Hugging Face's logo

Lvxue
/
finetuned-mt5-small-10epoch

Translation
Transformers
PyTorch
English
Romanian
mt5
text2text-generation
wmt16
Lvxue
Eval Results
Inference Endpoints
Model card Files Community
8
finetuned-mt5-small-10epoch / README.md
autoevaluator's picture
autoevaluator HF staff
Add evaluation results on the cs-en config and train split of wmt16
3d93faa
preview code
|
raw history blame
2.79 kB
metadata
language:
  - en
  - ro
license: apache-2.0
tags:
  - translation
  - wmt16
  - Lvxue
datasets:
  - wmt16
metrics:
  - sacrebleu
  - bleu
model-index:
  - name: Lvxue/finetuned-mt5-small-10epoch
    results:
      - task:
          type: translation
          name: Translation
        dataset:
          name: wmt16
          type: wmt16
          config: ro-en
          split: test
        metrics:
          - type: bleu
            value: 6.0012
            name: BLEU
            verified: true
          - type: loss
            value: 1.7407585382461548
            name: loss
            verified: true
          - type: gen_len
            value: 18.2281
            name: gen_len
            verified: true
      - task:
          type: translation
          name: Translation
        dataset:
          name: wmt16
          type: wmt16
          config: cs-en
          split: train
        metrics:
          - type: bleu
            value: 0.4125
            name: BLEU
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZTUxNmNhNjY2Nzk3ZTRkOTM0ZjQ4ZDI2ZTZkZTBkMzBkYjc1Y2EyNmVlOWEzYjc3ZjIzOTA5MDk4M2RiMzgxMyIsInZlcnNpb24iOjF9.XzXIZo3AlYuFeaJwBcHJPFML4Um4MER4XbL3RDuwwZeOMDnqnMfspLcrbyvozyZjSC8JWWvpgJ192JflgDv1Cw
          - type: loss
            value: 6.76282262802124
            name: loss
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNzI5ZDZkYzQxYTA3YzJjZWY0OWNhOGQ2YTIzOGJmMmY4NDQ4M2Y2ZTJmMWZiNWY0N2Y3Y2VjNDU4ZWMwZTM2OSIsInZlcnNpb24iOjF9.5yJy4N0hC3myiT0Q8kKNnKLbZnRiilaQA3dDxTNm7Sz-rNZqMTs9qUsX5S6Y4S4_hWD1Tny2xJle6o33IMDXBw
          - type: gen_len
            value: 18.4717
            name: gen_len
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZDdmZGE4ZTg5MGI1MmE3NjFmYzE2MGMyNWJiNGFiYjUwNWRjYjExNmUyNDQ2ZGY1OWJkNWQ4M2U2ZGVmMDYwMSIsInZlcnNpb24iOjF9.YX8bVSFJKTvCyeWRrgwt3W9EVbL9IcC6T4n2LGkz1AHc2aB3e-l8Wwg_yzfI_W3HBmVDveLGw6I7qUg1YgCcAA

finetuned-mt5-small-10epoch

This model is a fine-tuned version of google/mt5-small on the wmt16 ro-en dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7274

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: 5e-05
  • train_batch_size: 48
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10.0

Training results

Framework versions

  • Transformers 4.20.1
  • Pytorch 1.12.0+cu102
  • Datasets 2.3.2
  • Tokenizers 0.12.1