metadata
license: mit
base_model: facebook/mbart-large-50
tags:
- summarization
- generated_from_trainer
datasets:
- lr-sum
metrics:
- rouge
model-index:
- name: mbart-large-50-finetuned-lrsum-fr
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: lr-sum
type: lr-sum
config: fra
split: validation
args: fra
metrics:
- name: Rouge1
type: rouge
value: 0.2579
mbart-large-50-finetuned-lrsum-fr
This model is a fine-tuned version of facebook/mbart-large-50 on the lr-sum dataset. It achieves the following results on the evaluation set:
- Loss: 2.0341
- Rouge1: 0.2579
- Rouge2: 0.1232
- Rougel: 0.2142
- Rougelsum: 0.2153
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5.6e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 1.8023 | 1.0 | 141 | 1.2331 | 0.2511 | 0.115 | 0.205 | 0.2088 |
| 8.3626 | 2.0 | 282 | 1.3380 | 0.2601 | 0.1213 | 0.2106 | 0.2155 |
| 0.848 | 3.0 | 423 | 1.5333 | 0.2431 | 0.1109 | 0.2008 | 0.2022 |
| 0.4302 | 4.0 | 564 | 1.4443 | 0.2487 | 0.1153 | 0.204 | 0.2063 |
| 0.2181 | 5.0 | 705 | 1.6967 | 0.2445 | 0.1081 | 0.1977 | 0.2001 |
| 0.1131 | 6.0 | 846 | 1.8275 | 0.2704 | 0.1358 | 0.2249 | 0.2265 |
| 0.052 | 7.0 | 987 | 1.9579 | 0.2549 | 0.1161 | 0.2085 | 0.2099 |
| 0.0245 | 8.0 | 1128 | 2.0341 | 0.2579 | 0.1232 | 0.2142 | 0.2153 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1