update gptq tutorial

tutorials/notebooks/mct_features_notebooks/pytorch/example_pytorch_mobilenet_gptq.ipynb

@@ -176,8 +176,7 @@
     "GPTQ is a gradient-based optimization process, which requires representative dataset to perform inference and compute gradients. It is possible to define a separate representative dataset in addition to the one used for the PTQ statistics collection. A possible reason to do so is, for example, to use a larger dataset in the optimization process.\n",
     "\n",
     "In this tutorial we use the same representative dataset for both statistics collection and GPTQ. \n",
-    "A complete pass through the representative dataset generator constitutes an epoch (batch_size x n_iter samples).\n",
-    "In this example we use the same dataloader iterator for all epochs, i.e. different images are used in different epochs."
+    "A complete pass through the representative dataset generator constitutes an epoch (batch_size x n_iter samples)."
    ]
   },
   {
@@ -193,9 +192,9 @@
     "n_iter = 10\n",
     "\n",
     "dataloader = DataLoader(dataset, batch_size=batch_size, shuffle=True)\n",
-    "dataloader_iter = iter(dataloader)\n",
     "\n",
     "def representative_dataset_gen():\n",
+    "    dataloader_iter = iter(dataloader)\n",
     "    for _ in range(n_iter):\n",
     "        yield [next(dataloader_iter)[0]]\n"
    ]