PR tensorflow#510: Fixing dataset directory path and tensorflow 2.0 fit module

Please approve this CL. It will be submitted automatically, and its GitHub pull request will be marked as merged.

Imported from GitHub PR tensorflow#510

Existing code had a issue for dataset directory and also the current version of tensorflow 2.0 deprecated fit_generator and introduced fit method for the same functionality.

Copybara import of the project:

  - b3ac75f Fixing dataset directory path and tensorflow 2.0 fit modu... by sablevinayak <[email protected]>
  - 8af8226 Merge b3ac75f into 16a5a... by Vinayak Sable <[email protected]>

PiperOrigin-RevId: 757085715

Author: tensorflower-gardener

courses/udacity_intro_to_tensorflow_for_deep_learning/l05c01_dogs_vs_cats_without_augmentation.ipynb

@@ -230,7 +230,7 @@
       },
       "outputs": [],
       "source": [
-        "base_dir = os.path.join(os.path.dirname(zip_dir), 'cats_and_dogs_filtered')\n",
+        "base_dir = os.path.join(zip_dir, 'cats_and_dogs_filtered')\n",
         "train_dir = os.path.join(base_dir, 'train')\n",
         "validation_dir = os.path.join(base_dir, 'validation')\n",
         "\n",
@@ -591,7 +591,7 @@
       "outputs": [],
       "source": [
         "EPOCHS = 100\n",
-        "history = model.fit_generator(\n",
+        "history = model.fit(\n",
         "    train_data_gen,\n",
         "    steps_per_epoch=int(np.ceil(total_train / float(BATCH_SIZE))),\n",
         "    epochs=EPOCHS,\n",