CLI tool for evaluating educational questions via Incept API
Project description
Incept Eval
CLI tool for evaluating educational questions with configurable AI-powered modules.
Features
🎯 Multiple Evaluation Modules
- V3 Evaluation - Scaffolding and DI compliance
- Answer Verification - GPT-4 powered correctness checking
- EduBench Tasks - Educational quality benchmarks (QA, EC, IP, AG)
⚙️ Configurable
- Enable/disable modules per evaluation
- Custom module combinations
- Save configurations for reuse
🚀 Easy to Use
- Simple CLI interface
- JSON input/output
- Batch processing support
Installation
pip install incept-eval
Quick Start
1. Get API Key
Get your API key at https://incept.ai/api-keys
2. Configure
incept-eval configure YOUR_API_KEY
3. Evaluate
# Generate example
incept-eval example > questions.json
# Evaluate
incept-eval evaluate questions.json --pretty
Usage
Evaluate Questions
# Full evaluation (all modules)
incept-eval evaluate questions.json --output results.json --pretty
# Only V3 evaluation
incept-eval evaluate questions.json --no-answer-verify --edubench
# Only answer verification
incept-eval evaluate questions.json --no-v3 --edubench
# Custom modules
incept-eval evaluate questions.json --edubench QA --edubench EC
Quick Single Question
incept-eval quick-eval "What is 2+2?" "4" --grade 2 --pretty
Input Format
{
"request": {
"grade": 3,
"subject": "mathematics",
"instructions": "Generate multiplication problems",
"language": "english",
"count": 2
},
"questions": [
{
"type": "mcq",
"question": "What is 3 × 7?",
"answer": "21",
"difficulty": "medium",
"explanation": "Multiply 3 by 7",
"options": {"A": "18", "B": "21", "C": "24", "D": "28"},
"answer_choice": "B"
}
]
}
Authentication
Three ways to provide your API key:
1. Config file (recommended)
incept-eval configure YOUR_API_KEY
2. Environment variable
export INCEPT_API_KEY=YOUR_API_KEY
3. Command line
incept-eval evaluate questions.json --api-key YOUR_API_KEY
Commands
| Command | Description |
|---|---|
configure |
Save API key to config |
evaluate |
Evaluate questions from file |
quick-eval |
Quick single question evaluation |
example |
Generate example input |
modules |
Show available modules |
Evaluation Modules
V3 Evaluation
- Scaffolding quality assessment
- Direct Instruction (DI) compliance
- Pedagogical structure validation
Answer Verification
- GPT-4 powered correctness checking
- Mathematical accuracy validation
- Confidence scoring
EduBench Tasks
- QA: Question Answering evaluation
- EC: Error Correction analysis
- IP: Instructional Planning assessment
- AG: Answer Generation
Examples
Quality Assurance Pipeline
incept-eval evaluate questions.json --output qa_results.json --verbose
Fast Validation
incept-eval evaluate questions.json --no-v3 --edubench
Batch Processing
for file in questions/*.json; do
incept-eval evaluate "$file" --output "results/$(basename $file)"
done
Requirements
- Python >= 3.11
- Incept API key
Support
- Issues: GitHub Issues
- Documentation: CLI Usage Guide
- API Keys: https://incept.ai/api-keys
License
MIT License - see LICENSE file for details.
Made with ❤️ by the Incept Team
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file incept_eval-1.0.1.tar.gz.
File metadata
- Download URL: incept_eval-1.0.1.tar.gz
- Upload date:
- Size: 4.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/2.2.1 CPython/3.13.7 Darwin/24.3.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
7126daef8023e4de003478ea8aa51ca7211daa5463377dad99091b5ba1957dce
|
|
| MD5 |
caea23e774b59a76b1af04332af9e4d9
|
|
| BLAKE2b-256 |
dd36b4fba1368fe491efe3d44fb612242e9dc3694af9131f188ffa18024e7eb4
|
File details
Details for the file incept_eval-1.0.1-py3-none-any.whl.
File metadata
- Download URL: incept_eval-1.0.1-py3-none-any.whl
- Upload date:
- Size: 5.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/2.2.1 CPython/3.13.7 Darwin/24.3.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
ff5fc653439758de32daae604a774c31c08fdeb8dd29c5d81f4ee49691038d3f
|
|
| MD5 |
371a7b778be087d1866e66f17354f35b
|
|
| BLAKE2b-256 |
c0f9e7bebcdc0e5a19ca88f5bfea85ea6fce0acd0d546c682954d019226e0292
|