SeoyeonPark1223 commited on
Commit
4e5d372
·
verified ·
1 Parent(s): d04cb92

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -3
README.md CHANGED
@@ -97,6 +97,7 @@ tags:
97
  - Activation: Softmax (to output probabilities for each class).
98
  - **Model Performance:**
99
  ![Model Performance](./pics/audio-performance.png)
 
100
  1. Accuracy and Preprocessing (Table Summary)
101
  - The CNN model achieves the highest accuracy of 98.37% in the 8th configuration.
102
  - Key factors contributing to this performance:
@@ -106,9 +107,9 @@ tags:
106
  - Library: Keras is used for training and architecture implementation.
107
  - Sample Rate: A consistent sample rate of 16,000 Hz was maintained for all preprocessing steps.
108
  2. Confusion Matrix Analysis
109
- - High Precision: Minimal false positives suggest the model is very specific when identifying emergencies.
110
- - High Recall: Minimal false negatives indicate that most emergencies are correctly identified.
111
- ![Confusion Matrix](./pics/confusion-matrix.png)
112
 
113
 
114
  ## Model Usage
 
97
  - Activation: Softmax (to output probabilities for each class).
98
  - **Model Performance:**
99
  ![Model Performance](./pics/audio-performance.png)
100
+ ![Confusion Matrix](./pics/confusion-matrix.png)
101
  1. Accuracy and Preprocessing (Table Summary)
102
  - The CNN model achieves the highest accuracy of 98.37% in the 8th configuration.
103
  - Key factors contributing to this performance:
 
107
  - Library: Keras is used for training and architecture implementation.
108
  - Sample Rate: A consistent sample rate of 16,000 Hz was maintained for all preprocessing steps.
109
  2. Confusion Matrix Analysis
110
+ - High Precision: Minimal false positives suggest the model is very specific when identifying emergencies.
111
+ - High Recall: Minimal false negatives indicate that most emergencies are correctly identified.
112
+
113
 
114
 
115
  ## Model Usage