[Tests] fix sharding tests #8764
Conversation
| new_model = new_model.to(torch_device) | ||
|
|
||
| torch.manual_seed(0) | ||
| _, inputs_dict = self.prepare_init_args_and_inputs_for_common() |
There was a problem hiding this comment.
Otherwise the generator (which is required as an input when using the ConsistencyDecoderVAE will be consumed and its states will change in-place. So, we need to re-initialize here.
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
| del self.weight | ||
| self.W = nn.Parameter(torch.randn(embedding_size) * scale, requires_grad=False) | ||
|
|
||
| self.weight = self.W | ||
| del self.W |
There was a problem hiding this comment.
Necessary to mitigate the "tied weight tensors" error from safetensors.
There was a problem hiding this comment.
@sayakpaul curious about codes here, may i ask for more details if possible? ;)
| if seed is not None: | ||
| generator.manual_seed(0) |
There was a problem hiding this comment.
This was likely a mistake because if a seed is provided you would want to use that to do: generator.manual_seed(seed) and not generator.manual_seed(0) using a fixed seed.
SunMarc
left a comment
SunMarc
left a comment
There was a problem hiding this comment.
Thanks for fixing @sayakpaul !
What does this PR do?
So the following tests pass:
Some comments in line.