Upload folder using huggingface_hub

.env.example

@@ -0,0 +1 @@
+REPLICATE_API_TOKEN=

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+flagged/moodboard/06dd87dd9bbe75468871/Screenshot[[:space:]]2024-06-11[[:space:]]at[[:space:]]5.36.16PM.png filter=lfs diff=lfs merge=lfs -text
+flagged/moodboard/14c7c3a52cd71affacbe/Screenshot[[:space:]]2024-06-11[[:space:]]at[[:space:]]5.27.28PM.png filter=lfs diff=lfs merge=lfs -text
+flagged/moodboard/20a9ebffee3a984e4e92/Screenshot[[:space:]]2024-06-11[[:space:]]at[[:space:]]5.27.28PM.png filter=lfs diff=lfs merge=lfs -text
+flagged/moodboard/98d40109371f437df727/Screenshot[[:space:]]2024-06-11[[:space:]]at[[:space:]]5.36.16PM.png filter=lfs diff=lfs merge=lfs -text
+flagged/moodboard/fde8c7b6d7cdf2100e2e/Screenshot[[:space:]]2024-06-11[[:space:]]at[[:space:]]5.27.28PM.png filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,13 @@
+.vercel
+*.log
+*.pyc
+__pycache__
+
+# Environments 
+.env 
+.venv 
+env/ 
+venv/ 
+ENV/ 
+env.bak/ 
+venv.bak/ 

README.md

@@ -1,12 +1,34 @@
 ---
-title: Moodboard Gen Prompt
-emoji: 👀
-colorFrom: red
-colorTo: pink
+title: moodboard-gen-prompt
+app_file: api/index.py
 sdk: gradio
 sdk_version: 4.36.1
-app_file: app.py
-pinned: false
 ---
+# Alt Image Generator
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+[![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https://github.com/nutlope/alt-text-generator&env=REPLICATE_API_KEY&project-name=alt-tag-generator&repo-name=alt-tag-generator)
+
+This Flask API will generate a description for any image using AI. If you're looking for the TypeScript version, [click here](https://github.com/vercel/examples/tree/main/solutions/alt-tag-generator).
+
+![Alt Image Generator](ogimage.png)
+
+## How it works
+
+This project uses an ML modal from Salesforce called [BLIP](https://github.com/salesforce/BLIP) on [Replicate](https://replicate.com/) to generate relevant alt text for images. You can feed the Flask API endpoint an image as a query param and it will return a one sentence description of that image.
+
+## Running Locally
+
+After cloning the repo, go to [Replicate](https://replicate.com/) to make an account and put your API key in `.env`.
+
+Then, run the following in the command line and your application will be available at `http://localhost:3000`
+
+```bash
+npm i -g vercel
+vercel dev
+```
+
+To use the API route, go to the link below in your browser or run a curl command in your terminal to get a sample result. Feel free to replace the dub.sh link with a link to any image.
+
+```bash
+curl http://localhost:3000/generate?imageUrl=https://dub.sh/confpic
+```

api/index.py

@@ -0,0 +1,98 @@
+import os
+import base64
+import numpy as np
+from PIL import Image
+import io
+import requests
+
+import replicate
+from flask import Flask, request
+import gradio as gr
+from openai import OpenAI
+
+from dotenv import load_dotenv, find_dotenv
+
+# Locate the .env file
+dotenv_path = find_dotenv()
+
+load_dotenv(dotenv_path)
+
+OPENAI_API_KEY = os.getenv('OPENAI_API_KEY')
+REPLICATE_API_TOKEN = os.getenv('REPLICATE_API_TOKEN')
+
+
+
+client = OpenAI()
+
+
+def call_openai(pil_image):
+    # Save the PIL image to a bytes buffer
+    buffered = io.BytesIO()
+    pil_image.save(buffered, format="JPEG")
+    
+    # Encode the image to base64
+    image_data = base64.b64encode(buffered.getvalue()).decode('utf-8')
+    
+    response = client.chat.completions.create(
+        model="gpt-4o",
+        messages=[
+            {
+                "role": "user",
+                "content": [
+                    {"type": "text", "text": "You are a product designer. I've attached a moodboard here. In one sentence, what do all of these elements have in common? Answer from a design language perspective, if you were telling another designer to create something similar, including any repeating colors and materials and shapes and textures"},
+                    {
+                        "type": "image_url",
+                        "image_url": {
+                            "url": "data:image/jpeg;base64," + image_data,
+                        },
+                    },
+                ],
+            }
+        ],
+        max_tokens=300,
+    )
+    
+    return response.choices[0].message.content
+
+def image_classifier(moodboard, prompt):
+
+    # Convert the numpy array to a PIL image
+    pil_image = Image.fromarray(moodboard.astype('uint8'))
+    
+    openai_response = call_openai(pil_image)
+    openai_response = openai_response.replace('moodboard', '')
+        
+    # Call Stable Diffusion API with the response from OpenAI
+    input = {
+        "width": 768,
+        "height": 768,
+        "prompt": "high quality render of " + prompt + ", " + openai_response[20:],
+        "negative_prompt": "worst quality, low quality, illustration, 2d, painting, cartoons, sketch",
+        "refine": "expert_ensemble_refiner",
+        "apply_watermark": False,
+        "num_inference_steps": 25
+    }
+    
+    output = replicate.run(
+        "stability-ai/sdxl:7762fd07cf82c948538e41f63f77d685e02b063e37e496e96eefd46c929f9bdc",
+        input=input
+    )
+    
+    # Download the image from the URL
+    image_url = output[0]
+    print(image_url)
+    response = requests.get(image_url)
+    print(response)
+    img = Image.open(io.BytesIO(response.content))
+    
+    return img  # Return the image object
+
+
+# app = Flask(__name__)
+# os.environ.get("REPLICATE_API_TOKEN")
+
+# @app.route("/")
+# def index():
+
+demo = gr.Interface(fn=image_classifier, inputs=["image", "text"], outputs="image")
+demo.launch(share=True)

flagged/log.csv

@@ -0,0 +1,9 @@
+moodboard,prompt,output,flag,username,timestamp
+"{""path"":""flagged/moodboard/14c7c3a52cd71affacbe/Screenshot 2024-06-11 at 5.27.28PM.png"",""url"":""https://71d2f0a3d14e8c955e.gradio.live/file=/private/var/folders/yq/z7scmd9d1t5_v8ycgz9trrj80000gn/T/gradio/9ba7cc91249b0ba1ea627f60ec346bfafc2851a7/Screenshot 2024-06-11 at 5.27.28PM.png"",""size"":2794341,""orig_name"":""Screenshot 2024-06-11 at 5.27.28 PM.png"",""mime_type"":""""}",A white headphone,"{""path"":""flagged/output/bddb16aad6e2eb4ce2f4/image.PNG"",""url"":null,""size"":null,""orig_name"":""image.PNG"",""mime_type"":null}",,,2024-06-11 21:29:29.682875
+"{""path"":""flagged/moodboard/20a9ebffee3a984e4e92/Screenshot 2024-06-11 at 5.27.28PM.png"",""url"":""https://71d2f0a3d14e8c955e.gradio.live/file=/private/var/folders/yq/z7scmd9d1t5_v8ycgz9trrj80000gn/T/gradio/9ba7cc91249b0ba1ea627f60ec346bfafc2851a7/Screenshot 2024-06-11 at 5.27.28PM.png"",""size"":2794341,""orig_name"":""Screenshot 2024-06-11 at 5.27.28 PM.png"",""mime_type"":""""}",A white headphone designed by dieter rams,"{""path"":""flagged/output/c2deadeee2b177d78214/image.PNG"",""url"":null,""size"":null,""orig_name"":""image.PNG"",""mime_type"":null}",,,2024-06-11 21:31:16.888982
+"{""path"":""flagged/moodboard/fde8c7b6d7cdf2100e2e/Screenshot 2024-06-11 at 5.27.28PM.png"",""url"":""https://71d2f0a3d14e8c955e.gradio.live/file=/private/var/folders/yq/z7scmd9d1t5_v8ycgz9trrj80000gn/T/gradio/9ba7cc91249b0ba1ea627f60ec346bfafc2851a7/Screenshot 2024-06-11 at 5.27.28PM.png"",""size"":2794341,""orig_name"":""Screenshot 2024-06-11 at 5.27.28 PM.png"",""mime_type"":""""}",A white headphone designed by dieter rams,"{""path"":""flagged/output/5c91673d54010e53397b/image.PNG"",""url"":null,""size"":null,""orig_name"":""image.PNG"",""mime_type"":null}",,,2024-06-11 21:31:22.368689
+"{""path"":""flagged/moodboard/98d40109371f437df727/Screenshot 2024-06-11 at 5.36.16PM.png"",""url"":""https://71d2f0a3d14e8c955e.gradio.live/file=/private/var/folders/yq/z7scmd9d1t5_v8ycgz9trrj80000gn/T/gradio/04e231aa9e4e507cd24df24d3ab7b8b87a22d315/Screenshot 2024-06-11 at 5.36.16PM.png"",""size"":1326224,""orig_name"":""Screenshot 2024-06-11 at 5.36.16 PM.png"",""mime_type"":""""}",A white sleek and geometric headphone ,"{""path"":""flagged/output/1d9e6b07d5463259c1ef/image.PNG"",""url"":null,""size"":null,""orig_name"":""image.PNG"",""mime_type"":null}",,,2024-06-11 21:45:46.737039
+"{""path"":""flagged/moodboard/06dd87dd9bbe75468871/Screenshot 2024-06-11 at 5.36.16PM.png"",""url"":""https://71d2f0a3d14e8c955e.gradio.live/file=/private/var/folders/yq/z7scmd9d1t5_v8ycgz9trrj80000gn/T/gradio/04e231aa9e4e507cd24df24d3ab7b8b87a22d315/Screenshot 2024-06-11 at 5.36.16PM.png"",""size"":1326224,""orig_name"":""Screenshot 2024-06-11 at 5.36.16 PM.png"",""mime_type"":""""}",A pair of white headphones 3d view,"{""path"":""flagged/output/cb5c23947031acf21462/image.PNG"",""url"":null,""size"":null,""orig_name"":""image.PNG"",""mime_type"":null}",,,2024-06-11 21:54:38.615366
+,A person wearing a gorilla,,,,2024-06-11 23:07:57.065591
+,A chair with no arms,,,,2024-06-11 23:08:57.767527
+,A pair of white head phone,,,,2024-06-12 06:40:24.091034

requirements.txt

@@ -0,0 +1,4 @@
+Flask==3.0.0
+replicate==0.14.0
+gradio
+openai

vercel.json

@@ -0,0 +1,14 @@
+{
+  "builds": [
+    {
+      "src": "api/index.py",
+      "use": "@vercel/python"
+    }
+  ],
+  "routes": [
+    {
+      "src": "/(.*)",
+      "dest": "api/index.py"
+    }
+  ]
+}