Yaredoffice commited on
Commit
58df1c9
·
verified ·
1 Parent(s): c6be04e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -0
README.md CHANGED
@@ -53,6 +53,7 @@ This model addresses the challenge of digital recognition for the Geez script by
53
 
54
  The model is intended for direct use in digitizing handwritten Geez documents, educational language learning tools, and automated data entry systems. Users input a cropped image of a handwritten character, and the model returns the predicted character class and confidence score.
55
 
 
56
  ### Downstream Use [optional]
57
 
58
  <!-- This section is for model use when fine-tuned for a task, or when plugged into a larger ecosystem/app. -->
@@ -84,6 +85,20 @@ The model is **not** designed for:
84
 
85
  Users should implement a pre-processing pipeline to segment words into individual characters before feeding them into this model. Images should be normalized to 128x128 pixels and converted to grayscale.
86
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
87
  ## How to Get Started with the Model
88
 
89
  Use the code below to get started with the model.
 
53
 
54
  The model is intended for direct use in digitizing handwritten Geez documents, educational language learning tools, and automated data entry systems. Users input a cropped image of a handwritten character, and the model returns the predicted character class and confidence score.
55
 
56
+
57
  ### Downstream Use [optional]
58
 
59
  <!-- This section is for model use when fine-tuned for a task, or when plugged into a larger ecosystem/app. -->
 
85
 
86
  Users should implement a pre-processing pipeline to segment words into individual characters before feeding them into this model. Images should be normalized to 128x128 pixels and converted to grayscale.
87
 
88
+ ## Evaluation
89
+
90
+ ### Testing Data, Factors & Metrics
91
+
92
+ #### Metrics
93
+ - **Accuracy**: The primary metric used for evaluation.
94
+
95
+ #### Inference Performance
96
+ - **Single Image Inference**: 81% baseline accuracy.
97
+ - **Test-Time Augmentation (TTA)**:
98
+ - Configuration: 10 augmentations with majority voting.
99
+ - Result: Achieves approximately **90% classification accuracy**.
100
+ - Impact: Significantly reduces error rates caused by handwriting variability.
101
+
102
  ## How to Get Started with the Model
103
 
104
  Use the code below to get started with the model.