KrutikaBM commited on
Commit
efa49e8
1 Parent(s): b2a8f84

Upload 4 files

Browse files
configs/car-turn.yaml ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pretrained_model_path: "./checkpoints/stable-diffusion-v1-4"
2
+ output_dir: "./outputs/car-turn"
3
+
4
+ train_data:
5
+ video_path: "data/car-turn.mp4"
6
+ prompt: "a jeep car is moving on the road"
7
+ n_sample_frames: 24
8
+ width: 512
9
+ height: 512
10
+ sample_start_idx: 0
11
+ sample_frame_rate: 2
12
+
13
+ validation_data:
14
+ prompts:
15
+ - "a jeep car is moving on the beach"
16
+ - "a jeep car is moving on the snow"
17
+ - "a jeep car is moving on the road, cartoon style"
18
+ - "a sports car is moving on the road"
19
+ video_length: 24
20
+ width: 512
21
+ height: 512
22
+ num_inference_steps: 50
23
+ guidance_scale: 12.5
24
+ use_inv_latent: True
25
+ num_inv_steps: 50
26
+
27
+ learning_rate: 3e-5
28
+ train_batch_size: 1
29
+ max_train_steps: 500
30
+ checkpointing_steps: 1000
31
+ validation_steps: 100
32
+ trainable_modules:
33
+ - "attn1.to_q"
34
+ - "attn2.to_q"
35
+ - "attn_temp"
36
+
37
+ seed: 33
38
+ mixed_precision: fp16
39
+ use_8bit_adam: False
40
+ gradient_checkpointing: True
41
+ enable_xformers_memory_efficient_attention: True
configs/man-skiing.yaml ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pretrained_model_path: "./checkpoints/stable-diffusion-v1-4"
2
+ output_dir: "./outputs/man-skiing"
3
+
4
+ train_data:
5
+ video_path: "data/man-skiing.mp4"
6
+ prompt: "a man is skiing"
7
+ n_sample_frames: 24
8
+ width: 512
9
+ height: 512
10
+ sample_start_idx: 0
11
+ sample_frame_rate: 2
12
+
13
+ validation_data:
14
+ prompts:
15
+ - "mickey mouse is skiing on the snow"
16
+ - "spider man is skiing on the beach, cartoon style"
17
+ - "wonder woman, wearing a cowboy hat, is skiing"
18
+ - "a man, wearing pink clothes, is skiing at sunset"
19
+ video_length: 24
20
+ width: 512
21
+ height: 512
22
+ num_inference_steps: 50
23
+ guidance_scale: 12.5
24
+ use_inv_latent: True
25
+ num_inv_steps: 50
26
+
27
+ learning_rate: 3e-5
28
+ train_batch_size: 1
29
+ max_train_steps: 500
30
+ checkpointing_steps: 1000
31
+ validation_steps: 100
32
+ trainable_modules:
33
+ - "attn1.to_q"
34
+ - "attn2.to_q"
35
+ - "attn_temp"
36
+
37
+ seed: 33
38
+ mixed_precision: fp16
39
+ use_8bit_adam: False
40
+ gradient_checkpointing: True
41
+ enable_xformers_memory_efficient_attention: True
configs/man-surfing.yaml ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pretrained_model_path: "./checkpoints/stable-diffusion-v1-4"
2
+ output_dir: "./outputs/man-surfing"
3
+
4
+ train_data:
5
+ video_path: "data/man-surfing.mp4"
6
+ prompt: "a man is surfing"
7
+ n_sample_frames: 24
8
+ width: 512
9
+ height: 512
10
+ sample_start_idx: 0
11
+ sample_frame_rate: 1
12
+
13
+ validation_data:
14
+ prompts:
15
+ - "a panda is surfing"
16
+ - "a boy, wearing a birthday hat, is surfing"
17
+ - "a raccoon is surfing, cartoon style"
18
+ - "Iron Man is surfing in the desert"
19
+ video_length: 24
20
+ width: 512
21
+ height: 512
22
+ num_inference_steps: 50
23
+ guidance_scale: 12.5
24
+ use_inv_latent: True
25
+ num_inv_steps: 50
26
+
27
+ learning_rate: 3e-5
28
+ train_batch_size: 1
29
+ max_train_steps: 500
30
+ checkpointing_steps: 1000
31
+ validation_steps: 100
32
+ trainable_modules:
33
+ - "attn1.to_q"
34
+ - "attn2.to_q"
35
+ - "attn_temp"
36
+
37
+ seed: 33
38
+ mixed_precision: fp16
39
+ use_8bit_adam: False
40
+ gradient_checkpointing: True
41
+ enable_xformers_memory_efficient_attention: True
configs/rabbit-watermelon.yaml ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pretrained_model_path: "./checkpoints/stable-diffusion-v1-4"
2
+ output_dir: "./outputs/rabbit-watermelon"
3
+
4
+ train_data:
5
+ video_path: "data/rabbit-watermelon.mp4"
6
+ prompt: "a rabbit is eating a watermelon"
7
+ n_sample_frames: 24
8
+ width: 512
9
+ height: 512
10
+ sample_start_idx: 0
11
+ sample_frame_rate: 2
12
+
13
+ validation_data:
14
+ prompts:
15
+ - "a tiger is eating a watermelon"
16
+ - "a rabbit is eating an orange"
17
+ - "a rabbit is eating a pizza"
18
+ - "a puppy is eating an orange"
19
+ video_length: 24
20
+ width: 512
21
+ height: 512
22
+ num_inference_steps: 50
23
+ guidance_scale: 12.5
24
+ use_inv_latent: True
25
+ num_inv_steps: 50
26
+
27
+ learning_rate: 3e-5
28
+ train_batch_size: 1
29
+ max_train_steps: 500
30
+ checkpointing_steps: 1000
31
+ validation_steps: 100
32
+ trainable_modules:
33
+ - "attn1.to_q"
34
+ - "attn2.to_q"
35
+ - "attn_temp"
36
+
37
+ seed: 33
38
+ mixed_precision: fp16
39
+ use_8bit_adam: False
40
+ gradient_checkpointing: True
41
+ enable_xformers_memory_efficient_attention: True