Spaces:

nanotron
/

ultrascale-playbook

Running

nouamanetazi HF staff commited on 7 days ago

Commit

0a8f3ab

verified ·

1 Parent(s): ce06f49

embed predict_memory (#47)

Files changed (5) hide show

assets/images/predict_memory_tool.png ADDED Viewed

assets/images/profile_trace_annotated.png CHANGED Viewed

dist/assets/images/predict_memory_tool.png ADDED Viewed

dist/index.html CHANGED Viewed

@@ -226,19 +226,12 @@
         </div>
         <p>(Don't worry if you have no idea what's happening in this widget. That's why we're here.)</p>
-        <p>While this widget gives a theoretical breakdown the following tool can be used to predict the memory usage:</p>
-        <ul>
-          <li>
-            <p>
-              <a href="https://huggingface.co/spaces/nanotron/predict_memory">predict_memory</a>
-            </p>
-          </li>
-          <li>
-            <p>
-              <a href="https://pytorch.org/docs/stable/torch_cuda_memory.html">torch_cuda_memory</a>
-            </p>
-          </li>
-        </ul>
         <p><strong>Clear code implementations:</strong> theory is one thing, but we discover all kinds of edge cases and important details when we implement something. That’s why we link to implementation references where possible. Depending on the case, we’ll use two code references:</p>

         </div>
         <p>(Don't worry if you have no idea what's happening in this widget. That's why we're here.)</p>
+        <p>While this widget gives a theoretical breakdown we also made the <a href="https://huggingface.co/spaces/nanotron/predict_memory">following tool</a> that can be used to predict the memory usage during a training run:</p>
+        <a href="https://huggingface.co/spaces/nanotron/predict_memory">
+            <img src="/assets/images/predict_memory_tool.png" alt="Predict Memory Tool" />
+        </a>
         <p><strong>Clear code implementations:</strong> theory is one thing, but we discover all kinds of edge cases and important details when we implement something. That’s why we link to implementation references where possible. Depending on the case, we’ll use two code references:</p>

src/index.html CHANGED Viewed

@@ -226,19 +226,12 @@
         </div>
         <p>(Don't worry if you have no idea what's happening in this widget. That's why we're here.)</p>
-        <p>While this widget gives a theoretical breakdown the following tool can be used to predict the memory usage:</p>
-        <ul>
-          <li>
-            <p>
-              <a href="https://huggingface.co/spaces/nanotron/predict_memory">predict_memory</a>
-            </p>
-          </li>
-          <li>
-            <p>
-              <a href="https://pytorch.org/docs/stable/torch_cuda_memory.html">torch_cuda_memory</a>
-            </p>
-          </li>
-        </ul>
         <p><strong>Clear code implementations:</strong> theory is one thing, but we discover all kinds of edge cases and important details when we implement something. That’s why we link to implementation references where possible. Depending on the case, we’ll use two code references:</p>

         </div>
         <p>(Don't worry if you have no idea what's happening in this widget. That's why we're here.)</p>
+        <p>While this widget gives a theoretical breakdown we also made the <a href="https://huggingface.co/spaces/nanotron/predict_memory">following tool</a> that can be used to predict the memory usage during a training run:</p>
+        <a href="https://huggingface.co/spaces/nanotron/predict_memory">
+            <img src="/assets/images/predict_memory_tool.png" alt="Predict Memory Tool" />
+        </a>
         <p><strong>Clear code implementations:</strong> theory is one thing, but we discover all kinds of edge cases and important details when we implement something. That’s why we link to implementation references where possible. Depending on the case, we’ll use two code references:</p>