update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- rouge
|
| 6 |
+
model-index:
|
| 7 |
+
- name: tglobal-large-booksum-WIP5
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# tglobal-large-booksum-WIP5
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [pszemraj/tglobal-large-booksum-WIP4](https://huggingface.co/pszemraj/tglobal-large-booksum-WIP4) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 4.9519
|
| 19 |
+
- Rouge1: 21.8058
|
| 20 |
+
- Rouge2: 2.9343
|
| 21 |
+
- Rougel: 10.3717
|
| 22 |
+
- Rougelsum: 20.1537
|
| 23 |
+
- Gen Len: 106.055
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0004
|
| 43 |
+
- train_batch_size: 1
|
| 44 |
+
- eval_batch_size: 1
|
| 45 |
+
- seed: 31060
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 4
|
| 48 |
+
- gradient_accumulation_steps: 32
|
| 49 |
+
- total_train_batch_size: 128
|
| 50 |
+
- total_eval_batch_size: 4
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- num_epochs: 3.0
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Gen Len | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
| 58 |
+
|:-------------:|:-----:|:----:|:-------:|:---------------:|:-------:|:------:|:-------:|:---------:|
|
| 59 |
+
| 5.0389 | 0.99 | 37 | 219.03 | 5.1884 | 29.995 | 4.4045 | 12.8837 | 27.557 |
|
| 60 |
+
| 4.8986 | 1.0 | 75 | 5.1286 | 26.921 | 3.7193 | 11.3605| 25.3492 | 276.005 |
|
| 61 |
+
| 4.5928 | 2.0 | 150 | 4.9900 | 26.6667 | 3.7342 | 11.8223| 24.7087 | 178.775 |
|
| 62 |
+
| 4.6159 | 3.0 | 225 | 4.9519 | 21.8058 | 2.9343 | 10.3717| 20.1537 | 106.055 |
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
### Framework versions
|
| 66 |
+
|
| 67 |
+
- Transformers 4.25.0.dev0
|
| 68 |
+
- Pytorch 1.13.0+cu117
|
| 69 |
+
- Datasets 2.6.1
|
| 70 |
+
- Tokenizers 0.13.1
|