---
license: creativeml-openrail-m
base_model: kandinsky-community/kandinsky-2-2-prior
datasets:
- lambdalabs/pokemon-blip-captions
tags:
- kandinsky
- text-to-image
- diffusers
inference: true
---
    
# Finetuning - YiYiXu/kandinsky_prior_pokemon

This pipeline was finetuned from **kandinsky-community/kandinsky-2-2-prior** on the **lambdalabs/pokemon-blip-captions** dataset. Below are some example images generated with the finetuned pipeline using the following prompts: ['A robot pokemon, 4k photo']: 

![val_imgs_grid](./val_imgs_grid.png)


## Pipeline usage

You can use the pipeline like so:

```python
from diffusers import DiffusionPipeline
import torch

pipe_prior = DiffusionPipeline.from_pretrained("YiYiXu/kandinsky_prior_pokemon", torch_dtype=torch.float16)
pipe_t2i = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16)
prompt = "A robot pokemon, 4k photo"
image_embeds, negative_image_embeds = pipe_prior(prompt, guidance_scale=1.0).to_tuple()
image = pipe_t2i(image_embeds=image_embeds, negative_image_embeds=negative_image_embeds).images[0]
image.save("my_image.png")
```

## Training info

These are the key hyperparameters used during training:

* Epochs: 2
* Learning rate: 1e-05
* Batch size: 1
* Gradient accumulation steps: 1
* Image resolution: 768
* Mixed-precision: fp16


More information on all the CLI arguments and the environment are available on your [`wandb` run page](https://wandb.ai/yiyixu/text2image-fine-tune/runs/rmf7pvcm).