USER
Implement the API in my python script:
Choose a language to see the code snippets for interacting with the API.
1. Install the python client (docs) if you don't already have it installed.
copy
$ pip install gradio_client
2. Find the API endpoint below corresponding to your desired function in the app. Copy the code snippet, replacing the placeholder values with your own input data. If this is a private Space, you may need to pass your Hugging Face token as well (read more). Or use the
API Recorder
to automatically generate your API requests.
api_name: /flip_text
copy
from gradio_client import Client
client = Client("prodia/sdxl-stable-diffusion-xl")
result = client.predict(
prompt="space warrior, beautiful, female, ultrarealistic, soft lighting, 8k",
negative_prompt="3d, cartoon, anime, (deformed eyes, nose, ears, nose), bad anatomy, ugly",
model="sd_xl_base_1.0.safetensors [be9edd61]",
steps=20,
sampler="DPM++ 2M Karras",
cfg_scale=7,
resolution="1024x1024",
seed=-1,
api_name="/flip_text"
)
print(result)
Accepts 8 parameters:
prompt str Default: "space warrior, beautiful, female, ultrarealistic, soft lighting, 8k"
The input value that is provided in the "parameter_10" Textbox component.
negative_prompt str Default: "3d, cartoon, anime, (deformed eyes, nose, ears, nose), bad anatomy, ugly"
The input value that is provided in the "parameter_11" Textbox component.
model Literal['animagineXLV3_v30.safetensors [75f2f05b]', 'devlishphotorealism_sdxl15.safetensors [77cba69f]', 'dreamshaperXL10_alpha2.safetensors [c8afe2ef]', 'dynavisionXL_0411.safetensors [c39cc051]', 'juggernautXL_v45.safetensors [e75f5471]', 'realismEngineSDXL_v10.safetensors [af771c3f]', 'realvisxlV40.safetensors [f7fdcb51]', 'sd_xl_base_1.0.safetensors [be9edd61]', 'sd_xl_base_1.0_inpainting_0.1.safetensors [5679a81a]', 'turbovisionXL_v431.safetensors [78890989]'] Default: "sd_xl_base_1.0.safetensors [be9edd61]"
The input value that is provided in the "Stable Diffusion Checkpoint" Dropdown component.
steps float Default: 20
The input value that is provided in the "Sampling Steps" Slider component.
sampler Literal['DPM++ 2M Karras', 'DPM++ SDE Karras', 'DPM++ 2M SDE Exponential', 'DPM++ 2M SDE Karras', 'Euler a', 'Euler', 'LMS', 'Heun', 'DPM2', 'DPM2 a', 'DPM++ 2S a', 'DPM++ 2M', 'DPM++ SDE', 'DPM++ 2M SDE', 'DPM++ 2M SDE Heun', 'DPM++ 2M SDE Heun Karras', 'DPM++ 2M SDE Heun Exponential', 'DPM++ 3M SDE', 'DPM++ 3M SDE Karras', 'DPM++ 3M SDE Exponential', 'DPM fast', 'DPM adaptive', 'LMS Karras', 'DPM2 Karras', 'DPM2 a Karras', 'DPM++ 2S a Karras', 'Restart', 'DDIM', 'PLMS', 'UniPC'] Default: "DPM++ 2M Karras"
The input value that is provided in the "Sampling Method" Dropdown component.
cfg_scale float Default: 7
The input value that is provided in the "CFG Scale" Slider component.
resolution Literal['1024x1024', '1152x896', '1216x832', '1344x768', '1536x640', '640x1536', '768x1344', '832x1216'] Default: "1024x1024"
The input value that is provided in the "Resolution" Dropdown component.
seed float Default: -1
The input value that is provided in the "Seed" Number component.
Returns 1 element
filepath
The output value that appears in the "value_38" Image component.
Use the CSS for smooth design (libraries)ASSISTANT
Certainly! Below is a comprehensive Python script that demonstrates how to interact with the `/flip_text` API endpoint using the `gradio_client` library. This script includes handling for all required parameters, optional support for private Spaces via a Hugging Face token, and clear instructions for setup and usage.
## Table of Contents
1. [Prerequisites](#prerequisites)
2. [Installation](#installation)
3. [Configuration](#configuration)
4. [Python Script](#python-script)
5. [Usage](#usage)
6. [Notes on CSS Styling](#notes-on-css-styling)
7. [Complete Script](#complete-script)
---
### Prerequisites
- **Python 3.7 or higher**: Ensure that Python is installed on your system. You can download it from [python.org](https://www.python.org/downloads/).
- **Hugging Face Account (if accessing a private Space)**: If the Space you're accessing is private, you'll need a Hugging Face token with the appropriate permissions.
### Installation
1. **Create a Virtual Environment (Optional but Recommended)**
It's good practice to use a virtual environment to manage dependencies.
```bash
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
```
2. **Install Required Libraries**
Install `gradio_client` using `pip`:
```bash
pip install gradio_client
```
### Configuration
If you're accessing a **private Space** on Hugging Face, you'll need to provide your Hugging Face token. Here's how to set it up:
1. **Obtain Your Hugging Face Token**
- Log in to your [Hugging Face account](https://huggingface.co/).
- Navigate to [Access Tokens](https://huggingface.co/settings/tokens).
- Create a new token with the necessary permissions.
2. **Set the Token as an Environment Variable**
This approach enhances security by not hardcoding the token in your script.
- **Unix/Linux/MacOS:**
```bash
export HUGGINGFACE_TOKEN='your_token_here'
```
- **Windows (Command Prompt):**
```cmd
set HUGGINGFACE_TOKEN=your_token_here
```
- **Windows (PowerShell):**
```powershell
$env:HUGGINGFACE_TOKEN="your_token_here"
```
Alternatively, you can directly assign the token in the script (not recommended for sensitive environments).
### Python Script
Below is the Python script that interacts with the `/flip_text` API endpoint:
```python
import os
from gradio_client import Client
def flip_text_api(
prompt: str = "space warrior, beautiful, female, ultrarealistic, soft lighting, 8k",
negative_prompt: str = "3d, cartoon, anime, (deformed eyes, nose, ears, nose), bad anatomy, ugly",
model: str = "sd_xl_base_1.0.safetensors [be9edd61]",
steps: float = 20,
sampler: str = "DPM++ 2M Karras",
cfg_scale: float = 7,
resolution: str = "1024x1024",
seed: float = -1,
huggingface_token: str = None, # Optional: Only required for private Spaces
client_name: str = "prodia/sdxl-stable-diffusion-xl",
api_endpoint: str = "/flip_text"
) -> str:
"""
Interacts with the /flip_text API endpoint using gradio_client.
Parameters:
prompt (str): The main prompt for the model.
negative_prompt (str): The negative prompt to avoid certain outputs.
model (str): The model checkpoint to use.
steps (float): Number of sampling steps.
sampler (str): The sampling method.
cfg_scale (float): CFG scale value.
resolution (str): Output resolution.
seed (float): Seed for random number generator.
huggingface_token (str, optional): Hugging Face API token for private Spaces.
client_name (str): The name of the Gradio client/Space.
api_endpoint (str): The API endpoint to interact with.
Returns:
str: File path to the generated image.
"""
# Initialize the client
try:
if huggingface_token:
client = Client(client_name, token=huggingface_token)
else:
client = Client(client_name)
except Exception as e:
print(f"Error initializing Gradio Client: {e}")
return None
# Prepare the parameters
params = {
"prompt": prompt,
"negative_prompt": negative_prompt,
"model": model,
"steps": steps,
"sampler": sampler,
"cfg_scale": cfg_scale,
"resolution": resolution,
"seed": seed
}
try:
# Invoke the API
result = client.predict(
*params.values(),
api_name=api_endpoint
)
return result # This should be the filepath to the image
except Exception as e:
print(f"Error during API prediction: {e}")
return None
if __name__ == "__main__":
# Retrieve Hugging Face token from environment variable (if set)
hf_token = os.getenv('HUGGINGFACE_TOKEN') # Ensure you've set this if needed
# Example usage
image_path = flip_text_api(
prompt="A serene landscape with mountains and a river at sunset.",
negative_prompt="low resolution, blurry, distorted",
model="sd_xl_base_1.0.safetensors [be9edd61]",
steps=25,
sampler="Euler a",
cfg_scale=8.5,
resolution="1024x1024",
seed=42,
huggingface_token=hf_token # Pass the token if accessing a private Space
)
if image_path:
print(f"Generated image saved to: {image_path}")
else:
print("Failed to generate image.")
```
### Explanation of the Script
1. **Import Statements**
- `os`: To access environment variables.
- `gradio_client.Client`: To interact with the Gradio API.
2. **`flip_text_api` Function**
- **Parameters**: Accepts all eight parameters required by the API, with default values as specified.
- **Hugging Face Token**: Optional parameter to handle private Spaces.
- **Client Initialization**: Initializes the Gradio client with or without the token based on availability.
- **API Invocation**: Calls the `predict` method with the provided parameters and the specified API endpoint.
- **Error Handling**: Catches and prints errors during client initialization and API prediction.
3. **`__main__` Block**
- Retrieves the Hugging Face token from the environment variable `HUGGINGFACE_TOKEN`.
- Demonstrates an example usage of the `flip_text_api` function.
- Prints the path to the generated image or an error message.
### Usage
1. **Ensure Dependencies are Installed**
Make sure you've installed `gradio_client` as shown in the [Installation](#installation) section.
2. **Set Up Environment Variables (If Necessary)**
If you're accessing a private Space, set the `HUGGINGFACE_TOKEN` environment variable with your token.
3. **Run the Script**
Save the script to a file, e.g., `flip_text_api.py`, and run it using:
```bash
python flip_text_api.py
```
Upon successful execution, it will print the file path to the generated image.
### Notes on CSS Styling
CSS (Cascading Style Sheets) is primarily used for styling web pages and has no direct application in pure Python scripts. If you're building a web interface to interact with this Python script (e.g., using Flask, Django, or a frontend framework), you can apply CSS to enhance the design and user experience of that web interface.
If you intended to create a web-based application that uses this API, consider integrating this script with a web framework and applying CSS through that framework's standard practices.
### Complete Script
For convenience, here's the complete Python script again:
```python
import os
from gradio_client import Client
def flip_text_api(
prompt: str = "space warrior, beautiful, female, ultrarealistic, soft lighting, 8k",
negative_prompt: str = "3d, cartoon, anime, (deformed eyes, nose, ears, nose), bad anatomy, ugly",
model: str = "sd_xl_base_1.0.safetensors [be9edd61]",
steps: float = 20,
sampler: str = "DPM++ 2M Karras",
cfg_scale: float = 7,
resolution: str = "1024x1024",
seed: float = -1,
huggingface_token: str = None, # Optional: Only required for private Spaces
client_name: str = "prodia/sdxl-stable-diffusion-xl",
api_endpoint: str = "/flip_text"
) -> str:
"""
Interacts with the /flip_text API endpoint using gradio_client.
Parameters:
prompt (str): The main prompt for the model.
negative_prompt (str): The negative prompt to avoid certain outputs.
model (str): The model checkpoint to use.
steps (float): Number of sampling steps.
sampler (str): The sampling method.
cfg_scale (float): CFG scale value.
resolution (str): Output resolution.
seed (float): Seed for random number generator.
huggingface_token (str, optional): Hugging Face API token for private Spaces.
client_name (str): The name of the Gradio client/Space.
api_endpoint (str): The API endpoint to interact with.
Returns:
str: File path to the generated image.
"""
# Initialize the client
try:
if huggingface_token:
client = Client(client_name, token=huggingface_token)
else:
client = Client(client_name)
except Exception as e:
print(f"Error initializing Gradio Client: {e}")
return None
# Prepare the parameters
params = {
"prompt": prompt,
"negative_prompt": negative_prompt,
"model": model,
"steps": steps,
"sampler": sampler,
"cfg_scale": cfg_scale,
"resolution": resolution,
"seed": seed
}
try:
# Invoke the API
result = client.predict(
*params.values(),
api_name=api_endpoint
)
return result # This should be the filepath to the image
except Exception as e:
print(f"Error during API prediction: {e}")
return None
if __name__ == "__main__":
# Retrieve Hugging Face token from environment variable (if set)
hf_token = os.getenv('HUGGINGFACE_TOKEN') # Ensure you've set this if needed
# Example usage
image_path = flip_text_api(
prompt="A serene landscape with mountains and a river at sunset.",
negative_prompt="low resolution, blurry, distorted",
model="sd_xl_base_1.0.safetensors [be9edd61]",
steps=25,
sampler="Euler a",
cfg_scale=8.5,
resolution="1024x1024",
seed=42,
huggingface_token=hf_token # Pass the token if accessing a private Space
)
if image_path:
print(f"Generated image saved to: {image_path}")
else:
print("Failed to generate image.")
```
---
Feel free to customize the parameters as needed. If you encounter any issues or have further questions, don't hesitate to ask!