Update myapp.py
Browse files
myapp.py
CHANGED
|
@@ -1,63 +1,35 @@
|
|
| 1 |
from flask import Flask, request, jsonify, send_file
|
| 2 |
from flask_cors import CORS
|
| 3 |
-
import asyncio
|
| 4 |
-
import tempfile
|
| 5 |
import os
|
| 6 |
-
from threading import RLock
|
| 7 |
from huggingface_hub import InferenceClient
|
| 8 |
-
from PIL import Image # Import Pillow
|
| 9 |
from io import BytesIO # For converting image to bytes
|
|
|
|
| 10 |
|
|
|
|
| 11 |
myapp = Flask(__name__)
|
| 12 |
CORS(myapp) # Enable CORS for all routes
|
| 13 |
|
| 14 |
-
|
| 15 |
-
HF_TOKEN = os.environ.get("HF_TOKEN") # Hugging Face token
|
| 16 |
-
|
| 17 |
-
inference_timeout = 600 # Set timeout for inference
|
| 18 |
|
| 19 |
@myapp.route('/')
|
| 20 |
def home():
|
| 21 |
return "Welcome to the Image Background Remover!"
|
| 22 |
|
| 23 |
-
# Function to
|
| 24 |
-
def
|
| 25 |
-
return model_name if model_name in models else None
|
| 26 |
-
|
| 27 |
-
# Asynchronous function to perform inference
|
| 28 |
-
async def infer(client, prompt, seed=1, timeout=inference_timeout, model="prompthero/openjourney-v4"):
|
| 29 |
-
task = asyncio.create_task(
|
| 30 |
-
asyncio.to_thread(client.text_to_image, prompt=prompt, seed=seed, model=model)
|
| 31 |
-
)
|
| 32 |
-
await asyncio.sleep(0)
|
| 33 |
try:
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
result = None
|
| 41 |
-
|
| 42 |
-
if task.done() and result is not None:
|
| 43 |
-
with lock:
|
| 44 |
-
# Convert image result to bytes using Pillow
|
| 45 |
-
image_bytes = BytesIO()
|
| 46 |
-
# Assuming result is an image object from huggingface_hub
|
| 47 |
-
result.save(image_bytes, format='PNG') # Save the image to a BytesIO object
|
| 48 |
-
image_bytes.seek(0) # Go to the start of the byte stream
|
| 49 |
-
|
| 50 |
-
# Save the result image as a temporary file
|
| 51 |
-
temp_image = tempfile.NamedTemporaryFile(suffix=".png", delete=False)
|
| 52 |
-
with open(temp_image.name, "wb") as f:
|
| 53 |
-
f.write(image_bytes.read()) # Write the bytes to the temp file
|
| 54 |
-
|
| 55 |
-
return temp_image.name # Return the path to the saved image
|
| 56 |
-
return None
|
| 57 |
|
| 58 |
# Flask route for the API endpoint
|
| 59 |
@myapp.route('/generate_image', methods=['POST'])
|
| 60 |
-
def
|
| 61 |
data = request.get_json()
|
| 62 |
|
| 63 |
# Extract required fields from the request
|
|
@@ -68,23 +40,22 @@ def generate_image():
|
|
| 68 |
if not prompt:
|
| 69 |
return jsonify({"error": "Prompt is required"}), 400
|
| 70 |
|
| 71 |
-
# Get the model from all_models
|
| 72 |
-
model = get_model_from_name(model_name)
|
| 73 |
-
if not model:
|
| 74 |
-
return jsonify({"error": f"Model '{model_name}' not found in available models"}), 400
|
| 75 |
-
|
| 76 |
try:
|
| 77 |
-
#
|
| 78 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
if result_path:
|
| 83 |
-
return send_file(result_path, mimetype='image/png') # Send back the generated image file
|
| 84 |
else:
|
| 85 |
return jsonify({"error": "Failed to generate image"}), 500
|
| 86 |
except Exception as e:
|
| 87 |
-
print(f"Error in
|
| 88 |
return jsonify({"error": str(e)}), 500
|
| 89 |
|
| 90 |
# Add this block to make sure your app runs when called
|
|
|
|
| 1 |
from flask import Flask, request, jsonify, send_file
|
| 2 |
from flask_cors import CORS
|
|
|
|
|
|
|
| 3 |
import os
|
|
|
|
| 4 |
from huggingface_hub import InferenceClient
|
|
|
|
| 5 |
from io import BytesIO # For converting image to bytes
|
| 6 |
+
from PIL import Image # Import Pillow for image processing
|
| 7 |
|
| 8 |
+
# Initialize the Flask app
|
| 9 |
myapp = Flask(__name__)
|
| 10 |
CORS(myapp) # Enable CORS for all routes
|
| 11 |
|
| 12 |
+
# Initialize the InferenceClient with your Hugging Face token
|
| 13 |
+
HF_TOKEN = os.environ.get("HF_TOKEN") # Ensure to set your Hugging Face token in the environment
|
| 14 |
+
client = InferenceClient(token=HF_TOKEN)
|
|
|
|
| 15 |
|
| 16 |
@myapp.route('/')
|
| 17 |
def home():
|
| 18 |
return "Welcome to the Image Background Remover!"
|
| 19 |
|
| 20 |
+
# Function to generate an image from a prompt
|
| 21 |
+
def generate_image(prompt, seed=1, model="prompthero/openjourney-v4"):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
try:
|
| 23 |
+
# Generate the image using Hugging Face's inference API with the given model
|
| 24 |
+
result_image = client.text_to_image(prompt=prompt, seed=seed, model=model)
|
| 25 |
+
return result_image
|
| 26 |
+
except Exception as e:
|
| 27 |
+
print(f"Error generating image: {str(e)}")
|
| 28 |
+
return None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
# Flask route for the API endpoint
|
| 31 |
@myapp.route('/generate_image', methods=['POST'])
|
| 32 |
+
def generate_api():
|
| 33 |
data = request.get_json()
|
| 34 |
|
| 35 |
# Extract required fields from the request
|
|
|
|
| 40 |
if not prompt:
|
| 41 |
return jsonify({"error": "Prompt is required"}), 400
|
| 42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
try:
|
| 44 |
+
# Call the generate_image function with the custom model name
|
| 45 |
+
image = generate_image(prompt, seed, model_name)
|
| 46 |
+
|
| 47 |
+
if image:
|
| 48 |
+
# Save the image to a BytesIO object to send as response
|
| 49 |
+
image_bytes = BytesIO()
|
| 50 |
+
image.save(image_bytes, format='PNG')
|
| 51 |
+
image_bytes.seek(0) # Go to the start of the byte stream
|
| 52 |
|
| 53 |
+
# Send the generated image as a response with a download option
|
| 54 |
+
return send_file(image_bytes, mimetype='image/png', as_attachment=True, download_name='generated_image.png')
|
|
|
|
|
|
|
| 55 |
else:
|
| 56 |
return jsonify({"error": "Failed to generate image"}), 500
|
| 57 |
except Exception as e:
|
| 58 |
+
print(f"Error in generate_api: {str(e)}") # Log the error
|
| 59 |
return jsonify({"error": str(e)}), 500
|
| 60 |
|
| 61 |
# Add this block to make sure your app runs when called
|