Upload folder using huggingface_hub
Browse files- README.md +25 -0
- config.yaml +43 -0
README.md
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: creativeml-openrail-m
|
3 |
+
base_model: CompVis/stable-diffusion-v1-4
|
4 |
+
training_prompt: A bottle doing a series of translations
|
5 |
+
tags:
|
6 |
+
- stable-diffusion
|
7 |
+
- stable-diffusion-diffusers
|
8 |
+
- text-to-image
|
9 |
+
- diffusers
|
10 |
+
- text-to-video
|
11 |
+
- tune-a-video
|
12 |
+
inference: false
|
13 |
+
---
|
14 |
+
|
15 |
+
# Tune-A-Video - product-ads
|
16 |
+
|
17 |
+
## Model description
|
18 |
+
- Base model: [CompVis/stable-diffusion-v1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4)
|
19 |
+
- Training prompt: A bottle doing a series of translations
|
20 |
+
|
21 |
+
|
22 |
+
|
23 |
+
## Related papers:
|
24 |
+
- [Tune-A-Video](https://arxiv.org/abs/2212.11565): One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation
|
25 |
+
- [Stable-Diffusion](https://arxiv.org/abs/2112.10752): High-Resolution Image Synthesis with Latent Diffusion Models
|
config.yaml
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
pretrained_model_path: checkpoints/CompVis/stable-diffusion-v1-4
|
2 |
+
output_dir: /home/user/app/experiments/product-ads
|
3 |
+
train_data:
|
4 |
+
video_path: /tmp/gradio/e14d86daa040bee397ec527f496bcf79f217d00e/Tom Ford - Fougere
|
5 |
+
DArgent Spec Ad 1.mp3
|
6 |
+
prompt: A bottle doing a series of translations
|
7 |
+
n_sample_frames: 8
|
8 |
+
width: 512
|
9 |
+
height: 512
|
10 |
+
sample_start_idx: 0
|
11 |
+
sample_frame_rate: 1
|
12 |
+
validation_data:
|
13 |
+
prompts:
|
14 |
+
- A bottle doing a series of translations
|
15 |
+
video_length: 8
|
16 |
+
width: 512
|
17 |
+
height: 512
|
18 |
+
num_inference_steps: 50
|
19 |
+
guidance_scale: 7.5
|
20 |
+
validation_steps: 100
|
21 |
+
trainable_modules:
|
22 |
+
- attn1.to_q
|
23 |
+
- attn2.to_q
|
24 |
+
- attn_temp
|
25 |
+
train_batch_size: 1
|
26 |
+
max_train_steps: 300
|
27 |
+
learning_rate: 3.5e-05
|
28 |
+
scale_lr: false
|
29 |
+
lr_scheduler: constant
|
30 |
+
lr_warmup_steps: 0
|
31 |
+
adam_beta1: 0.9
|
32 |
+
adam_beta2: 0.999
|
33 |
+
adam_weight_decay: 0.01
|
34 |
+
adam_epsilon: 1.0e-08
|
35 |
+
max_grad_norm: 1.0
|
36 |
+
gradient_accumulation_steps: 1
|
37 |
+
gradient_checkpointing: true
|
38 |
+
checkpointing_steps: 1000
|
39 |
+
resume_from_checkpoint: null
|
40 |
+
mixed_precision: fp16
|
41 |
+
use_8bit_adam: false
|
42 |
+
enable_xformers_memory_efficient_attention: true
|
43 |
+
seed: 87217
|