict/_weighted-precision": 0.9854002360022739, | |
| "predict/_weighted-recall": 0.8890909776278342, | |
| "predict_samples": 28218, | |
| "train_CITATION_f1": 0.9814787122684838, | |
| "train_CITATION_precision": 0.9844650624611626, | |
| "train_CITATION_recall": 0.978510425318936, | |
| "train_LAW_f1": 0.9211885809936669, | |
| "train_LAW_precision": 0.8550646934923544, | |
| "train_LAW_recall": 0.9983966169255928, | |
| "train_accuracy_normalized": 0.9905667906095162, | |
| "train_loss": 0.026775669306516647, | |
| "train_macro-f1": 0.9515974221562933, | |
| "train_macro-precision": 0.9884535211222644, | |
| "train_macro-recall": 0.9202279237061262, | |
| "train_micro-f1": 0.9429566347321088, | |
| "train_micro-precision": 0.9909403074465961, | |
| "train_micro-recall": 0.8994053033460294, | |
| "train_runtime": 1297.076, | |
| "train_samples": 90101, | |
| "train_samples_per_second": 69.465, | |
| "train_steps_per_second": 0.543, | |
| "train_weighted-f1": 0.9419101385291693, | |
| "train_weighted-precision": 0.9916707876514618, | |
| "train_weighted-recall": 0.8994053033460294 | |
| } |