Update README.md

README.md

@@ -1,8 +1,9 @@
 ---
 license: creativeml-openrail-m
 ---
+# Model Card for Kanji_ETL9G
 
-Based on:
+### Summary:
 ```
 ETL9G
  : 607200 samples
@@ -13,49 +14,41 @@ ETL9G
  : image_height: 64px
 ```
 
-I was testing a few more samples locally with the below.   Note: The results of the model are encoded. 
 
-Model Card for Kanji_ETL9G
-Model Details
-Model Name: Kanji_ETL9G
-Version: 1.0.0
-Model Type: Neural Network (Specify architecture, e.g., Fully Connected Feedforward Neural Network)
-Framework: PyTorch
-Model Description
+
+## Model Details
+- **Model Name:** Kanji_ETL9G
+- **Version:** 1.0.0
+- **Model Type:** Neural Network 
+- **Framework:** PyTorch
+
+## Model Description
 This model is trained on a dataset derived from the ETL9G dataset to recognize Kanji characters from 64x64 grayscale images. The primary use-case is for optical character recognition (OCR) for handwritten Kanji characters.
 
-Intended Use
+## Intended Use
 The primary application of this model is for OCR tasks to recognize handwritten Kanji characters in images, with potential extensions for applications like smart dictionary lookup, handwriting-based user authentication, and so on.
 
-Limitations
+## Limitations
 This model might have limitations regarding:
+- Variability in handwriting styles not present in the training set. (200 samples per character/class were used)
+- Noises and artifacts in input images.
+- Characters written in unconventional ways.
+
+## Data Details
+
+### Training Data:
+- **Dataset:** Derived from the ETL9G dataset (http://etlcdb.db.aist.go.jp/specification-of-etl-9)
+- **Size:** 607200 samples
+- **Data Type:** 64x64 grayscale images of handwritten Kanji characters (images were resized from 128x127 due to technical limitations)
+- **Labels:** 3036 unique characters (classes)
+
+## Model Files
+- **PyTorch Model:** Kanji_ETL9G.pth
+- **ONNX Model:** Kanji_ETL9G.onnx
+- **CoreML Model:** next effort....
 
-Variability in handwriting styles not present in the training set.
-Noises and artifacts in input images.
-Characters written in unconventional ways.
-Data Details
-Training Data:
-Dataset: Derived from the ETL9G dataset
-Size: (Specify the number of samples in the training dataset)
-Data Type: 64x64 grayscale images of handwritten Kanji characters
-Labels: Kanji characters (Specify the total number of unique characters)
-Validation and Test Data:
-Validation Size: (Specify the number of samples in the validation dataset)
-Test Size: (Specify the number of samples in the test dataset)
-Data Type: Same as training
-Labels: Same as training
-Metrics
-Accuracy: (Your test accuracy here, e.g., 95.4%)
-Precision: (If calculated)
-Recall: (If calculated)
-F1 Score: (If calculated)
-Model Files
-PyTorch Model: Kanji_ETL9G.pth
-ONNX Model: (If you've saved it in ONNX format)
-CoreML Model: (If you've converted it to CoreML)
-Usage
-python
-Copy code
+## Usage
+```python
 import torch
 model = torch.load('Kanji_ETL9G.pth')
 model.eval()
@@ -63,7 +56,3 @@ model.eval()
 # Assuming input image tensor is `input_tensor`
 output = model(input_tensor)
 predicted_label = torch.argmax(output).item()
-Maintainers
-Your Name (Your email/contact details)
-Licensing
-Specify the license under which you're releasing the model. If unsure, consider common licenses like MIT, Apache 2.0, etc.