File size: 11,391 Bytes
2666ddd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
<br><br>


<!-- Save as: teachable-machine-widget.html -->
<div class="teachable-machine-widget w-full max-w-2xl mx-auto my-8 font-sans bg-zinc-950 text-zinc-100 rounded-2xl overflow-hidden border border-zinc-800 shadow-2xl">
  
  <!-- Dependencies (Load these in your main page <head> if possible) -->
  <script src="https://cdn.tailwindcss.com"></script>
  <script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@4.22.0/dist/tf.min.js"></script>
  <script src="https://cdn.jsdelivr.net/npm/@tensorflow-models/mobilenet@2.1.1"></script>
  <script src="https://cdn.jsdelivr.net/npm/@tensorflow-models/knn-classifier@1.2.6"></script>
  <link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap" rel="stylesheet">

  <style>

  

body {

background-color: black;

}

    /* Scoped styles for the widget */

    .teachable-machine-widget { font-family: 'Inter', sans-serif; }

    .teachable-machine-widget button { touch-action: manipulation; user-select: none; -webkit-user-select: none; }

    .tm-pulse { animation: tm-pulse-ring 1.5s cubic-bezier(0.24, 0, 0.38, 1) infinite; }

    @keyframes tm-pulse-ring {

      0% { box-shadow: 0 0 0 0 rgba(255, 255, 255, 0.2); }

      70% { box-shadow: 0 0 0 6px rgba(255, 255, 255, 0); }

      100% { box-shadow: 0 0 0 0 rgba(255, 255, 255, 0); }

    }

  </style>

  <!-- Widget Header -->
  <div class="px-5 py-4 bg-zinc-900 border-b border-zinc-800 flex items-center justify-between">
    <div class="flex items-center gap-3">
      <div class="w-3 h-3 rounded-full bg-indigo-500 shadow-[0_0_10px_rgba(99,102,241,0.5)]"></div>
      <h2 class="text-sm font-semibold tracking-wide text-zinc-200 uppercase">Teachable Machine</h2>
    </div>
    <div id="tm-status-badge" class="flex items-center gap-2 px-2 py-1 bg-zinc-950 rounded text-[10px] font-medium text-zinc-500 border border-zinc-800">
      <span class="w-1.5 h-1.5 rounded-full bg-zinc-600"></span>
      <span id="tm-status-text">Loading...</span>
    </div>
  </div>

  <!-- Main Content Area -->
  <div class="p-5 bg-zinc-950/50">
    
    <!-- Video & Prediction Overlay Wrapper -->
    <div class="relative rounded-xl overflow-hidden bg-black border border-zinc-800 aspect-video shadow-inner">
      <video id="tm-webcam" class="w-full h-full object-cover transform scale-x-[-1]" autoplay playsinline muted></video>
      
      <!-- Prediction Overlay (Bottom of video) -->
      <div class="absolute bottom-0 left-0 right-0 bg-gradient-to-t from-black/90 to-transparent pt-12 pb-3 px-4 flex items-end justify-between">
        <div>
          <div class="text-[10px] text-zinc-400 uppercase tracking-wider mb-0.5">I see</div>
          <div id="tm-result" class="text-xl font-bold text-white leading-none">...</div>
        </div>
        <div class="text-right">
          <div class="text-[10px] text-zinc-400 uppercase tracking-wider mb-1">Confidence</div>
          <div class="w-24 h-1.5 bg-zinc-800 rounded-full overflow-hidden">
            <div id="tm-conf-bar" class="h-full bg-indigo-500 w-0 transition-all duration-300"></div>
          </div>
        </div>
      </div>
    </div>

    <!-- Controls Compact Grid -->
    <div class="mt-4 grid grid-cols-3 gap-3">
      
      <!-- Button A -->
      <button id="tm-btn-a" class="group relative flex flex-col items-center justify-center p-3 rounded-xl bg-zinc-900 border border-zinc-800 hover:border-red-500/50 hover:bg-zinc-800 transition-all active:scale-[0.98]">
        <div class="absolute inset-0 rounded-xl border-2 border-transparent group-hover:border-red-500/20 pointer-events-none"></div>
        <div class="mb-1 text-xs font-medium text-red-400">Class A</div>
        <div class="text-[10px] text-zinc-500">Red Object</div>
        <div class="mt-2 px-2 py-0.5 bg-zinc-950 rounded text-[10px] text-zinc-300 font-mono border border-zinc-800">
          <span id="tm-count-a">0</span> samples
        </div>
      </button>

      <!-- Button B -->
      <button id="tm-btn-b" class="group relative flex flex-col items-center justify-center p-3 rounded-xl bg-zinc-900 border border-zinc-800 hover:border-blue-500/50 hover:bg-zinc-800 transition-all active:scale-[0.98]">
        <div class="absolute inset-0 rounded-xl border-2 border-transparent group-hover:border-blue-500/20 pointer-events-none"></div>
        <div class="mb-1 text-xs font-medium text-blue-400">Class B</div>
        <div class="text-[10px] text-zinc-500">Blue Object</div>
        <div class="mt-2 px-2 py-0.5 bg-zinc-950 rounded text-[10px] text-zinc-300 font-mono border border-zinc-800">
          <span id="tm-count-b">0</span> samples
        </div>
      </button>

      <!-- Button Idle -->
      <button id="tm-btn-i" class="group relative flex flex-col items-center justify-center p-3 rounded-xl bg-zinc-900 border border-zinc-800 hover:border-zinc-500/50 hover:bg-zinc-800 transition-all active:scale-[0.98]">
        <div class="absolute inset-0 rounded-xl border-2 border-transparent group-hover:border-zinc-500/20 pointer-events-none"></div>
        <div class="mb-1 text-xs font-medium text-zinc-400">Background</div>
        <div class="text-[10px] text-zinc-500">Idle</div>
        <div class="mt-2 px-2 py-0.5 bg-zinc-950 rounded text-[10px] text-zinc-300 font-mono border border-zinc-800">
          <span id="tm-count-i">0</span> samples
        </div>
      </button>

    </div>

    <!-- Footer / Reset -->
    <div class="mt-4 flex items-center justify-between text-[10px] text-zinc-500">
      <p>Hold buttons to train</p>
      <button id="tm-reset-btn" class="hover:text-red-400 transition-colors underline decoration-zinc-800 hover:decoration-red-900">
        Reset Data
      </button>
    </div>
  </div>

  <!-- Logic -->
  <script>

    (function() {

      // Namespace variables to avoid conflicts if you have multiple scripts on the page

      const TM_LABELS = ["Class A", "Class B", "Background"];

      const TM_TOPK = 10;

      let tmNet = null;

      let tmKnn = null;

      let tmIsTraining = false;

      let tmTrainingClass = -1;

      let tmPredicting = false;



      // Elements

      const get = (id) => document.getElementById(id);

      const els = {

        webcam: get('tm-webcam'),

        statusText: get('tm-status-text'),

        statusBadge: get('tm-status-badge'),

        result: get('tm-result'),

        confBar: get('tm-conf-bar'),

        counts: [get('tm-count-a'), get('tm-count-b'), get('tm-count-i')],

        btns: [get('tm-btn-a'), get('tm-btn-b'), get('tm-btn-i')],

        reset: get('tm-reset-btn')

      };



      function setStatus(msg, type="neutral") {

        els.statusText.textContent = msg;

        if(type === 'ready') els.statusBadge.querySelector('span').className = "w-1.5 h-1.5 rounded-full bg-emerald-500 shadow-[0_0_8px_rgba(16,185,129,0.6)]";

        else if(type === 'loading') els.statusBadge.querySelector('span').className = "w-1.5 h-1.5 rounded-full bg-amber-500 animate-pulse";

        else els.statusBadge.querySelector('span').className = "w-1.5 h-1.5 rounded-full bg-zinc-600";

      }



      function updateUI(label, conf) {

        if (!label || label === "—") {

          els.result.textContent = "...";

          els.result.className = "text-xl font-bold text-zinc-500 leading-none";

          els.confBar.style.width = "0%";

          return;

        }



        els.result.textContent = label;

        let color = "text-white";

        let barColor = "bg-indigo-500";

        

        if(label === TM_LABELS[0]) { color = "text-red-400"; barColor = "bg-red-500"; }

        if(label === TM_LABELS[1]) { color = "text-blue-400"; barColor = "bg-blue-500"; }

        if(label === TM_LABELS[2]) { color = "text-zinc-400"; barColor = "bg-zinc-500"; }



        els.result.className = `text-xl font-bold ${color} leading-none transition-colors duration-200`;

        els.confBar.className = `h-full ${barColor} transition-all duration-200`;

        els.confBar.style.width = `${conf}%`;

      }



      function updateCounts() {

        const counts = tmKnn ? tmKnn.getClassExampleCount() : {};

        els.counts.forEach((el, idx) => el.textContent = counts[idx] || 0);

      }



      async function init() {

        try {

          setStatus("Camera...", "loading");

          

          // 1. Setup Webcam

          const stream = await navigator.mediaDevices.getUserMedia({ video: true, audio: false });

          els.webcam.srcObject = stream;

          await new Promise(r => els.webcam.onloadedmetadata = () => r());



          // 2. Load Models

          setStatus("Models...", "loading");

          tmKnn = knnClassifier.create();

          tmNet = await mobilenet.load({ version: 2, alpha: 0.50 });



          // 3. Setup Events

          els.btns.forEach((btn, idx) => {

            const start = (e) => {

              if (e.cancelable) e.preventDefault();

              tmIsTraining = true;

              tmTrainingClass = idx;

              btn.classList.add("tm-pulse", "border-white");

            };

            const stop = (e) => {

              if (e.cancelable) e.preventDefault();

              tmIsTraining = false;

              tmTrainingClass = -1;

              btn.classList.remove("tm-pulse", "border-white");

            };

            btn.addEventListener("mousedown", start);

            btn.addEventListener("mouseup", stop);

            btn.addEventListener("mouseleave", stop);

            btn.addEventListener("touchstart", start, {passive:false});

            btn.addEventListener("touchend", stop, {passive:false});

          });



          els.reset.addEventListener('click', () => {

            if(tmKnn) tmKnn.clearAllClasses();

            updateCounts();

            updateUI(null, 0);

          });



          setStatus("Ready", "ready");

          

          // 4. Loops

          trainLoop();

          predictLoop();

          tmPredicting = true;



        } catch (e) {

          console.error(e);

          setStatus("Error", "error");

        }

      }



      async function trainLoop() {

        while (true) {

          if (tmIsTraining && tmTrainingClass !== -1) {

            const img = tf.browser.fromPixels(els.webcam);

            const activation = tmNet.infer(img, true);

            tmKnn.addExample(activation, tmTrainingClass);

            img.dispose();

            activation.dispose();

            updateCounts();

          }

          await new Promise(r => setTimeout(r, 70)); // Throttle training

        }

      }



      async function predictLoop() {

        while (true) {

          if (tmPredicting && tmKnn.getNumClasses() > 0) {

            const img = tf.browser.fromPixels(els.webcam);

            const activation = tmNet.infer(img, true);

            try {

              const res = await tmKnn.predictClass(activation, TM_TOPK);

              const label = TM_LABELS[res.classIndex];

              const conf = res.confidences[res.classIndex] * 100;

              updateUI(label, conf);

            } catch(e) {}

            img.dispose();

            activation.dispose();

          }

          await tf.nextFrame();

        }

      }



      init();

    })();

  </script>
</div>