Instructions to use heitorrosa/logun-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use heitorrosa/logun-base with PEFT:
Task type is invalid.
- Transformers
How to use heitorrosa/logun-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="heitorrosa/logun-base")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("heitorrosa/logun-base") model = AutoModelForMaskedLM.from_pretrained("heitorrosa/logun-base", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Fix model description and dataset reference
Browse filesThe model card states the model was trained on an 'unknown dataset' and has 'More information needed' placeholders, but the YAML frontmatter correctly identifies the dataset as `heitorrosa/cvm-corpus`. I updated the description to reflect this and removed the generic placeholders to make the card informative.
---
*Opened automatically by [BananaMindBot](https://huggingface.co/BananaMindBot). Close it if it is not useful - no hard feelings.*
README.md
CHANGED
|
@@ -15,26 +15,23 @@ datasets:
|
|
| 15 |
- heitorrosa/cvm-corpus
|
| 16 |
---
|
| 17 |
|
| 18 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 19 |
-
should probably proofread and complete it, then remove this comment. -->
|
| 20 |
-
|
| 21 |
# logun-base
|
| 22 |
|
| 23 |
-
This model is a fine-tuned version of [Itau-Unibanco/NorBERTo-base](https://huggingface.co/Itau-Unibanco/NorBERTo-base) on
|
| 24 |
It achieves the following results on the evaluation set:
|
| 25 |
- Loss: 0.4919
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
| 29 |
-
|
| 30 |
|
| 31 |
## Intended uses & limitations
|
| 32 |
|
| 33 |
-
|
| 34 |
|
| 35 |
## Training and evaluation data
|
| 36 |
|
| 37 |
-
|
| 38 |
|
| 39 |
## Training procedure
|
| 40 |
|
|
@@ -83,4 +80,4 @@ The following hyperparameters were used during training:
|
|
| 83 |
- Transformers 5.16.1
|
| 84 |
- Pytorch 2.11.0+cu128
|
| 85 |
- Datasets 4.0.0
|
| 86 |
-
- Tokenizers 0.23.1
|
|
|
|
| 15 |
- heitorrosa/cvm-corpus
|
| 16 |
---
|
| 17 |
|
|
|
|
|
|
|
|
|
|
| 18 |
# logun-base
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [Itau-Unibanco/NorBERTo-base](https://huggingface.co/Itau-Unibanco/NorBERTo-base) on the `heitorrosa/cvm-corpus` dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 0.4919
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
| 26 |
+
This is a LoRA adapter for the NorBERTo-base model, fine-tuned for fill-mask tasks using the CVM corpus.
|
| 27 |
|
| 28 |
## Intended uses & limitations
|
| 29 |
|
| 30 |
+
Intended for Portuguese text classification or fill-mask tasks related to the CVM domain. Limitations include the specific domain focus and the small batch size used during training.
|
| 31 |
|
| 32 |
## Training and evaluation data
|
| 33 |
|
| 34 |
+
Trained on the `heitorrosa/cvm-corpus` dataset.
|
| 35 |
|
| 36 |
## Training procedure
|
| 37 |
|
|
|
|
| 80 |
- Transformers 5.16.1
|
| 81 |
- Pytorch 2.11.0+cu128
|
| 82 |
- Datasets 4.0.0
|
| 83 |
+
- Tokenizers 0.23.1
|