mispeech
/

dashengtokenizer

dashengtokenizer

feature-extraction

audio-classification

signal-processing

Model card Files Files and versions

Heinrich Dinkel commited on Feb 14

Commit

5340ead

·

1 Parent(s): 0f35cd3

updated

Files changed (1) hide show

notebook.ipynb +1 -1

notebook.ipynb CHANGED Viewed

@@ -214,7 +214,7 @@
     "    classifier.to(device)\n",
     "    \n",
     "    # Training setup\n",
-    "    optimizer = torch.optim.Adam(classifier.parameters(), lr=1e-3)\n",
     "    criterion = nn.CrossEntropyLoss()\n",
     "    \n",
     "    # Training loop\n",

     "    classifier.to(device)\n",
     "    \n",
     "    # Training setup\n",
+    "    optimizer = torch.optim.Adam(classifier.parameters(), lr=8e-3)\n",
     "    criterion = nn.CrossEntropyLoss()\n",
     "    \n",
     "    # Training loop\n",