• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1---
2name: TensorFlow Lite Converter Issue
3about: Use this template for reporting issues during model conversion to TFLite
4labels: 'TFLiteConverter'
5
6---
7
8### 1. System information
9
10- OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
11- TensorFlow installation (pip package or built from source):
12- TensorFlow library (version, if pip package or github SHA, if built from source):
13
14### 2. Code
15
16Provide code to help us reproduce your issues using one of the following options:
17
18#### Option A: Reference colab notebooks
19
201)  Reference [TensorFlow Model Colab](https://colab.research.google.com/gist/ymodak/e96a4270b953201d5362c61c1e8b78aa/tensorflow-datasets.ipynb?authuser=1): Demonstrate how to build your TF model.
212)  Reference [TensorFlow Lite Model Colab](https://colab.research.google.com/gist/ymodak/0dfeb28255e189c5c48d9093f296e9a8/tensorflow-lite-debugger-colab.ipynb): Demonstrate how to convert your TF model to a TF Lite model (with quantization, if used) and run TFLite Inference (if possible).
22
23```
24(You can paste links or attach files by dragging & dropping them below)
25- Provide links to your updated versions of the above two colab notebooks.
26- Provide links to your TensorFlow model and (optionally) TensorFlow Lite Model.
27```
28
29#### Option B: Paste your code here or provide a link to a custom end-to-end colab
30
31```
32(You can paste links or attach files by dragging & dropping them below)
33- Include code to invoke the TFLite Converter Python API and the errors.
34- Provide links to your TensorFlow model and (optionally) TensorFlow Lite Model.
35```
36
37### 3. Failure after conversion
38If the conversion is successful, but the generated model is wrong, then state what is wrong:
39
40- Model produces wrong results and/or has lesser accuracy.
41- Model produces correct results, but it is slower than expected.
42
43### 4. (optional) RNN conversion support
44If converting TF RNN to TFLite fused RNN ops, please prefix [RNN] in the title.
45
46### 5. (optional) Any other info / logs
47Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.
48