tonylianlong / animegan.js Goto Github PK

When I converted saved_model into tfjs, there was a problem. I think you might have encountered it. How did you solve it?
the xshell scripts is :
CUDA_VISIBLE_DEVICES=0 tensorflowjs_converter --input_format=tf_saved_model --output_format=tfjs_graph_model --signature_name='custom_signature' --output_node_names="output" --saved_model_tags=serve ./Hayao_model-64 ./web_model

the error infomation is:
ValueError: Unsupported Ops in the model before optimization MirrorPad (base) Workstation:/ada/AnimeGAN_js$

The MirrorPad corresponds to the reflect mode of tf. pad in TF.
Can you give me some help

hi can the weights be converted to pb or onnx for mobile app and will it be under 10mb thanks

First of all. I really appreciate to you for sharing nice codes. It was really impressive to me about utilizing tensorflow-js model. I am pretty newbe of tensorflow js and I want to be good at exporting python model into javascript via tf-js model.

1. Is there any tutorial or lecture to learn about exporting tf-js model using javascripts? otherwise how did you learn?
2. Which develop environment do you prefer to debug javascript?( like yarn + chrome). Because visualizing parameters of model is really hard to me.

While i try to make an image w/o resize i got an error and monitors blinks for a second.
Task manager shows 12GB VRAM overflow.
Error encountered while generating image: Error: Failed to compile fragment shader.

Most likely i'm already out of trends 👍
Gotta wait for 3090

Hi,
I have something to tell you.
Because there is no corresponding paper for open source AnimeGANv2, I recently made a technical description page of AnimeGANv2 using GitHub page, and found that GitHub page can't be accessed normally in China. Therefore, I plan to stop doing AnimeGANv2 online use website. If it's convenient for you, I hope you can implement it like v1.