peft-finetune-flan-t5-mc-question-generation
Browse files
README.md
CHANGED
|
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 1.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -42,32 +42,48 @@ The following hyperparameters were used during training:
|
|
| 42 |
- total_train_batch_size: 32
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
-
- num_epochs:
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
-
| 2.
|
| 52 |
-
| 1.
|
| 53 |
-
| 1.
|
| 54 |
-
| 1.
|
| 55 |
-
| 1.
|
| 56 |
-
| 1.
|
| 57 |
-
| 1.
|
| 58 |
-
| 1.
|
| 59 |
-
| 1.
|
| 60 |
-
| 1.
|
| 61 |
-
| 1.
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
| 1.
|
| 65 |
-
| 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
| 69 |
|
| 70 |
-
- Transformers 4.
|
| 71 |
- Pytorch 2.0.0
|
| 72 |
- Datasets 2.1.0
|
| 73 |
- Tokenizers 0.13.3
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 1.1306
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 42 |
- total_train_batch_size: 32
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
+
- num_epochs: 4
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
+
| 2.2907 | 0.13 | 100 | 1.7315 |
|
| 52 |
+
| 1.8012 | 0.25 | 200 | 1.3366 |
|
| 53 |
+
| 1.5077 | 0.38 | 300 | 1.2346 |
|
| 54 |
+
| 1.419 | 0.51 | 400 | 1.2027 |
|
| 55 |
+
| 1.3772 | 0.64 | 500 | 1.1884 |
|
| 56 |
+
| 1.3566 | 0.76 | 600 | 1.1770 |
|
| 57 |
+
| 1.3348 | 0.89 | 700 | 1.1696 |
|
| 58 |
+
| 1.3307 | 1.02 | 800 | 1.1624 |
|
| 59 |
+
| 1.3247 | 1.15 | 900 | 1.1586 |
|
| 60 |
+
| 1.3139 | 1.27 | 1000 | 1.1537 |
|
| 61 |
+
| 1.3048 | 1.4 | 1100 | 1.1507 |
|
| 62 |
+
| 1.3045 | 1.53 | 1200 | 1.1476 |
|
| 63 |
+
| 1.2999 | 1.65 | 1300 | 1.1451 |
|
| 64 |
+
| 1.2978 | 1.78 | 1400 | 1.1425 |
|
| 65 |
+
| 1.2903 | 1.91 | 1500 | 1.1407 |
|
| 66 |
+
| 1.2897 | 2.04 | 1600 | 1.1409 |
|
| 67 |
+
| 1.2881 | 2.16 | 1700 | 1.1386 |
|
| 68 |
+
| 1.2845 | 2.29 | 1800 | 1.1374 |
|
| 69 |
+
| 1.2749 | 2.42 | 1900 | 1.1360 |
|
| 70 |
+
| 1.2846 | 2.55 | 2000 | 1.1349 |
|
| 71 |
+
| 1.281 | 2.67 | 2100 | 1.1339 |
|
| 72 |
+
| 1.282 | 2.8 | 2200 | 1.1331 |
|
| 73 |
+
| 1.2786 | 2.93 | 2300 | 1.1326 |
|
| 74 |
+
| 1.2799 | 3.06 | 2400 | 1.1322 |
|
| 75 |
+
| 1.2777 | 3.18 | 2500 | 1.1320 |
|
| 76 |
+
| 1.2767 | 3.31 | 2600 | 1.1316 |
|
| 77 |
+
| 1.2716 | 3.44 | 2700 | 1.1313 |
|
| 78 |
+
| 1.2762 | 3.56 | 2800 | 1.1309 |
|
| 79 |
+
| 1.2723 | 3.69 | 2900 | 1.1305 |
|
| 80 |
+
| 1.2741 | 3.82 | 3000 | 1.1304 |
|
| 81 |
+
| 1.2762 | 3.95 | 3100 | 1.1306 |
|
| 82 |
|
| 83 |
|
| 84 |
### Framework versions
|
| 85 |
|
| 86 |
+
- Transformers 4.33.1
|
| 87 |
- Pytorch 2.0.0
|
| 88 |
- Datasets 2.1.0
|
| 89 |
- Tokenizers 0.13.3
|