Update README.md

README.md

@@ -19,21 +19,6 @@ This is an INT8  PyTorch model quantized by [intel/nlp-toolkit](https://github.c
 
 The original fp32 model comes from the fine-tuned model [Intel/bert-base-uncased-mrpc](https://huggingface.co/Intel/bert-base-uncased-mrpc).
 
-#### Training hyperparameters
-
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- num_epochs: 3.0
-- train_batch_size: 8
-- eval_batch_size: 8
-- eval_steps: 100
-- load_best_model_at_end: True
-- metric_for_best_model: f1
-- early_stopping_patience = 6
-- early_stopping_threshold = 0.001
-
 ### Test result
 
 - Batch size = 8
@@ -42,7 +27,7 @@ The following hyperparameters were used during training:
 |   |INT8|FP32|
 |---|:---:|:---:|
 | **Throughput (samples/sec)**  |24.263|11.202|
-| **Accuracy (eval-accuracy)** |0.9153|0.9042|
+| **Accuracy (eval-f1)** |0.9153|0.9042|
 | **Model size (MB)**  |174|418|
 
 ### Load with nlp-toolkit:
@@ -56,3 +41,18 @@ int8_model = OptimizedModel.from_pretrained(
 
 Notes:  
  - The INT8 model has better performance than the FP32 model when the CPU is fully occupied. Otherwise, there will be the illusion that INT8 is inferior to FP32.
+
+### Training hyperparameters
+    
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+- train_batch_size: 8
+- eval_batch_size: 8
+- eval_steps: 100
+- load_best_model_at_end: True
+- metric_for_best_model: f1
+- early_stopping_patience = 6
+- early_stopping_threshold = 0.001