Upload folder using huggingface_hub

README.md

@@ -0,0 +1,25 @@
+---
+license: creativeml-openrail-m
+base_model: CompVis/stable-diffusion-v1-4
+training_prompt: A bottle doing a series of translations
+tags:
+- stable-diffusion
+- stable-diffusion-diffusers
+- text-to-image
+- diffusers
+- text-to-video
+- tune-a-video
+inference: false
+---
+
+# Tune-A-Video - product-ads
+
+## Model description
+- Base model: [CompVis/stable-diffusion-v1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4)
+- Training prompt: A bottle doing a series of translations
+
+
+
+## Related papers:
+- [Tune-A-Video](https://arxiv.org/abs/2212.11565): One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation
+- [Stable-Diffusion](https://arxiv.org/abs/2112.10752): High-Resolution Image Synthesis with Latent Diffusion Models

config.yaml

@@ -0,0 +1,43 @@
+pretrained_model_path: checkpoints/CompVis/stable-diffusion-v1-4
+output_dir: /home/user/app/experiments/product-ads
+train_data:
+  video_path: /tmp/gradio/e14d86daa040bee397ec527f496bcf79f217d00e/Tom Ford - Fougere
+    DArgent Spec Ad 1.mp3
+  prompt: A bottle doing a series of translations
+  n_sample_frames: 8
+  width: 512
+  height: 512
+  sample_start_idx: 0
+  sample_frame_rate: 1
+validation_data:
+  prompts:
+  - A bottle doing a series of translations
+  video_length: 8
+  width: 512
+  height: 512
+  num_inference_steps: 50
+  guidance_scale: 7.5
+validation_steps: 100
+trainable_modules:
+- attn1.to_q
+- attn2.to_q
+- attn_temp
+train_batch_size: 1
+max_train_steps: 300
+learning_rate: 3.5e-05
+scale_lr: false
+lr_scheduler: constant
+lr_warmup_steps: 0
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_weight_decay: 0.01
+adam_epsilon: 1.0e-08
+max_grad_norm: 1.0
+gradient_accumulation_steps: 1
+gradient_checkpointing: true
+checkpointing_steps: 1000
+resume_from_checkpoint: null
+mixed_precision: fp16
+use_8bit_adam: false
+enable_xformers_memory_efficient_attention: true
+seed: 87217