Upload folder using huggingface_hub

README.md

@@ -0,0 +1,52 @@
+---
+tags:
+- text-to-image
+- flux
+- lora
+- diffusers
+- template:sd-lora
+- fluxgym
+widget:
+-   output:
+        url: sample/bigass_002600_00_20241122081233.png
+    text: B1g455 The image is a high-resolution photograph featuring a woman taking
+        a mirror selfie in what appears to be a modern, minimalist bedroom. The subject
+        is a light-skinned woman with a curvy physique, including a prominent, rounded
+        buttocks and a flat stomach. She has long, dark hair pulled back in a high
+        ponytail. She is topless, wearing only a very minimal thong bikini that accentuates
+        her hips and buttocks. Her breasts are not visible in this pose. The woman
+        is holding a smartphone with a pink case, capturing the reflection in the
+        mirror.She has a tattoo on her left hip. Her makeup is subtle, with a focus
+        on her eyes and lips.
+-   output:
+        url: sample/bigass_002600_01_20241122081516.png
+    text: B1g455 The image is a high-resolution photograph featuring a young woman
+        posed provocatively on all fours on a green yoga mat in a minimalist room
+        with beige walls. She is positioned with her upper body leaning forward, accentuating
+        her curvy figure, including her prominent buttocks and thighs. She has long,
+        wavy dark hair with bangs and is wearing a black, long-sleeved, cropped turtleneck
+        sweater that covers her arms but leaves her lower body completely exposed.
+        Her skin tone is light, and she has a tattoo on her left hip featuring a simple,
+        geometric design. Her makeup is bold, with dark eyeliner, mascara, and nude
+        lipstick, highlighting her full, pouty lips.
+base_model: black-forest-labs/FLUX.1-dev
+instance_prompt: B1g455
+license: other
+license_name: flux-1-dev-non-commercial-license
+license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
+---
+
+# BigAss
+
+A Flux LoRA trained on a local computer with [Fluxgym](https://github.com/cocktailpeanut/fluxgym)
+
+<Gallery />
+
+## Trigger words
+
+You should use `B1g455` to trigger the image generation.
+
+## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, Forge, etc.
+
+Weights for this model are available in Safetensors format.
+

bigass-000001.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5df856f36e844d7f81fe3fca18a9275dbd3e38d78773e32b4a9421f99b55b4b
+size 18250260

bigass-000002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f30479e77c8bb00866b9fcc8fc7370ec1cc74515d4c7736ba41c93f8b9868ee
+size 18250260

bigass-000003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14bb36e9c710e4dc4d2a116c58cadb9a1e333377684784effabc91fc1cf722a0
+size 18250260

bigass-000004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70b08b79a4b40f6ec3cae93bf739e2e996dc908b909e5f16d9ab14e79cb6b000
+size 18250260

bigass-000005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f55336438050de414cded7e8d6642bc0432eca27a377ebc28a147a2dfa89cdc
+size 18250260

bigass-000006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca4c5a8f95055a4a8690701af9ac077734fc4d67a1ebe9ba23c3585a900b1f40
+size 18250260

bigass-000007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80b67bf88664cd971f8ffd9687057588afca40b70e15a907c1663daff1381738
+size 18250260

bigass-000008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fec73f64f0d416ee995a2cb97639d3a0a5a91f6245a3f98bebc8a48e3df231a8
+size 18250260

bigass-000009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f17c36fcb73d12b54912bac2c5e5afd2e18677521c7707557ca461b2e687299
+size 18250260

bigass-000010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bee62ed75e55ea596171d7f9fd3edf2b888db2b02f46d1e74ffbd84bd3a37200
+size 18250260

bigass-000011.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70b56993edc7298ebbde54358ff0a9a9fc1d0c1284ff381129a779223563b51d
+size 18250260

bigass-000012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:721f676f8fd4f9f242afde1b71566f0edd49f3d04a89dcb4b1238b240affa549
+size 18250260

bigass-000013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b82597e9c259b2d164b7e7cd3200e72ea881476bd7a70e9884ff4015c5bff73
+size 18250260

bigass-000014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d3dcb2fc665ec6999914667455f37f1867db2c42eb32d6b518973d29a0fdace
+size 18250260

bigass-000015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74cd4476aeaebbdf716c5e4c57b84122d901a6e637eb91da12ef0ead9345db15
+size 18250260

bigass-000016.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2ef2edd05dba77777f11df842fee0f0962d67d7c07e9ab8046ea56a9f2ba011
+size 18250260

bigass-000017.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47d3590fbd375bb5c74886547553309065dc9de7325eec4d1936594e2c3f3542
+size 18250260

bigass-000018.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:357346722ecd46201d8a2c37e027610c08bd15a7e18d66e7a22cf0610ffce6d3
+size 18250260

bigass-000019.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3bfde43acd8df4011749bac9fa09706e8e977d93fc81ce403a8576864e855cb
+size 18250260

bigass.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a61ecd9eebc3a26e9882029deec2e53e2f5e355d827223b5d76ee5a6cd12f3c3
+size 18250260

dataset.toml

@@ -0,0 +1,14 @@
+[general]
+shuffle_caption = false
+caption_extension = '.txt'
+keep_tokens = 1
+
+[[datasets]]
+resolution = 512
+batch_size = 1
+keep_tokens = 1
+
+  [[datasets.subsets]]
+  image_dir = 'C:\pinokio\api\fluxgym.git\datasets\bigass'
+  class_tokens = 'B1g455'
+  num_repeats = 4

sample_prompts.txt

@@ -0,0 +1,3 @@
+B1g455 The image is a high-resolution photograph featuring a woman taking a mirror selfie in what appears to be a modern, minimalist bedroom. The subject is a light-skinned woman with a curvy physique, including a prominent, rounded buttocks and a flat stomach. She has long, dark hair pulled back in a high ponytail. She is topless, wearing only a very minimal thong bikini that accentuates her hips and buttocks. Her breasts are not visible in this pose. The woman is holding a smartphone with a pink case, capturing the reflection in the mirror.She has a tattoo on her left hip. Her makeup is subtle, with a focus on her eyes and lips.
+
+B1g455 The image is a high-resolution photograph featuring a young woman posed provocatively on all fours on a green yoga mat in a minimalist room with beige walls. She is positioned with her upper body leaning forward, accentuating her curvy figure, including her prominent buttocks and thighs. She has long, wavy dark hair with bangs and is wearing a black, long-sleeved, cropped turtleneck sweater that covers her arms but leaves her lower body completely exposed. Her skin tone is light, and she has a tattoo on her left hip featuring a simple, geometric design. Her makeup is bold, with dark eyeliner, mascara, and nude lipstick, highlighting her full, pouty lips.

train.bat

@@ -0,0 +1,39 @@
+accelerate launch ^
+  --mixed_precision bf16 ^
+  --num_cpu_threads_per_process 1 ^
+  sd-scripts/flux_train_network.py ^
+  --pretrained_model_name_or_path "C:\pinokio\api\fluxgym.git\models\unet\bdsqlsz\flux1-dev2pro-single\flux1-dev2pro.safetensors" ^
+  --clip_l "C:\pinokio\api\fluxgym.git\models\clip\clip_l.safetensors" ^
+  --t5xxl "C:\pinokio\api\fluxgym.git\models\clip\t5xxl_fp16.safetensors" ^
+  --ae "C:\pinokio\api\fluxgym.git\models\vae\ae.sft" ^
+  --cache_latents_to_disk ^
+  --save_model_as safetensors ^
+  --sdpa --persistent_data_loader_workers ^
+  --max_data_loader_n_workers 2 ^
+  --seed 42 ^
+  --gradient_checkpointing ^
+  --mixed_precision bf16 ^
+  --save_precision bf16 ^
+  --network_module networks.lora_flux ^
+  --network_dim 4 ^
+  --optimizer_type adafactor ^
+  --optimizer_args "relative_step=False" "scale_parameter=False" "warmup_init=False" ^
+  --split_mode ^
+  --network_args "train_blocks=single" ^
+  --lr_scheduler constant_with_warmup ^
+  --max_grad_norm 0.0 ^--sample_prompts="C:\pinokio\api\fluxgym.git\outputs\bigass\sample_prompts.txt" --sample_every_n_steps="200" ^
+  --learning_rate 8e-4 ^
+  --cache_text_encoder_outputs ^
+  --cache_text_encoder_outputs_to_disk ^
+  --fp8_base ^
+  --highvram ^
+  --max_train_epochs 20 ^
+  --save_every_n_epochs 1 ^
+  --dataset_config "C:\pinokio\api\fluxgym.git\outputs\bigass\dataset.toml" ^
+  --output_dir "C:\pinokio\api\fluxgym.git\outputs\bigass" ^
+  --output_name bigass ^
+  --timestep_sampling shift ^
+  --discrete_flow_shift 3.1582 ^
+  --model_prediction_type raw ^
+  --guidance_scale 1 ^
+  --loss_type l2 ^