Upload fusion_t2i_CLIP_interrogator.ipynb

Google Colab Notebooks/fusion_t2i_CLIP_interrogator.ipynb

@@ -389,21 +389,21 @@
         "# @title \t⚄ Use a pre-encoded prompt + image pair from the fusion gen (note: NSFW!)\n",
         "# @markdown 🖼️ Choose a pre-encoded reference\n",
         "index = 708 # @param {type:\"slider\", min:0, max:1666, step:1}\n",
-        "\n",
         "PROMPT_INDEX = index\n",
-        "\n",
-        "# @markdown ⚖️ Set the value for C in the reference  <br> <br> sim = C* text_enc + image_enc*(1-C) <br><br>\n",
-        "\n",
+        "# @markdown ⚖️ Set the value for C in the reference  <br> <br> sim = C* text_enc + image_enc*(1-C) <br>\n",
         "C = 0.5 # @param {type:\"slider\", min:0, max:1, step:0.01}\n",
+        "# @markdown -----------\n",
+        "# @markdown 📝 Enhance similarity to this prompt(optional)\n",
+        "POS = '' # @param {type:'string'}\n",
+        "strength = 1 # @param {type:\"slider\", min:-5, max:5, step:0.1}\n",
+        "pos_strength = strength\n",
+        "# @markdown -----------\n",
         "\n",
         "# @markdown 🚫 Penalize similarity to this prompt(optional)\n",
-        "\n",
         "NEG = '' # @param {type:'string'}\n",
-        "strength = 1 # @param {type:\"slider\", min:-5, max:5, step:0.1}\n",
-        "\n",
-        "# @markdown Calculate most similiar items using above settings?\n",
-        "\n",
-        "\n",
+        "strength = 1 # @param {type:\"slider\", min:-50, max:50, step:0.1}\n",
+        "neg_strength = strength\n",
+        "# @markdown -----------\n",
         "# @title ⚙️📝 Print the results (Advanced)\n",
         "list_size = 1000 # param {type:'number'}\n",
         "start_at_index = 0 # param {type:'number'}\n",
@@ -413,9 +413,9 @@
         "print_Descriptions = True # param {type:\"boolean\"}\n",
         "compact_Output = True # param {type:\"boolean\"}\n",
         "\n",
-        "# @markdown -----------\n",
-        "# @markdown 📝 Printing options\n",
-        "newline_Separator = False # @param {type:\"boolean\"}\n",
+        "\n",
+        "# markdown Printing options\n",
+        "newline_Separator = False # param {type:\"boolean\"}\n",
         "\n",
         "import random\n",
         "list_size2 = 1000 # param {type:'number'}\n",
@@ -482,12 +482,30 @@
         "    neg_sims =  torch.matmul(text_tensor, text_features_NEG.t())\n",
         "  #------#\n",
         "\n",
+        "  pos_sims = 0*sims\n",
+        "  if(POS != ''):\n",
+        "\n",
+        "    # Get text features for user input\n",
+        "    inputs = tokenizer(text = POS, padding=True, return_tensors=\"pt\")\n",
+        "    text_features_POS = model.get_text_features(**inputs)\n",
+        "    text_features_POS = text_features_POS/text_features_POS.norm(p=2, dim=-1, keepdim=True)\n",
+        "\n",
+        "    # text-similarity\n",
+        "    pos_sims =  torch.matmul(text_tensor, text_features_POS.t())\n",
+        "  #------#\n",
+        "\n",
         "  # plus image-similarity\n",
-        "  sims = sims +  (1-C) * torch.matmul(text_tensor, image_features_A.t()) * logit_scale\n",
+        "  img_sims = torch.matmul(text_tensor, image_features_A.t()) * logit_scale\n",
+        "  sims = sims +  (1-C) * img_sims\n",
+        "\n",
+        "\n",
+        "  # plus POS-similarity\n",
+        "  sims = sims + pos_strength*pos_sims\n",
         "\n",
         "\n",
         "  # minus NEG-similarity\n",
-        "  sims = sims - strength*neg_sims\n",
+        "  sims = sims - neg_strength*neg_sims\n",
+        "\n",
         "\n",
         "  # Sort the items\n",
         "  sorted , indices = torch.sort(sims,dim=0 , descending=True)\n",