multimind-sdk
Version:
This SDK gives JavaScript/TypeScript developers full access to advanced AI features like agent orchestration, RAG, and fine-tuning ā without needing to manage backend code.
206 lines ā¢ 8.22 kB
JavaScript
import MultiMindSDK from '../index.js';
async function comprehensiveDemo() {
const sdk = new MultiMindSDK();
try {
console.log('š MultiMind SDK Comprehensive Demo\n');
await sdk.initialize();
// 1. SDK Information
console.log('1. š SDK Information');
const sdkInfo = await sdk.getSDKInfo();
console.log('SDK Version:', sdkInfo.version);
console.log('Features:', sdkInfo.features.length);
console.log('Initialized:', sdkInfo.initialized);
// 2. Health Check
console.log('\n2. š„ Health Check');
const health = await sdk.healthCheck();
console.log('Status:', health.status);
console.log('Message:', health.message);
// 3. Basic Agent Generation
console.log('\n3. š¤ Basic Agent Generation');
try {
const response = await sdk.generateWithAgent("Explain quantum computing in simple terms", { model: "gpt-3.5-turbo", temperature: 0.7, maxTokens: 200 });
console.log('Response:', response);
}
catch (error) {
console.log('Basic agent failed:', error.message);
}
// 4. Advanced Fine-tuning
console.log('\n4. šÆ Advanced Fine-tuning Demo');
try {
const fineTuneResult = await sdk.advancedFineTune({
baseModelName: "bert-base-uncased",
outputDir: "./output/finetune",
method: "lora",
epochs: 3,
learningRate: 0.001,
batchSize: 16,
loraConfig: {
r: 16,
alpha: 32,
dropout: 0.1,
targetModules: ["query", "value"]
}
});
console.log('Fine-tuning result:', fineTuneResult);
}
catch (error) {
console.log('Advanced fine-tuning failed:', error.message);
}
// 5. Advanced RAG System
console.log('\n5. š Advanced RAG System');
try {
const documents = [
{
text: "MultiMind SDK is a comprehensive AI development toolkit that unifies fine-tuning, RAG, and agent orchestration.",
metadata: { type: "introduction", source: "docs" }
},
{
text: "The SDK supports advanced fine-tuning methods including LoRA, Adapters, and Prefix Tuning.",
metadata: { type: "features", source: "docs" }
},
{
text: "RAG capabilities include document processing, vector storage, and hybrid retrieval.",
metadata: { type: "features", source: "docs" }
}
];
await sdk.addDocumentsToRAG(documents);
console.log('Documents added to RAG');
const ragResponse = await sdk.queryAdvancedRAG({
query: "What are the main features of MultiMind SDK?",
topK: 3,
includeMetadata: true
});
console.log('RAG Response:', ragResponse);
}
catch (error) {
console.log('Advanced RAG failed:', error.message);
}
// 6. Model Conversion
console.log('\n6. š Model Conversion');
try {
const conversionResult = await sdk.pytorchToONNX("./models/sample_model.pt", "./models/sample_model.onnx", {
quantization: {
method: "int8",
targetDevice: "cpu"
},
graphOptimization: {
fuseOperations: true,
removeUnusedNodes: true,
optimizeMemory: true
}
});
console.log('Model conversion result:', conversionResult);
}
catch (error) {
console.log('Model conversion failed:', error.message);
}
// 7. Compliance Monitoring
console.log('\n7. š Compliance Monitoring');
try {
const complianceResult = await sdk.checkCompliance({
modelId: "model_123",
dataCategories: ["text", "user_data"],
useCase: "customer_support",
region: "EU"
});
console.log('Compliance check result:', complianceResult);
}
catch (error) {
console.log('Compliance check failed:', error.message);
}
// 8. Advanced Agent with Tools
console.log('\n8. š ļø Advanced Agent with Tools');
try {
const agentResponse = await sdk.runAdvancedAgent("Calculate 15 * 23 and then search for information about quantum computing", { context: "mathematical and scientific inquiry" });
console.log('Advanced Agent Response:', agentResponse);
}
catch (error) {
console.log('Advanced agent failed:', error.message);
}
// 9. Model Client System
console.log('\n9. š§ Model Client System');
try {
// LSTM Model Client
const lstmClient = await sdk.createLSTMModelClient({
modelPath: "./models/lstm_model.pt",
modelName: "custom_lstm",
maxLength: 512,
temperature: 0.7
});
console.log('LSTM client created');
// MoE Model Client
const moeClient = await sdk.createMoEModelClient({
experts: {
"expert1": { modelName: "gpt-3.5-turbo" },
"expert2": { modelName: "claude-3" }
},
router: (input) => input.length > 100 ? "expert2" : "expert1",
loadBalancing: true
});
console.log('MoE client created');
// MultiModal Client
const mmClient = await sdk.createMultiModalClient({
textClient: lstmClient,
fusionStrategy: "attention"
});
console.log('MultiModal client created');
}
catch (error) {
console.log('Model client system failed:', error.message);
}
// 10. Gateway API
console.log('\n10. š Gateway API');
try {
const gatewayResult = await sdk.startGateway({
host: "0.0.0.0",
port: 8000,
enableMiddleware: true,
corsEnabled: true,
rateLimit: 100
});
console.log('Gateway started:', gatewayResult);
// Stop gateway after demo
setTimeout(async () => {
await sdk.stopGateway();
console.log('Gateway stopped');
}, 5000);
}
catch (error) {
console.log('Gateway failed:', error.message);
}
// 11. Model Evaluation
console.log('\n11. š Model Evaluation');
try {
const evaluation = await sdk.evaluateModel({
model: "gpt-3.5-turbo",
task: "text-generation",
dataset: "test-dataset",
metrics: ["perplexity", "accuracy", "f1"]
});
console.log('Model evaluation:', evaluation);
}
catch (error) {
console.log('Model evaluation failed:', error.message);
}
// 12. Model Comparison
console.log('\n12. āļø Model Comparison');
try {
const comparison = await sdk.compareModels(["gpt-3.5-turbo", "claude-3", "mistral"], "text-classification", "benchmark-dataset");
console.log('Model comparison:', comparison);
}
catch (error) {
console.log('Model comparison failed:', error.message);
}
console.log('\nā
Comprehensive demo completed successfully!');
}
catch (error) {
console.error('ā Demo failed:', error);
}
finally {
await sdk.close();
console.log('\nš SDK closed');
}
}
// Run the comprehensive demo
comprehensiveDemo().catch(console.error);
//# sourceMappingURL=comprehensive-demo.js.map