Training / Validation Curves
Summary
Class Counts
Confusion Matrix
low mid high
Per-class Recall
Derived from the confusion matrix (recall = TP / support).
Sample Batch (train)
Written from sample_train.png exported by your script.
Sample Predictions (test)
Generated as sample_pred.png to show model predictions (title: T=true, P=pred).
Wrong Examples (if any)
Only visible if wrong_examples.png exists.
Hard Examples (low-confidence corrects)
Correct predictions with confidence < 0.60 (if available).
📊 Manual Test Results
When you run manual test in your local machine, be sure to make background black and the digits white color.
Result of manual test is 8/10 correct predictions.
💻 Run It on Your Computer
Want to try the trained model yourself? Download the project and start the local Flask server.
⬇️ Download Project (ZIP)- Unzip the folder.
- Open a terminal inside the project.
- First of all if you don't have python in your device follow this steps.
- (For WINDOWS)
- Install python:
winget install -e --id Python.Python.3.12 - After install python check if installed:
python --version - (For MAC)
- If you don't have brew install it before installing python:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)" - Install python:
brew install python - After install python check if installed:
python3 --version - THAN (Optional)
- (For WINDOWS)
- Create a virtual environment:
py -3 -m venv .venv - (If you use PowerShell)
.\.venv\Scripts\Activate.ps1- (If you use CMD)
.\.venv\Scripts\activate.bat- (For MAC)
- Create a virtual environment:
python3 -m venv env && source env/bin/activate - THAN
- Install requirements:
pip install -r requirements.txt - Run the local server:
python serve.py - Return here and click the button below.