metadata
base_model: UrukHan/t5-russian-summarization
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: last_run_UrukHan_t5-russian-summarization
results: []
last_run_UrukHan_t5-russian-summarization
This model is a fine-tuned version of UrukHan/t5-russian-summarization on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6429
- F1: 0.4241
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: 0.00012
- train_batch_size: 4
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7.0
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.8758 | 0.6667 | 200 | 0.9457 | 0.3173 |
| 0.7212 | 1.3333 | 400 | 0.8009 | 0.3486 |
| 0.5636 | 2.0 | 600 | 0.7738 | 0.3699 |
| 0.4931 | 2.6667 | 800 | 0.7189 | 0.3951 |
| 0.4748 | 3.3333 | 1000 | 0.6766 | 0.4020 |
| 0.4125 | 4.0 | 1200 | 0.6658 | 0.4143 |
| 0.3326 | 4.6667 | 1400 | 0.6481 | 0.4243 |
| 0.2676 | 5.3333 | 1600 | 0.6494 | 0.4063 |
| 0.3643 | 6.0 | 1800 | 0.6383 | 0.4175 |
| 0.2064 | 6.6667 | 2000 | 0.6429 | 0.4241 |
Framework versions
- Transformers 4.43.1
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1