Hugging Face's logo

Spaces:
MohamedRashad
/
Game-Items-Generator
Running on Zero

App Files Community
Game-Items-Generator / app.py

Commit History

Update image generation pipeline by adding Autoencoder models and random seed for variability
90c99b7
Running

MohamedRashad commited on

Reduce inference steps to 1 in generate_item_image function
2c55a9a

MohamedRashad commited on

Remove commented-out code and add print statement in generate_item_image function
196afc8

MohamedRashad commited on

Update model pipeline and adjust inference steps in generate_item_image function
a5480b1

MohamedRashad commited on

Update generate_item_image function to reduce inference steps and image dimensions
cce8702

MohamedRashad commited on

Reduce inference steps in generate_item_image function and remove print statement
5b9fc85

MohamedRashad commited on

Add print statement to indicate image generation in generate_item_image function
d65d459

MohamedRashad commited on

Remove unused imports and clean up generate_item_image function in app.py
474e44a

MohamedRashad commited on

Update image generation pipeline in app.py to use new model and enable sequential CPU offload
9200fea

MohamedRashad commited on

Remove sequential CPU offload from image generation pipeline in app.py
699c5a8

MohamedRashad commited on

Enhance image generation pipeline by enabling tiling, slicing, and sequential CPU offload in app.py
36a40e7

MohamedRashad commited on

Update app.py to refine image generation and adjust demo sharing settings; modify requirements for xformers and restore utils3d dependency
3712078

MohamedRashad commited on

Add check for CUDA availability in app.py
0467139

MohamedRashad commited on

Add nvidia-smi command to app.py for GPU monitoring
b1b4917

MohamedRashad commited on

Set environment variables for CUDA and Hugging Face paths; increase image prompt height and enable sharing for demo launch
f4f0e5b

MohamedRashad commited on

Refactor app.py to optimize model loading and memory management; added AutoencoderTiny and adjusted GPU settings.
8961b93

MohamedRashad commited on

Install flash-attn without build isolation in app.py and remove it from requirements.txt
449f481

MohamedRashad commited on

Update SDK version in README and revert model output to LitModel3D in app.py
c63aa42

MohamedRashad commited on

Replace LitModel3D with gr.Model3D for 3D asset generation in app.py
2175b47

MohamedRashad commited on

Enhance image generation by moving FluxPipeline and AutoencoderKL to GPU; update requirements to include gradio_litmodel3d
b1e0411

MohamedRashad commited on

Move FluxPipeline to GPU and back to CPU during image generation; remove unused AutoencoderKL reference
9459cdf

MohamedRashad commited on

Refactor FluxPipeline and AutoencoderKL initialization to remove device assignment
5ee125e

MohamedRashad commited on

Integrate FluxPipeline and AutoencoderKL for enhanced image generation; add live preview helper functions and update requirements
1273148

MohamedRashad commited on

Implement temporary directory cleanup every 10 minutes and remove unused kaolin imports
61757c6

MohamedRashad commited on

Update t2i_client initialization to use the merged model and reduce inference steps
191990f

MohamedRashad commited on

Remove HF_TOKEN from t2i_client initialization and update model reference
06fd2b2

MohamedRashad commited on

Update t2i_client initialization to include HF_TOKEN from environment variables
c07df32

MohamedRashad commited on

Replace LitModel3D with gr.Model3D and restore functionality for GLB extraction
b54b529

MohamedRashad commited on

Comment out unused LitModel3D import and related functionality in app.py
cb71ec8

MohamedRashad commited on

Increase num_inference_steps in generate_item_image function to improve image quality
28f466d

MohamedRashad commited on

Update SDK version, remove FluxPipeline, and clean up requirements
689e677

MohamedRashad commited on

Update image generation to use InferenceClient and adjust requirements
87af913

MohamedRashad commited on

Replace FluxPipeline with InferenceClient for image generation and update related code
a5e543c

MohamedRashad commited on

Refactor GPU decorator usage in generate_item_image function
5eedf0a

MohamedRashad commited on

Update Flux pipeline initialization to load on CPU and reflect device in log
5ba7022

MohamedRashad commited on

Add logging for Flux pipeline device initialization in app.py
9cb3d0f

MohamedRashad commited on

Refactor image generation in app.py to streamline processing and enhance performance
38f03cc

MohamedRashad commited on

Remove duplicate import of spaces in app.py
df48381

MohamedRashad commited on

Remove redundant torch import and streamline device initialization in app.py
09d1730

MohamedRashad commited on

Refactor app.py to initialize FLUX pipeline and VAE on CUDA, enhancing image generation capabilities
e8fcedb

MohamedRashad commited on

Move device initialization for FLUX pipeline and VAE to CPU, add CUDA support in generate_item_image function
5d7365b

MohamedRashad commited on

Add trial_id variable initialization in generate_item_image function
d997e94

MohamedRashad commited on

Implement FLUX pipeline integration and add helper functions for image generation
cf723be

MohamedRashad commited on

Remove commented-out code for generating T2I prompt in app.py
985163d

MohamedRashad commited on

Integrate Gradio client for LLM predictions and update requirements
c767532

MohamedRashad commited on

Add initial module structure and base classes for samplers and representations
73c350d

MohamedRashad commited on