-
Notifications
You must be signed in to change notification settings - Fork 2k
Description
Please make sure that this is a build/installation issue. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:build_template
System information
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Codesandbox
- Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device:
- TensorFlow.js installed from (npm or script link): npm
- TensorFlow.js version: 1.7.4
- CUDA/cuDNN version:
Describe the problem
I found a problem that my models worked in Python, but I did not work in TFJS. My steps were below:
Step 1) Create model
I created the object detection model with my custom dataset and efficientdet-d0.
You can see more details from my Colab belew:
https://colab.research.google.com/drive/1fjnVbDnHnGnILuvDTv7zAwUwIMntrC7p?usp=sharing
Step 2) convert saved_model to web_model
tensorflowjs_converter \
--input_format=tf_saved_model \
--output_format=tfjs_graph_model \
--signature_name=serving_default \
--saved_model_tags=serve \
saved_model \
web_modelStep 3) import it on my Code sandbox but it didn't work
https://codesandbox.io/s/object-detection-with-tensorflow-js-forked-6r45k?file=/src/index.js
| Python | JS |
|---|---|
![]() |
![]() |
Any other info / logs
When I replaced my model with Kangaroo-detector from this blog post...And it worked?! I have no idea what makes this kinda difference.


