-
Hi, I am having trouble in training a QuaNet quantifier for multiclass (20) data. Everything works fine with where my dataset only has 2 classes. It looks like the ACC quantifier is not able to aggreg…
-
### 1. System information
- OS Platform and Distribution: Ubuntu 18.04 (and Android for inference)
- TensorFlow installation: pip
- TensorFlow library: version 1.14.0
### 2. Code
I wish to …
-
I implement a crnn model with keras 2.0.8 that structure like below:
```
Layer (type) Output Shape Param # Connected to
=======================…
-
The BLSTM layer is defined in Keras like:
```
x2 = Bidirectional(LSTM(128,return_sequences=True),name='blstm')(x1)
```
which has 128 units. The input shape of BLSTM is [32,32,512] and the output …
-
Could this TCN implementation hypothetically be modified so that it can be wrapped by `tf.keras.layers.Bidirectional` ?
Currently it is not possible, as layers need the `go_backwards` attribute in …
-
I am trying to convert a keras model to coreml, but it fails every time with this error:
0 : input_1,
1 : dense_1,
2 : dense_1__activation__,
Neural Network compiler 0: 140 …
-
This toy example
```
import sys
import keras
from keras import Sequential
from keras.activations import linear
from keras.engine import InputLayer
from keras.layers import Dense
from keras.los…
-
I use command：
python -m tf2onnx.convert \
--graphdef pb_path \
--output onnx_path \
--inputs images:0 \
--outputs rnn_output:0 \
--opset 15
log:
Use `tf.compat.v1.graph_util.extract_sub_gra…
-
I'm using pytorch export bi-lstm onnx model, cant import, there is an error (barracuda package Version 3.0.0 - March 01, 2022);
But if I set the pytorch lstm parameter 'bidirectional' to False, the…
-
Repro:
```
benchmarks/torchbench.py -d cuda --inductor --training --float32 --no-skip -k demucs
benchmarks/torchbench.py -d cuda --inductor --training --float32 --no-skip -k tts_angular
```
Error…