Update Part_1_Data_Preperation.ipynb

recipes/quickstart/Multi-Modal-RAG/notebooks/Part_1_Data_Preperation.ipynb

@@ -9,7 +9,7 @@
     "\n",
     "To make the experience consistent, we will use [this link]() for getting access to our dataset. To credit, thanks to the author [here]() for making it available. \n",
     "\n",
-    "The author of this series empathises with all Kagglers aspiring to be Grandmasters: Please upvote the dataset version on Kaggle if you enjoy this course."
+    "As thanks to original author-Please upvote the dataset version on Kaggle if you enjoy this course."
    ]
   },
   {
@@ -43,7 +43,7 @@
     "\n",
     "Let's first download the dataset and set our variables to point to it. \n",
     "\n",
-    "Remember, this is something you will change, don't rush the shift+enter fingers yet! It will come soon!"
+    "Remember, this is something you will change, don't rush the shift+enter fingers yet! Please also set your hf-token in the line below"
    ]
   },
   {
@@ -72,7 +72,7 @@
     "\n",
     "- PIL: For handling images to be passed to our Llama model\n",
     "- Huggingface Tranformers: For running the model\n",
-    "- Concurrent Library: Because 405B suggested its useful for speedups and we want to look smart when doing OS stuff :) "
+    "- Concurrent Library: To look smart when doing OS stuff by using concurrency :) "
    ]
   },
   {
@@ -362,9 +362,7 @@
   {
    "cell_type": "markdown",
    "id": "db3d7f11-e5d2-49e3-a607-188f2f43379c",
-   "metadata": {
-    "jp-MarkdownHeadingCollapsed": true
-   },
+   "metadata": {},
    "source": [
     "## EDA\n",
     "\n",
@@ -1137,48 +1135,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
-   "id": "1de59227-6042-441b-a1f8-b19ce83f7c45",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "USER_TEXT = \"\"\"\n",
-    "You are an expert fashion captioner, we are writing descriptions of clothes, look at the image closely and write a caption for it.\n",
-    "\n",
-    "Write the following Title, Size, Category, Gender, Type, Description in JSON FORMAT, PLEASE DO NOT FORGET JSON, I WILL BE VERY SAD AND CRY\n",
-    "\n",
-    "ALSO START WITH THE JSON AND NOT ANY THING ELSE, FIRST CHAR IN YOUR RESPONSE IS ITS OPENING BRACE, I WILL DRINK CHAI IF YOU FOLLOW THIS\n",
-    "\n",
-    "FOLLOW THESE STEPS CLOSELY WHEN WRITING THE CAPTION: \n",
-    "1. Only start your response with a dictionary like the example below, nothing else, I NEED TO PARSE IT LATER, SO DONT ADD ANYTHING ELSE-IT WILL BREAK MY CODE AND I WILL BE VERY SAD \n",
-    "Remember-DO NOT SAY ANYTHING ELSE ABOUT WHAT IS GOING ON, just the opening brace is the first thing in your response nothing else ok?\n",
-    "2. REMEMBER TO CLOSE THE DICTIONARY WITH '}'BRACE, IT GOES AFTER THE END OF DESCRIPTION-YOU ALWAYS FORGET IT, THIS WILL CAUSE A FIRE ON A PRODUCTION SERVER BEING USE BY MILLIONS\n",
-    "3. If you cant tell the size from image, guess it! its okay but dont literally write that you guessed it\n",
-    "4. Do not make the caption very literal, all of these are product photos, DO NOT CAPTION HOW OR WHERE THEY ARE PLACED, FOCUS ON WRITING ABOUT THE PIECE OF CLOTHING\n",
-    "5. BE CREATIVE WITH THE DESCRIPTION BUT FOLLOW EVERYTHING CLOSELY FOR STRUCTURE\n",
-    "6. Return your answer in dictionary format, see the example below\n",
-    "7. Please do NOT add new lines or tabs in the JSON\n",
-    "8. I REPEAT DO NOT GIVE ME YOUR EXPLAINATION START WITH THE JSON\n",
-    "\n",
-    "{\"Title\": \"Title of item of clothing\", \"Size\": {'S', 'M', 'L', 'XL'}, #select one randomly if you cant tell from the image. DO NOT TELL ME YOU ESTIMATE OR GUESSED IT ONLY THE LETTER IS ENOUGH\", Category\":  {T-Shirt, Shoes, Tops, Pants, Jeans, Shorts, Skirts, Shoes, Footwear}, \"Gender\": {M, F, U}, \"Type\": {Casual, Formal, Work Casual, Lounge}, \"Description\": \"Write it here\"}\n",
-    "\n",
-    "Example: ALWAYS RETURN ANSWERS IN THE DICTIONARY FORMAT BELOW OK?\n",
-    "\n",
-    "{\"Title\": \"Casual White pant with logo on it\", \"size\": \"L\", \"Category\": \"Jeans\", \"Gender\": \"U\", \"Type\": \"Work Casual\", \"Description\": \"Write it here, this is where your stuff goes\"} \n",
-    "\"\"\""
-   ]
-  },
-  {
-   "cell_type": "code",
    "execution_count": 34,
    "id": "ab307328-ad5e-436e-a3d5-30bfb8e24a34",
    "metadata": {},
    "outputs": [],
    "source": [
-    "USER_TEXT_OPTION_2 = \"\"\"\n",
+    "USER_TEXT_OPTION = \"\"\"\n",
     "You are an expert fashion captioner, we are writing descriptions of clothes, look at the image closely and write a caption for it.\n",
     "\n",
-    "Write the following Title, Size, Category, Gender, Type, Description in JSON FORMAT, PLEASE DO NOT FORGET JSON,\n",
+    "Write the following Title, Size, Category, Gender, Type, Description in JSON FORMAT, PLEASE DO NOT FORGET JSON, \n",
     "\n",
     "ALSO START WITH THE JSON AND NOT ANY THING ELSE, FIRST CHAR IN YOUR RESPONSE IS ITS OPENING BRACE\n",
     "\n",
@@ -1793,7 +1758,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.5"
+   "version": "3.11.10"
   }
  },
  "nbformat": 4,