Construct a Absolutely Autonomous Native Fleet-Upkeep Evaluation Agent Utilizing SmolAgents and Qwen Mannequin

On this tutorial, we stroll by way of the method of making a totally autonomous fleet-analysis agent utilizing SmolAgents and a neighborhood Qwen mannequin. We generate telemetry knowledge, load it by way of a customized device, and let our agent cause, analyze, and visualize upkeep dangers with none exterior API calls. At every step of implementation, we see how the agent interprets structured logs, applies logical filters, detects anomalies, and eventually produces a transparent visible warning for fleet managers. Try the FULL CODES right here.

print("⏳ Putting in libraries... (approx 30-60s)")
!pip set up smolagents transformers speed up bitsandbytes ddgs matplotlib pandas -q


import os
import pandas as pd
import matplotlib.pyplot as plt
from smolagents import CodeAgent, Software, TransformersModel

We set up all required libraries and import the core modules we depend on for constructing our agent. We arrange SmolAgents, Transformers, and fundamental data-handling instruments to course of telemetry and run the native mannequin easily. At this stage, we put together the environment and guarantee all the things masses accurately earlier than shifting forward. Try the FULL CODES right here.

fleet_data = {
   "truck_id": ["T-101", "T-102", "T-103", "T-104", "T-105"],
   "driver": ["Ali", "Sara", "Mike", "Omar", "Jen"],
   "avg_speed_kmh": [65, 70, 62, 85, 60],
   "fuel_efficiency_kml": [3.2, 3.1, 3.3, 1.8, 3.4],
   "engine_temp_c": [85, 88, 86, 105, 84],
   "last_maintenance_days": [30, 45, 120, 200, 15]
}
df = pd.DataFrame(fleet_data)
df.to_csv("fleet_logs.csv", index=False)
print("✅ 'fleet_logs.csv' created.")

We generate the dummy fleet dataset that our agent will later analyze. We create a small however sensible set of telemetry fields, convert it right into a DataFrame, and put it aside as a CSV file. Right here, we set up the core knowledge supply that drives the agent’s reasoning and predictions. Try the FULL CODES right here.

class FleetDataTool(Software):
   identify = "load_fleet_logs"
   description = "Hundreds automobile telemetry logs from 'fleet_logs.csv'. Returns the information abstract."
   inputs = {}
   output_type = "string"


   def ahead(self):
       attempt:
           df = pd.read_csv("fleet_logs.csv")
           return f"Columns: {record(df.columns)}nData Pattern:n{df.to_string()}"
       besides Exception as e:
           return f"Error loading logs: {e}"

We outline the FleetDataTool, which acts because the bridge between the agent and the underlying telemetry file. We give the agent the flexibility to load and examine the CSV file to grasp its construction. This device turns into the muse for each subsequent evaluation the mannequin performs. Try the FULL CODES right here.

print("⏳ Downloading & Loading Native Mannequin (approx 60-90s)...")
mannequin = TransformersModel(
   model_id="Qwen/Qwen2.5-Coder-1.5B-Instruct",
   device_map="auto",
   max_new_tokens=2048
)
print("✅ Mannequin loaded on GPU.")


agent = CodeAgent(
   instruments=[FleetDataTool()],
   mannequin=mannequin,
   add_base_tools=True
)


print("n🤖 Agent is analyzing fleet knowledge... (Examine the 'Agent' output under)n")


question = """
1. Load the fleet logs.
2. Discover the truck with the worst gasoline effectivity (lowest 'fuel_efficiency_kml').
3. For that truck, test whether it is overdue for upkeep (threshold is 90 days).
4. Create a bar chart evaluating the 'fuel_efficiency_kml' of ALL vans.
5. Spotlight the worst truck in RED and others in GRAY on the chart.
6. Save the chart as 'maintenance_alert.png'.
"""
response = agent.run(question)


print(f"n📝 FINAL REPORT: {response}")

We load the Qwen2.5 native mannequin and initialize our CodeAgent with the customized device. We then craft an in depth question outlining the reasoning steps we wish the agent to comply with and execute it end-to-end. That is the place we watch the agent assume, analyze, compute, and even plot, absolutely autonomously. Try the FULL CODES right here.

if os.path.exists("maintenance_alert.png"):
   print("n📊 Displaying Generated Chart:")
   img = plt.imread("maintenance_alert.png")
   plt.determine(figsize=(10, 5))
   plt.imshow(img)
   plt.axis('off')
   plt.present()
else:
   print("⚠️ No chart picture discovered. Examine the agent logs above.")

We test whether or not the agent efficiently saved the generated upkeep chart and show it if out there. We visualize the output immediately within the pocket book, permitting us to substantiate that the agent accurately carried out knowledge evaluation and plotting. This offers us a clear, interpretable outcome from your entire workflow.

In conclusion, we constructed an clever end-to-end pipeline that permits a neighborhood mannequin to autonomously load knowledge, consider fleet well being, determine the highest-risk automobile, and generate a diagnostic chart for actionable insights. We witness how simply we will prolong this framework to real-world datasets, combine extra complicated instruments, or add multi-step reasoning capabilities for security, effectivity, or predictive upkeep use circumstances. Ultimately, we respect how SmolAgents empowers us to create sensible agentic programs that execute actual code, cause over actual telemetry, and ship insights instantly.

Try the FULL CODES right here. Additionally, be happy to comply with us on Twitter and don’t overlook to hitch our 100k+ ML SubReddit and Subscribe to our E-newsletter. Wait! are you on telegram? now you may be a part of us on telegram as properly.

Michal Sutter is a knowledge science skilled with a Grasp of Science in Knowledge Science from the College of Padova. With a strong basis in statistical evaluation, machine studying, and knowledge engineering, Michal excels at reworking complicated datasets into actionable insights.