TencentARC Photomaker x Colab
I'm having trouble pulling TencentARC Photomaker into Colab. Is there an established colab I can check out to see where I went wrong?
Thank you, I have it on my Colab but when running all the steps on G4, I still get: Is there a way for the colab to launch a gradio environment? to make uploading files and other easier?
RuntimeError Traceback (most recent call last)
in <cell line: 13>()
11 start_merge_step = 30
12
---> 13 images = pipe(
14 prompt=prompt,
15 input_id_images=input_id_images,
17 frames
/usr/local/lib/python3.10/dist-packages/diffusers/models/attention_processor.py in call(self, attn, hidden_states, encoder_hidden_states, attention_mask, temb, scale)
1257 # the output of sdp = (batch, num_heads, seq_len, head_dim)
1258 # TODO: add support for attn.scale when we move to Torch 2.1
-> 1259 hidden_states = F.scaled_dot_product_attention(
1260 query, key, value, attn_mask=attention_mask, dropout_p=0.0, is_causal=False
1261 )
RuntimeError: cutlassF: no kernel found to launch!
Maybe the CUDA memory issue
So my step is to connect to T4 GPU
- Execute all the blocks on the left and that's it?
- Perhaps change folder to get the different files
Am I supposed to get this:
/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning:
The secret HF_TOKEN does not exist in your Colab secrets.
To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.
You will be able to reuse this secret in all of your notebooks.
Please note that authentication is recommended but still optional to access public models or datasets.
Follow this PR for solving the problem:
https://huggingface.co/spaces/TencentARC/PhotoMaker-Style/discussions/10