Upload folder using huggingface_hub

README.md

@@ -23,15 +23,37 @@ instance_prompt: TOK
 
 <Gallery />
 
-Trained on Replicate using:
+## About this LoRA
 
-https://replicate.com/ostris/flux-dev-lora-trainer/train
+This is a [LoRA](https://replicate.com/docs/guides/working-with-loras) for the FLUX.1-dev text-to-image model. It can be used with diffusers or ComfyUI.
+
+It was trained on [Replicate](https://replicate.com/) using AI toolkit: https://replicate.com/ostris/flux-dev-lora-trainer/train
 
 
 ## Trigger words
+
 You should use `TOK` to trigger the image generation.
 
 
+## Run this LoRA with an API using Replicate
+
+```py
+import replicate
+
+input = {
+    "prompt": "TOK",
+    "lora_weights": "https://huggingface.co/gride29/flux-custom-smaller/resolve/main/lora.safetensors"
+}
+
+output = replicate.run(
+    "black-forest-labs/flux-dev-lora",
+    input=input
+)
+for index, item in enumerate(output):
+    with open(f"output_{index}.webp", "wb") as file:
+        file.write(item.read())
+```
+
 ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
 
 ```py
@@ -40,7 +62,19 @@ import torch
 
 pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
 pipeline.load_lora_weights('gride29/flux-custom-smaller', weight_name='lora.safetensors')
-image = pipeline('your prompt').images[0]
+image = pipeline('TOK').images[0]
 ```
 
 For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
+
+
+## Training details
+
+- Steps: 1000
+- Learning rate: 0.0004
+- LoRA rank: 16
+
+
+## Contribute your own examples
+
+You can use the [community tab](https://huggingface.co/gride29/flux-custom-smaller/discussions) to add images that show off what you’ve made with this LoRA.

config.yaml

@@ -12,7 +12,7 @@ config:
       linear_alpha: 16
     save:
       dtype: float16
-      save_every: 51
+      save_every: 1001
       max_step_saves_to_keep: 1
     datasets:
     - folder_path: input_images
@@ -27,7 +27,7 @@ config:
       - 1024
     train:
       batch_size: 1
-      steps: 50
+      steps: 1000
       gradient_accumulation_steps: 1
       train_unet: true
       train_text_encoder: false
@@ -46,7 +46,7 @@ config:
       quantize: false
     sample:
       sampler: flowmatch
-      sample_every: 51
+      sample_every: 1001
       width: 1024
       height: 1024
       prompts: []

lora.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e8a4fe59d015a03e0c30a6c2a341ebeceb9f557959260557f7275f5756ef751
-size 171969408
+oid sha256:9ea2f682e622b2887fd3be71036045b37648efbc6370d62e213568e5a0739e37
+size 171969416