working tech stack advisor app

.dockerignore

@@ -0,0 +1,38 @@
+# Ignore virtual environments
+.venv/
+venv/
+
+# Python cache
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+*.pyc
+*.pyo
+*.pyd
+*.pdb
+*.egg-info/
+
+# Jupyter Notebook checkpoints (if applicable)
+.ipynb_checkpoints/
+
+# MacOS metadata
+.DS_Store
+
+# VSCode or other IDE settings
+.vscode/
+.idea/
+
+# Git files
+.git
+.gitignore
+
+# Test folders or data artifacts (if any)
+*.log
+*.csv
+*.tsv
+*.xlsx
+
+# Docker-specific
+Dockerfile.*
+.dockerignore

app.py

@@ -0,0 +1,31 @@
+import gradio as gr
+import pickle
+import numpy as np
+
+# Load trained model and encoders
+model = pickle.load(open("model.pkl", "rb"))
+encoders = pickle.load(open("encoders.pkl", "rb"))
+
+def recommend_stack(project_type, team_size, perf_need, experience):
+    pt = encoders["project_type"].transform([project_type])[0]
+    pn = encoders["perf_need"].transform([perf_need])[0]
+    ex = encoders["experience"].transform([experience])[0]
+    input_data = np.array([[pt, team_size, pn, ex]])
+    pred_encoded = model.predict(input_data)[0]
+    return f"🔧 Recommended Tech Stack: {encoders['stack'].inverse_transform([pred_encoded])[0]}"
+
+demo = gr.Interface(
+    fn=recommend_stack,
+    inputs=[
+        gr.Radio(["Web App", "API", "ML App", "Real-time App"], label="Project Type"),
+        gr.Slider(1, 10, step=1, label="Team Size"),
+        gr.Radio(["Low", "Medium", "High"], label="Performance Need"),
+        gr.Radio(["Beginner", "Intermediate", "Expert"], label="Experience Level")
+    ],
+    outputs="text",
+    title="Tech Stack Advisor",
+    description="Get a recommended tech stack based on your project and team!"
+)
+
+demo.launch(server_name="0.0.0.0", server_port=7860)
+

requirements.txt

@@ -0,0 +1,4 @@
+scikit-learn==1.3.2
+pandas==2.1.4
+numpy==1.26.4
+gradio==4.44.1

train.py

@@ -0,0 +1,36 @@
+import pandas as pd
+from sklearn.preprocessing import LabelEncoder
+from sklearn.tree import DecisionTreeClassifier
+import pickle
+
+# Sample synthetic data
+data = {
+    "project_type": ["Web App", "API", "ML App", "Real-time App", "Web App"],
+    "team_size": [3, 2, 5, 6, 1],
+    "perf_need": ["Medium", "Low", "Medium", "High", "Low"],
+    "experience": ["Intermediate", "Beginner", "Expert", "Expert", "Beginner"],
+    "stack": ["Django + PostgreSQL", "Flask + SQLite", "FastAPI + TensorFlow", "Node.js + Redis", "Django + SQLite"]
+}
+
+df = pd.DataFrame(data)
+
+# Encode categorical variables
+encoders = {}
+for col in ["project_type", "perf_need", "experience", "stack"]:
+    le = LabelEncoder()
+    df[col] = le.fit_transform(df[col])
+    encoders[col] = le
+
+# Train model
+X = df[["project_type", "team_size", "perf_need", "experience"]]
+y = df["stack"]
+model = DecisionTreeClassifier()
+model.fit(X, y)
+
+# Save model
+with open("model.pkl", "wb") as f:
+    pickle.dump(model, f)
+
+# Save encoders for use in app
+with open("encoders.pkl", "wb") as f:
+    pickle.dump(encoders, f)