End of training

README.md

@@ -5,8 +5,6 @@ language:
 base_model: Bhuvana/t5-base-spellchecker
 tags:
 - generated_from_trainer
-metrics:
-- rouge
 model-index:
 - name: T5 Sindhi Spell Checker - Fahad Maqsood Qazi
   results: []
@@ -18,13 +16,6 @@ should probably proofread and complete it, then remove this comment. -->
 # T5 Sindhi Spell Checker - Fahad Maqsood Qazi
 
 This model is a fine-tuned version of [Bhuvana/t5-base-spellchecker](https://huggingface.co/Bhuvana/t5-base-spellchecker) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 3.4755
-- Rouge1: 16.5098
-- Rouge2: 13.9875
-- Rougel: 16.4217
-- Rougelsum: 16.5185
-- Gen Len: 19.9968
 
 ## Model description
 
@@ -44,22 +35,18 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 64
+- train_batch_size: 16
+- eval_batch_size: 100
 - seed: 42
-- gradient_accumulation_steps: 8
+- gradient_accumulation_steps: 4
 - total_train_batch_size: 64
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- training_steps: 2000
+- training_steps: 10
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
-|:-------------:|:------:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
-| 30.6232       | 0.0138 | 1000 | 3.6027          | 16.3965 | 13.9006 | 16.3126 | 16.4006   | 19.9988 |
-| 28.9499       | 0.0276 | 2000 | 3.4755          | 16.5098 | 13.9875 | 16.4217 | 16.5185   | 19.9968 |
 
 
 ### Framework versions