Errors during "Generate animation frames"

[VicRejkia]

Everything worked fine till i got to the Generate animation frames part. It stopped on the fetching variations at this loop: for _ in range(tot_variations):

Fetching variations
1
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
[<ipython-input-12-ccdc85f3c37d>](https://localhost:8080/#) in <module>
    153         tot_variations -= curr_variation_count  # only generate variations we still need
    154         for _ in range(tot_variations):
--> 155             img = get_variations_w_init(prompt, init_image, start_schedule=(1-image_consistency))[0]
    156             save_frame(
    157                 img,

3 frames
[/usr/local/lib/python3.8/dist-packages/torch/autograd/grad_mode.py](https://localhost:8080/#) in decorate_context(*args, **kwargs)
     25         def decorate_context(*args, **kwargs):
     26             with self.clone():
---> 27                 return func(*args, **kwargs)
     28         return cast(F, decorate_context)
     29 

TypeError: __call__() missing 1 required positional argument: 'image'

[dmarx]

i'm wondering if maybe this was a bug due to a change in huggingface or something like that? I swear I was able to reproduce this not that long ago, but now I can't. Can you double check if this is still an issue for you? I'm wondering if maybe this was caused by something in an up-stream dependency which has now fixed the problem.

[SchmitzAndrew]

I got a very similar issue today. I was able to reproduce it after reloading the notebook and rerunning everything.

Fetching variations
1
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
[<ipython-input-20-6970025e684d>](https://localhost:8080/#) in <module>
    153         tot_variations -= curr_variation_count  # only generate variations we still need
    154         for _ in range(tot_variations):
--> 155             img = get_variations_w_init(prompt, init_image, start_schedule=(1-image_consistency))[0]
    156             save_frame(
    157                 img,

3 frames
[/usr/local/lib/python3.8/dist-packages/torch/autograd/grad_mode.py](https://localhost:8080/#) in decorate_context(*args, **kwargs)
     25         def decorate_context(*args, **kwargs):
     26             with self.clone():
---> 27                 return func(*args, **kwargs)
     28         return cast(F, decorate_context)
     29 

TypeError: __call__() got an unexpected keyword argument 'init_image'

[mjea1980]

Also having this same issue. Weird thing is I was able to generate a video yesterday, and then when I went back today hoping to do another, this issue is popping up.

Anybody found a work around?

[mjea1980]

Fetching variations 1

TypeError Traceback (most recent call last) in 153 tot_variations -= curr_variationcount # only generate variations we still need 154 for in range(tot_variations): --> 155 img = get_variations_w_init(prompt, init_image, start_schedule=(1-image_consistency))[0] 156 save_frame( 157 img,

3 frames /usr/local/lib/python3.8/dist-packages/torch/autograd/grad_mode.py in decorate_context(*args, kwargs) 25 def decorate_context(*args, *kwargs): 26 with self.clone(): ---> 27 return func(args, kwargs) 28 return cast(F, decorate_context) 29

TypeError: call() got an unexpected keyword argument 'init_image'

[mjea1980]

I got a very similar issue today. I was able to reproduce it after reloading the notebook and rerunning everything.

Fetching variations
1
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
[<ipython-input-20-6970025e684d>](https://localhost:8080/#) in <module>
    153         tot_variations -= curr_variation_count  # only generate variations we still need
    154         for _ in range(tot_variations):
--> 155             img = get_variations_w_init(prompt, init_image, start_schedule=(1-image_consistency))[0]
    156             save_frame(
    157                 img,

3 frames
[/usr/local/lib/python3.8/dist-packages/torch/autograd/grad_mode.py](https://localhost:8080/#) in decorate_context(*args, **kwargs)
     25         def decorate_context(*args, **kwargs):
     26             with self.clone():
---> 27                 return func(*args, **kwargs)
     28         return cast(F, decorate_context)
     29 

TypeError: __call__() got an unexpected keyword argument 'init_image'

Hey Andrew, I'm having this issue as well. Were you able to find a solution?

[SchmitzAndrew]

Unfortunately not. I've been able to generate videos in the past so I'm not sure what the cause is. Please let me know if you come up with anything.

[mjea1980]

Same with me. I had no issues 2 days ago, then yesterday that issue popped up. Hopefully it just kind of fixes itself. lol

[SchmitzAndrew]

@mjea1980 ,

I tried the solution proposed in https://github.com/dmarx/video-killed-the-radio-star/pull/96 but wasn't able to get it to work. Might be worth a shot.

[H2whoaMusic]

Hi all. New poster. I actually just posted the same problem in a separate thread as I was desperately looking for solutions and stumbled across this forum. So it looks like a general issue at the moment.

[mjea1980]

@H2whoaMusic @SchmitzAndrew I tried to put together a video this morning, and still no dice. Its generates such cool images, I wish it would just finish off the final step!! haha. Hopefully one day soon it starts working again :)

@dmarx any idea what the problem could be ? Looks like its been a known issue for a couple months now..

[H2whoaMusic]

@mjea1980 I know, I'm in the same boat. I've generated some truly fantastic images. @dmarx was kind enough to guide me on how to modify the storyboard to make really bespoke sequences, and I'm dying to get it pulled together!

[ianhirsch]

I'm getting the same issue and was able to create a video about a month ago! not sure what changed here

[DerisoryPull]

Same issue here! That...was not a good timing to buy Colab pro

[H2whoaMusic]

Same issue here! That...was not a good timing to buy Colab pro

Lol, same. But I have faith that a solution will be found. It's way out of my very limited coding experience, but it's such an impressive project that I don't doubt that it will get resolved.

[mjea1980]

Tried again today, still not working. I changed to a "write" token. Exact same issue. Disappointing because it generated some really cool images that I would love to use!!

[dmarx]

thanks for your patience folks and sorry for the delay, issue should be fixed now. Will close this after I've got confirmation from one of y'all that it's working for you as well.

@ianhirsch @DerisoryPull @H2whoaMusic @mjea1980 @SchmitzAndrew @VicRejkia

[ianhirsch]

You’re the man! I’ll give it a go this afternoon and report back 🫡

On Wed, Mar 22, 2023 at 3:46 PM David Marx @.***> wrote:

thanks for your patience folks and sorry for the delay, issue should be fixed now. Will close this after I've got confirmation from one of y'all that it's working for you as well.

@ianhirsch https://github.com/ianhirsch @DerisoryPull https://github.com/DerisoryPull @H2whoaMusic https://github.com/H2whoaMusic @mjea1980 https://github.com/mjea1980 @SchmitzAndrew https://github.com/SchmitzAndrew @VicRejkia https://github.com/VicRejkia

— Reply to this email directly, view it on GitHub https://github.com/dmarx/video-killed-the-radio-star/issues/94#issuecomment-1480160226, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJPPL2CDZK5U3A4FFYAR54DW5NJJ5ANCNFSM6AAAAAASYVXGUQ . You are receiving this because you were mentioned.Message ID: @.***>

[DerisoryPull]

Lol My Colab pro expired a couple hours ago...time to start the engines again I guess

[H2whoaMusic]

So far, so good. Created a new project and taken it right through to generating animation images. Currently running and appears to be working. Actively generating images, which it was not doing before. Will confirm once completed, but looks good. Thank you!

[H2whoaMusic]

I am now compiling my video! Was able to generate the animation frames successfully. Huge thanks to you, @dmarx