File size: 1,888 Bytes
c5bdd8b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
from fastapi import FastAPI
import fal_client
import gradio as gr
import threading
import os
from pathlib import Path
from public_instagram_loader import public_instagram_loader, InstagramData

app = FastAPI()

# Set up the FAL API Key
FAL_KEY = os.getenv("FAL_KEY")
#fal_client.set_api_key(FAL_KEY)

# FastAPI route
@app.get("/")
def greet_json():
    return {"AI-InstagramPhotos": "Generate photos from your favorite Instagram account using FAL API."}

# Function to submit Instagram data to FAL API
def submit_to_fal(instagram_data: InstagramData):
    handler = fal_client.submit(
        "fal-ai/flux-lora-fast-training",
        arguments={
            "images_data_url": instagram_data.image_urls,
            "trigger_word": instagram_data.profile_info['username'],
            "is_style": True
        },
    )
    result = handler.get()
    return result

# Gradio interface for interacting with the Instagram loader and FAL
def gradio_interface():
    def generate_lora_from_instagram(username: str):
        # Load Instagram data
        instagram_data = public_instagram_loader(username)
        if instagram_data is None:
            return "Profile not found or an error occurred."
        # Pass Instagram data to FAL API for LoRA training
        fal_result = submit_to_fal(instagram_data)
        # Return the URL of the trained LoRA weights
        return f"Trained LoRA weights: {fal_result['diffusers_lora_file']['url']}"

    iface = gr.Interface(
        fn=generate_lora_from_instagram, 
        inputs=gr.Textbox(label="Enter Instagram username"),
        outputs=gr.Textbox(label="LoRA URL"), 
        title="Instagram LoRA Generator using FAL"
    )
    
    # Launch Gradio on a different port (7861)
    iface.launch(server_name="0.0.0.0", server_port=7861)

# Run Gradio in a separate thread
thread = threading.Thread(target=gradio_interface)
thread.start()