%PDF-1.4 %âãÏÓ 1 0 obj << /Type /Catalog /Pages 2 0 R >> endobj 2 0 obj << /Type /Pages /Count 3 /Kids [5 0 R 7 0 R 9 0 R] >> endobj 3 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >> endobj 4 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica-Bold >> endobj 5 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 6 0 R >> endobj 6 0 obj << /Length 4814 >> stream BT /F2 22 Tf 0.06 0.08 0.12 rg 1 0 0 1 46 789.89 Tm (How to Setup TinyML on Raspberry Pi Pico - Quick) Tj ET BT /F2 22 Tf 0.06 0.08 0.12 rg 1 0 0 1 46 762.89 Tm (Steps) Tj ET BT /F2 11 Tf 0.72 0.14 0.18 rg 1 0 0 1 46 725.89 Tm (TechRounder PDF Edition) Tj ET BT /F1 9.5 Tf 0.36 0.39 0.46 rg 1 0 0 1 46 709.89 Tm (Live article: https://www.techrounder.com/how-to/how-to-setup-tinyml-on-raspberry-pi-pico-quick-steps/) Tj ET q 0.82 0.85 0.9 RG 1 w 46 691.39 m 549.28 691.39 l S Q BT /F1 10 Tf 0.24 0.27 0.32 rg 1 0 0 1 46 679.39 Tm (By Vipin PG | Published August 28, 2025 | Updated January 4, 2026 | Format: Guide | 4 min read) Tj ET BT /F2 13 Tf 0.72 0.14 0.18 rg 1 0 0 1 46 656.39 Tm (Quick answer) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 636.39 Tm (Artificial Intelligence \(AI\) is no longer limited to powerful servers and cloud platforms. With the rise of) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 621.39 Tm (TinyML \(Tiny Machine Learning\), we can now run machine learning models directly on small devices) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 606.39 Tm (like microcontrollers.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 581.39 Tm (Artificial Intelligence \(AI\) is no longer limited to powerful servers and cloud platforms. With the rise of) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 566.39 Tm (TinyML \(Tiny Machine Learning\), we can now run machine learning models directly on small devices) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 551.39 Tm (like microcontrollers. Imagine your motion sensor recognizing gestures, your smart gadget detecting) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 536.39 Tm (voice commands, or a tiny board analyzing patterns-all without needing internet or cloud support.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 521.39 Tm (That's the magic of TinyML.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 499.39 Tm (One of the best boards to start this journey is the Raspberry Pi Pico, a low-cost, low-power) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 484.39 Tm (microcontroller that's easy to use and ideal for beginners. In this article, we'll help you to understand) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 469.39 Tm (everything you need to set up TinyML on Raspberry Pi Pico step by step-even if you're completely) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 454.39 Tm (new to the subject.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 432.39 Tm (By the end, you'll know how to prepare your Pico, install the right libraries, run sample models, and) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 417.39 Tm (even explore simple projects.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 389.39 Tm (What You Need Before Starting) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 365.39 Tm (Before diving in, make sure you have the following:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 343.39 Tm (- Raspberry Pi Pico or Pico W \(wireless version works too\)) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 326.59 Tm (- Micro-USB cable to connect Pico to your computer) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 309.79 Tm (- Computer with internet access \(Windows, macOS, or Linux\)) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 292.99 Tm (- Arduino IDE \(or Thonny if you prefer MicroPython\)) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 276.19 Tm (- TinyML libraries \(TensorFlow Lite for Microcontrollers\)) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 259.39 Tm (- Optional sensors : microphone, accelerometer, or temperature sensor \(for advanced projects\)) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 242.59 Tm (Tip: Stick to the basics first. Start with the "Hello World" TinyML model before experimenting with) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 227.59 Tm (sensors.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 199.59 Tm (Step 1: Setting Up the Raspberry Pi Pico) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 175.59 Tm (1. Download and install Arduino IDE) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 158.79 Tm (- Visit the Arduino website.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 141.99 Tm (- Download the latest version for your operating system.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 125.19 Tm (- Install it like any other application.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 108.39 Tm (2. Add Raspberry Pi Pico support) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 91.59 Tm (- Open Arduino IDE -> go to File -> Preferences .) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 1 of 3) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/how-to-setup-tinyml-on-raspberry-pi-pico-quick-steps.pdf) Tj ET endstream endobj 7 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 8 0 R >> endobj 8 0 obj << /Length 4815 >> stream BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 789.89 Tm (- In "Additional Board Manager URLs," add:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 776.09 Tm (`https://github.com/earlephilhower/arduino-pico/releases/download/global/package_rp2040_index.json) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 762.29 Tm (`) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 745.49 Tm (- Go to Tools -> Board -> Board Manager , search for "Raspberry Pi RP2040," and install it.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 728.69 Tm (3. Connect Pico to your computer) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 711.89 Tm (- Hold the BOOTSEL button on the Pico while plugging it into your PC.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 695.09 Tm (- It will appear as a USB drive.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 678.29 Tm (- Once installed, select the correct port in Arduino IDE under Tools -> Port .) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 661.49 Tm (Your Raspberry Pi Pico is now ready to be programmed with Arduino IDE.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 633.49 Tm (Step 2: Installing TinyML Libraries) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 609.49 Tm (Now let's bring machine learning into the picture.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 587.49 Tm (1. In Arduino IDE, go to Sketch -> Include Library -> Manage Libraries .) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 570.69 Tm (2. Search for TensorFlow Lite for Microcontrollers .) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 553.89 Tm (3. Install the library.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 537.09 Tm (This library allows the Pico to run pre-trained ML models in a very lightweight way.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 509.09 Tm (Step 3: Running a Sample TinyML Model) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 485.09 Tm (To make sure everything works, let's start with the Hello World example provided by TensorFlow Lite.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 463.09 Tm (1. Go to File -> Examples -> Arduino_TensorFlowLite -> hello_world .) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 446.29 Tm (2. Upload the sketch to your Pico.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 429.49 Tm (3. Open Serial Monitor in Arduino IDE.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 412.69 Tm (You should see a sine wave pattern being generated. This might look simple, but it proves your Pico can) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 397.69 Tm (now run a neural network model.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 375.69 Tm (Why sine waves? It's a lightweight test model that checks if your board and libraries are working) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 360.69 Tm (correctly before moving on to more complex models.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 332.69 Tm (Step 4: Running an Interactive Example \(Optional\)) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 308.69 Tm (Once you're comfortable, try running a gesture recognition or keyword spotting model.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 286.69 Tm (- Gesture Recognition:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 269.89 Tm (- Use an accelerometer with your Pico.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 253.09 Tm (- Load the example `magic_wand` project from TensorFlow Lite.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 236.29 Tm (- Move your board in shapes \(like "O" or "Z"\) and watch the model recognize gestures.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 219.49 Tm (- Keyword Spotting:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 202.69 Tm (- With a microphone attached, train the board to detect simple words like "yes" or "no.") Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 62 185.89 Tm (- Useful for building voice-activated IoT projects .) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 163.09 Tm (Step 5: Training Your Own Model \(Optional Advanced Step\)) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 139.09 Tm (If you're curious, you can also train your own machine learning model:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 117.09 Tm (1. Use TensorFlow on your PC to train a model \(for example, detecting different sensor patterns\).) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 100.29 Tm (2. Convert the model to TensorFlow Lite format .) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 83.49 Tm (3. Upload it to your Pico with Arduino IDE.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 2 of 3) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/how-to-setup-tinyml-on-raspberry-pi-pico-quick-steps.pdf) Tj ET endstream endobj 9 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 10 0 R >> endobj 10 0 obj << /Length 2966 >> stream BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 789.89 Tm (This way, you're not limited to pre-built examples-you can design custom AI projects.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 761.89 Tm (Troubleshooting Tips) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 737.89 Tm (- Library not found? Double-check that "TensorFlow Lite for Microcontrollers" is installed via Library) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 724.09 Tm (Manager.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 707.29 Tm (- Upload fails? Reconnect Pico in BOOTSEL mode and reselect the port in Arduino IDE.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 690.49 Tm (- Out of memory error? Start with smaller models. The Pico has limited RAM, so avoid large neural) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 676.69 Tm (networks.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 653.89 Tm (Practical Use Cases of TinyML on Pico) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 629.89 Tm (Here are some exciting things you can build once TinyML is up and running:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 607.89 Tm (- Smart Sensors -> Detect unusual motion, vibrations, or temperature patterns.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 591.09 Tm (- Voice Commands -> Control appliances with basic keyword recognition.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 574.29 Tm (- Gesture Control -> Use hand movements to control lights or gadgets.) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 557.49 Tm (- Wearables -> Low-power activity tracking without cloud dependence.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 534.69 Tm (Conclusion) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 510.69 Tm (You've just learned how to set up TinyML on Raspberry Pi Pico from scratch. From installing the) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 495.69 Tm (Arduino IDE to running your first TinyML model, the process is surprisingly simple once broken into) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 480.69 Tm (steps. With a bit of practice, you can train your own models and build creative AI-powered projects) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 465.69 Tm (right on a tiny board.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 443.69 Tm (TinyML is opening doors for AI at the edge, making it possible to bring intelligence into everyday) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 428.69 Tm (devices without heavy hardware or constant internet access. Whether you're a student, hobbyist, or) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 413.69 Tm (developer, the Raspberry Pi Pico is a perfect playground to explore this new frontier.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 3 of 3) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/how-to-setup-tinyml-on-raspberry-pi-pico-quick-steps.pdf) Tj ET endstream endobj xref 0 11 0000000000 65535 f 0000000015 00000 n 0000000064 00000 n 0000000133 00000 n 0000000203 00000 n 0000000278 00000 n 0000000420 00000 n 0000005285 00000 n 0000005427 00000 n 0000010293 00000 n 0000010436 00000 n trailer << /Size 11 /Root 1 0 R >> startxref 13454 %%EOF