alexsha
/

t5-large-finetuned-English-to-BASH

Text2Text Generation

Generated from Trainer

Inference Endpoints

text-generation-inference

Model card Files Files and versions Metrics Training metrics Community

alexsha commited on Feb 25, 2023

Commit

005b019

•

1 Parent(s): 74e4932

update model card README.md

Files changed (1) hide show

README.md +10 -15

README.md CHANGED Viewed

@@ -14,9 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6147
-- Nl2bash M: 0.776
-- Gen Len: 14.4407
 ## Model description
@@ -35,28 +35,23 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 0.0001
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Nl2bash M | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
-| 1.5433        | 1.0   | 561  | 0.9192          | 0.6019    | 13.6592 |
-| 0.9018        | 2.0   | 1122 | 0.7196          | 0.6793    | 14.3167 |
-| 0.6917        | 3.0   | 1683 | 0.6409          | 0.7169    | 14.3827 |
-| 0.5842        | 4.0   | 2244 | 0.6094          | 0.7384    | 14.1775 |
-| 0.5013        | 5.0   | 2805 | 0.6041          | 0.7577    | 14.2194 |
-| 0.4295        | 6.0   | 3366 | 0.5921          | 0.7632    | 14.4202 |
-| 0.3762        | 7.0   | 3927 | 0.5983          | 0.7625    | 14.4853 |
-| 0.3366        | 8.0   | 4488 | 0.6032          | 0.7707    | 14.4755 |
-| 0.2779        | 9.0   | 5049 | 0.6084          | 0.7725    | 14.3666 |
-| 0.2654        | 10.0  | 5610 | 0.6147          | 0.776     | 14.4407 |
 ### Framework versions

 This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.7646
+- Nl2bash M: 0.7822
+- Gen Len: 14.4665
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 4e-05
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
+- num_epochs: 5
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Nl2bash M | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
+| 0.2312        | 1.0   | 561  | 0.6370          | 0.7743    | 14.4202 |
+| 0.1836        | 2.0   | 1122 | 0.6767          | 0.7782    | 14.4585 |
+| 0.1445        | 3.0   | 1683 | 0.7171          | 0.7805    | 14.3586 |
+| 0.1263        | 4.0   | 2244 | 0.7457          | 0.78      | 14.479  |
+| 0.1092        | 5.0   | 2805 | 0.7646          | 0.7822    | 14.4665 |
 ### Framework versions