update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: afl-3.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- lg-ner
|
| 7 |
+
metrics:
|
| 8 |
+
- precision
|
| 9 |
+
- recall
|
| 10 |
+
- f1
|
| 11 |
+
- accuracy
|
| 12 |
+
model-index:
|
| 13 |
+
- name: luganda-ner-v5
|
| 14 |
+
results:
|
| 15 |
+
- task:
|
| 16 |
+
name: Token Classification
|
| 17 |
+
type: token-classification
|
| 18 |
+
dataset:
|
| 19 |
+
name: lg-ner
|
| 20 |
+
type: lg-ner
|
| 21 |
+
config: lug
|
| 22 |
+
split: test
|
| 23 |
+
args: lug
|
| 24 |
+
metrics:
|
| 25 |
+
- name: Precision
|
| 26 |
+
type: precision
|
| 27 |
+
value: 0.8502710027100271
|
| 28 |
+
- name: Recall
|
| 29 |
+
type: recall
|
| 30 |
+
value: 0.8428475486903962
|
| 31 |
+
- name: F1
|
| 32 |
+
type: f1
|
| 33 |
+
value: 0.8465430016863407
|
| 34 |
+
- name: Accuracy
|
| 35 |
+
type: accuracy
|
| 36 |
+
value: 0.959089589080877
|
| 37 |
+
---
|
| 38 |
+
|
| 39 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 40 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 41 |
+
|
| 42 |
+
# luganda-ner-v5
|
| 43 |
+
|
| 44 |
+
This model is a fine-tuned version of [masakhane/afroxlmr-large-ner-masakhaner-1.0_2.0](https://huggingface.co/masakhane/afroxlmr-large-ner-masakhaner-1.0_2.0) on the lg-ner dataset.
|
| 45 |
+
It achieves the following results on the evaluation set:
|
| 46 |
+
- Loss: 0.2328
|
| 47 |
+
- Precision: 0.8503
|
| 48 |
+
- Recall: 0.8428
|
| 49 |
+
- F1: 0.8465
|
| 50 |
+
- Accuracy: 0.9591
|
| 51 |
+
|
| 52 |
+
## Model description
|
| 53 |
+
|
| 54 |
+
More information needed
|
| 55 |
+
|
| 56 |
+
## Intended uses & limitations
|
| 57 |
+
|
| 58 |
+
More information needed
|
| 59 |
+
|
| 60 |
+
## Training and evaluation data
|
| 61 |
+
|
| 62 |
+
More information needed
|
| 63 |
+
|
| 64 |
+
## Training procedure
|
| 65 |
+
|
| 66 |
+
### Training hyperparameters
|
| 67 |
+
|
| 68 |
+
The following hyperparameters were used during training:
|
| 69 |
+
- learning_rate: 2e-05
|
| 70 |
+
- train_batch_size: 8
|
| 71 |
+
- eval_batch_size: 8
|
| 72 |
+
- seed: 42
|
| 73 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 74 |
+
- lr_scheduler_type: linear
|
| 75 |
+
- num_epochs: 6
|
| 76 |
+
|
| 77 |
+
### Training results
|
| 78 |
+
|
| 79 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 80 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 81 |
+
| No log | 1.0 | 261 | 0.2276 | 0.7703 | 0.6441 | 0.7015 | 0.9353 |
|
| 82 |
+
| 0.3176 | 2.0 | 522 | 0.1848 | 0.8431 | 0.7542 | 0.7962 | 0.9545 |
|
| 83 |
+
| 0.3176 | 3.0 | 783 | 0.1871 | 0.8564 | 0.8173 | 0.8364 | 0.9576 |
|
| 84 |
+
| 0.0753 | 4.0 | 1044 | 0.2015 | 0.8691 | 0.8294 | 0.8488 | 0.9614 |
|
| 85 |
+
| 0.0753 | 5.0 | 1305 | 0.2325 | 0.8616 | 0.8361 | 0.8487 | 0.9584 |
|
| 86 |
+
| 0.0261 | 6.0 | 1566 | 0.2328 | 0.8503 | 0.8428 | 0.8465 | 0.9591 |
|
| 87 |
+
|
| 88 |
+
|
| 89 |
+
### Framework versions
|
| 90 |
+
|
| 91 |
+
- Transformers 4.27.4
|
| 92 |
+
- Pytorch 1.13.1+cu116
|
| 93 |
+
- Datasets 2.11.0
|
| 94 |
+
- Tokenizers 0.13.2
|