Update README.md

README.md

@@ -10,12 +10,35 @@ tags:
 - lora
 inference: true
 ---
+
+use the following code to generate image
+```python
+import os
+from diffusers import StableDiffusionPipeline
+import torch
+import random
+
+prompt="可爱的,黄色的老鼠,小精灵,卡通"
+seed=random.randint(1,2147483647)
+print(seed)
+pipe = StableDiffusionPipeline.from_pretrained("IDEA-CCNL/Taiyi-Stable-Diffusion-1B-Anime-Chinese-v0.1",
+                                               torch_dtype=torch.float16,)
+model_path = "miluELK/Taiyi-sd-pokemon-LoRA-zh-512-v2"
+
+pipe.unet.load_attn_procs(model_path)
+pipe.to("cuda")
+generator = torch.Generator("cuda").manual_seed(seed)
+pipe.safety_checker = lambda images, clip_input: (images, False)
+image = pipe(prompt, generator=generator, num_inference_steps=50, guidance_scale=7.5).images[0]
+image.save("./" + prompt + ".png")
+image
+```
     
 # LoRA text2image fine-tuning - miluELK/Taiyi-sd-pokemon-LoRA-zh-512-v2
 These are LoRA adaption weights for IDEA-CCNL/Taiyi-Stable-Diffusion-1B-Anime-Chinese-v0.1. The weights were fine-tuned on the svjack/pokemon-blip-captions-en-zh dataset. You can find some example images in the following. 
 
-![img_0](./image_0.png)
 ![img_1](./image_1.png)
 ![img_2](./image_2.png)
-![img_3](./image_3.png)
+![img_3](./image3.png)
+![img_4](./image4.png)