CoolFace
Modelpublic

datasetsANDmodels/purpose-extraction

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face
0likes12downloads
README.md65 linesDownload Raw Back to root
1---2license: apache-2.03base_model: t5-large4tags:5- generated_from_trainer6model-index:7- name: purpose_extractor8  results: []9---10This model extracts purpose from intent.11 12# purpose_extractor13 14This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on an unknown dataset.15It achieves the following results on the evaluation set:16- Loss: 0.020317 18## Model description19 20More information needed21 22## Intended uses & limitations23 24More information needed25 26## Training and evaluation data27 28More information needed29 30## Training procedure31 32### Training hyperparameters33 34The following hyperparameters were used during training:35- learning_rate: 2e-0536- train_batch_size: 437- eval_batch_size: 438- seed: 4239- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0840- lr_scheduler_type: linear41- num_epochs: 1042 43### Training results44 45| Training Loss | Epoch | Step | Validation Loss |46|:-------------:|:-----:|:----:|:---------------:|47| 1.1179        | 1.0   | 109  | 0.6176          |48| 0.09          | 2.0   | 218  | 0.1340          |49| 0.0982        | 3.0   | 327  | 0.0781          |50| 0.0015        | 4.0   | 436  | 0.0522          |51| 0.3695        | 5.0   | 545  | 0.0406          |52| 0.0051        | 6.0   | 654  | 0.0310          |53| 0.0294        | 7.0   | 763  | 0.0251          |54| 0.0027        | 8.0   | 872  | 0.0228          |55| 0.015         | 9.0   | 981  | 0.0209          |56| 0.0303        | 10.0  | 1090 | 0.0203          |57 58 59### Framework versions60 61- Transformers 4.35.262- Pytorch 2.1.163- Datasets 2.15.064- Tokenizers 0.15.065