Upload sd_token_similarity_calculator.ipynb

sd_token_similarity_calculator.ipynb

@@ -122,41 +122,30 @@
         "!git clone https://huggingface.co/datasets/codeShare/text-to-image-prompts\n"
       ],
       "metadata": {
-        "id": "rUXQ73IbonHY",
-        "outputId": "5ce85c43-136a-49b5-f316-a74e70085e88",
-        "colab": {
-          "base_uri": "https://localhost:8080/"
-        }
+        "id": "rUXQ73IbonHY"
       },
-      "execution_count": 1,
-      "outputs": [
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "/content\n",
-            "Cloning into 'text-to-image-prompts'...\n",
-            "remote: Enumerating objects: 521, done.\u001b[K\n",
-            "remote: Counting objects: 100% (518/518), done.\u001b[K\n",
-            "remote: Compressing objects: 100% (492/492), done.\u001b[K\n",
-            "remote: Total 521 (delta 91), reused 0 (delta 0), pack-reused 3 (from 1)\u001b[K\n",
-            "Receiving objects: 100% (521/521), 2.51 MiB | 6.14 MiB/s, done.\n",
-            "Resolving deltas: 100% (91/91), done.\n",
-            "Filtering content: 100% (95/95), 305.98 MiB | 37.71 MiB/s, done.\n"
-          ]
-        }
-      ]
+      "execution_count": null,
+      "outputs": []
     },
     {
       "cell_type": "code",
       "source": [
         "# @title ✳️ Select items for the vocab\n",
         "\n",
-        "prompt_features = True # @param {\"type\":\"boolean\",\"placeholder\":\"🦜\"}\n",
-        "civitai_blue_set = True # @param {\"type\":\"boolean\",\"placeholder\":\"📘\"}\n",
-        "suffix = True # @param {\"type\":\"boolean\",\"placeholder\":\"🔹\"}\n",
+        "prompt_features = False # @param {\"type\":\"boolean\",\"placeholder\":\"🦜\"}\n",
+        "civitai_blue_set = False # @param {\"type\":\"boolean\",\"placeholder\":\"📘\"}\n",
+        "suffix = False # @param {\"type\":\"boolean\",\"placeholder\":\"🔹\"}\n",
         "prefix = False # @param {\"type\":\"boolean\",\"placeholder\":\"🔸\"}\n",
         "emojis = True # @param {\"type\":\"boolean\",\"placeholder\":\"😃\"}\n",
+        "#------#\n",
+        "\n",
+        "#These are borked\n",
+        "celebs = False # param {\"type\":\"boolean\",\"placeholder\":\"🆔👨\"}\n",
+        "celebs_young = False # param {\"type\":\"boolean\",\"placeholder\":\"🔸\"}\n",
+        "#-------#\n",
+        "\n",
+        "#-----#\n",
+        "female_fullnames = True # @param {\"type\":\"boolean\",\"placeholder\":\"😃\"}\n",
         "debug = False\n",
         "\n",
         "#------#\n",
@@ -165,30 +154,38 @@
         "nA = 0\n",
         "#--------#\n",
         "\n",
+        "\n",
+        "if celebs:\n",
+        "  url = '/content/text-to-image-prompts/names/celebs/mixed'\n",
+        "  prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
+        "#--------#\n",
+        "\n",
+        "if celebs_young :\n",
+        "  url = '/content/text-to-image-prompts/names/celebs/young'\n",
+        "  prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
+        "#--------#\n",
+        "\n",
+        "if female_fullnames:\n",
+        "  url = '/content/text-to-image-prompts/names/fullnames'\n",
+        "  prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
+        "#--------#\n",
+        "\n",
+        "\n",
         "if prompt_features:\n",
         "  url = '/content/text-to-image-prompts/civitai-prompts/green'\n",
         "  prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
-        "  if debug:\n",
-        "    print(prompts[f'{nA}'])\n",
-        "    print(text_encodings[f'{nA}'])\n",
         "#--------#\n",
         "\n",
         "\n",
         "if emojis:\n",
         "  url = '/content/text-to-image-prompts/tokens/emoji'\n",
         "  prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
-        "  if debug:\n",
-        "    print(prompts[f'{nA}'])\n",
-        "    print(text_encodings[f'{nA}'])\n",
         "#--------#\n",
         "\n",
         "\n",
         "if civitai_blue_set:\n",
         "  url = '/content/text-to-image-prompts/civitai-prompts/blue'\n",
         "  prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
-        "  if debug:\n",
-        "    print(prompts[f'{nA}'])\n",
-        "    print(text_encodings[f'{nA}'])\n",
         "#--------#\n",
         "\n",
         "if suffix :\n",
@@ -196,9 +193,6 @@
         "  for item in ['common','average','rare','weird','exotic'] :\n",
         "    url =  tmp  + item\n",
         "    prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '')\n",
-        "    if debug:\n",
-        "      print(prompts[f'{nA}'])\n",
-        "      print(text_encodings[f'{nA}'])\n",
         "#------#\n",
         "\n",
         "if prefix :\n",
@@ -206,9 +200,6 @@
         "  for item in ['common','average','rare','weird','exotic'] :\n",
         "    url =  tmp  + item\n",
         "    prompts , text_encodings, nA = append_from_url(prompts , text_encodings, nA , url , '-')\n",
-        "    if debug:\n",
-        "      print(prompts[f'{nA}'])\n",
-        "      print(text_encodings[f'{nA}'])\n",
         "#------#\n",
         "\n",
         "if debug:\n",
@@ -232,7 +223,7 @@
       "cell_type": "code",
       "source": [
         "# @title 📝 Get Prompt text_encoding similarity to the pre-calc. text_encodings\n",
-        "prompt = \" a fast car on the road \" # @param {\"type\":\"string\",\"placeholder\":\"Write a prompt\"}\n",
+        "prompt = \"Beyonce and Lady Gaga\" # @param {\"type\":\"string\",\"placeholder\":\"Write a prompt\"}\n",
         "\n",
         "from transformers import AutoTokenizer\n",
         "tokenizer = AutoTokenizer.from_pretrained(\"openai/clip-vit-large-patch14\", clean_up_tokenization_spaces = False)\n",
@@ -259,7 +250,7 @@
       "metadata": {
         "id": "xc-PbIYF428y"
       },
-      "execution_count": null,
+      "execution_count": 9,
       "outputs": []
     },
     {
@@ -368,13 +359,13 @@
       ],
       "metadata": {
         "id": "ke6mZ1RZDOeB",
-        "outputId": "d8ef4589-8393-4001-ff35-c0c30646a576",
+        "outputId": "4e2c2845-a26f-4966-b560-51b2bd059192",
         "colab": {
           "base_uri": "https://localhost:8080/",
           "height": 1000
         }
       },
-      "execution_count": null,
+      "execution_count": 11,
       "outputs": [
         {
           "output_type": "display_data",
@@ -418,7 +409,7 @@
       "metadata": {
         "id": "rebogpoyOG8k"
       },
-      "execution_count": null,
+      "execution_count": 12,
       "outputs": []
     },
     {
@@ -426,7 +417,7 @@
       "source": [
         "# @title 🖼️ Print the results\n",
         "list_size = 100 # @param {type:'number'}\n",
-        "start_at_index = 100 # @param {type:'number'}\n",
+        "start_at_index = 0 # @param {type:'number'}\n",
         "print_Similarity = True # @param {type:\"boolean\"}\n",
         "print_Prompts = True # @param {type:\"boolean\"}\n",
         "print_Prefix = True # @param {type:\"boolean\"}\n",
@@ -466,10 +457,121 @@
         "#-------#"
       ],
       "metadata": {
-        "id": "JkzncP8SgKtS"
+        "id": "JkzncP8SgKtS",
+        "outputId": "16a3365b-1ae4-4ce3-b4aa-9f3a03e87735",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        }
       },
-      "execution_count": null,
-      "outputs": []
+      "execution_count": 13,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "{Tabytha Christion |\n",
+            "Neera Stimpert |\n",
+            "Eris Ororke |\n",
+            "Meelah Onan |\n",
+            "Khalia Okins |\n",
+            "Niajah Olshan |\n",
+            "Aariel Gutterman |\n",
+            "Indyia Orsborn |\n",
+            "Mairin Elfstrom |\n",
+            "Iliani Roshan |\n",
+            "Sapphire Bargstadt |\n",
+            "Maelin Kiel |\n",
+            "Janeah Easterwood |\n",
+            "Eris Kindelan |\n",
+            "Zanari Guillette |\n",
+            "Kree Chotibai |\n",
+            "Khaia Dutter |\n",
+            "Millennia Ekman |\n",
+            "Saphirah Kess |\n",
+            "Kahlan Romanczuk |\n",
+            "Jaza Rowton |\n",
+            "Niara Bargneare |\n",
+            "Janiaya Gudgel |\n",
+            "Antigone Kiddle |\n",
+            "Sariana Dyreson |\n",
+            "Sigourney Mamula |\n",
+            "Starlett Guill |\n",
+            "Yordanos Oien |\n",
+            "Halia Charadan |\n",
+            "Nephtalie Olsby |\n",
+            "Cortana Bartle |\n",
+            "Jaeda Echof |\n",
+            "Esthela Khalsa |\n",
+            "Varya Stimits |\n",
+            "Amaurie Romanok |\n",
+            "Saran Startz |\n",
+            "Xitlali Killian |\n",
+            "Karma Marchildon |\n",
+            "Liberti Olivars |\n",
+            "Aleen Rotch |\n",
+            "Ryenn Stanaland |\n",
+            "Jarae Grun |\n",
+            "Daja Orsburn |\n",
+            "Sariana Gwyn |\n",
+            "Roxann Gyllensten |\n",
+            "Yerania Stivason |\n",
+            "Karmyn Oroark |\n",
+            "Kaiulani Gulreg |\n",
+            "Clary Barbre |\n",
+            "Zamiya Guardarrama |\n",
+            "Aquila Roloson |\n",
+            "Jashara Kernan |\n",
+            "Serenity Barer |\n",
+            "Melynna Dygert |\n",
+            "Chakira Olvey |\n",
+            "Avagrace Rought |\n",
+            "Halo Barne |\n",
+            "Malori Guljar |\n",
+            "Tashai Barvosa |\n",
+            "Rani Okeson |\n",
+            "Ariadne Barias |\n",
+            "Zoraida Romanoff |\n",
+            "Loghan Orozeo |\n",
+            "Zurii Kesler |\n",
+            "Aseret Battani |\n",
+            "Genesi Elion |\n",
+            "Rin Stirna |\n",
+            "Kitara Gutshall |\n",
+            "Xiamara Omo |\n",
+            "Zamiya Barbur |\n",
+            "Diem Guerrera |\n",
+            "Janecia Choma |\n",
+            "Rain Eddens |\n",
+            "Neela Barella |\n",
+            "Zhane Gulfan |\n",
+            "Aeva Omeara |\n",
+            "Krosby Barbour |\n",
+            "Lilith Chitu |\n",
+            "Zerah Rogg |\n",
+            "Venus Roupe |\n",
+            "Nuriya Ogg |\n",
+            "Natallie Marabella |\n",
+            "Amyrah Elfrink |\n",
+            "Hailynn Gulnaz |\n",
+            "Zaniah Orrick |\n",
+            "Naia Gunanidhi |\n",
+            "Tanayjah Barryman |\n",
+            "Noomi Stavrides |\n",
+            "Lanesha Groot |\n",
+            "Daniah Stokan |\n",
+            "Avaeh Chanchuan |\n",
+            "Miaya Groote |\n",
+            "Kora Chebahtah |\n",
+            "Jessabelle Rosenhagen |\n",
+            "Zyrah Guley |\n",
+            "Eliette Guardipee |\n",
+            "Myha Charleton |\n",
+            "Ishitha Steckelberg |\n",
+            "Ayrionna Rothfuss |\n",
+            "Honor Olan }\n"
+          ]
+        }
+      ]
     },
     {
       "cell_type": "code",
@@ -1346,7 +1448,7 @@
           "base_uri": "https://localhost:8080/"
         }
       },
-      "execution_count": 5,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",