Vishnuvardhan Janapati

I am able to reproduce the error. Please check the [gist here](https://colab.research.google.com/gist/jvishnuvardhan/606043f7066fa56ca7a7f4b62af361db/untitled.ipynb). The following is the error trace ``` Epoch 1/10 1/1 [==============================] - 0s 377ms/step - loss: 7.7125 ---------------------------------------------------------------------------...

@jacksonloper Currently there is some description [here](https://www.tensorflow.org/guide/gpu). You can set the device placement for any op. You can also check some useful info [here](https://www.tensorflow.org/guide/variable). Setting `set_log_device_placement` will place the ops...

This is still an issue. I ran the code with `tf-nightly` 2.9.0-dev20220218. [Here](https://colab.research.google.com/gist/jvishnuvardhan/dca4038de3baf142f40a42c95eaac56e/untitled.ipynb) is the gist for our reference. Thanks

I was able to reproduce the issue. Please check the [gist here](https://colab.research.google.com/gist/jvishnuvardhan/63f9ed182854ad7732ca265cca5c0d2a/test.ipynb). output from the code is as shown below ### Training with Estimator ``` INFO:tensorflow:loss = 0.027672563, step =...

Did this patch get merged? I cannot see the `show_trainable` argument in keras `master` https://github.com/mfidabel/keras/blob/master/keras/utils/layer_utils.py I tried using `show_trainable` in a simple example from TF tutorial (with `tf-nightly`), and it...

Reopened it. Sorry for the inconvenience. Removed the label. Thanks

@vyang2968 Can you please describe your question with more details? Do you want to download the rubber duckey model? or download the `eager_few_shot_od_training_tf2_colab.ipynb`? Thanks

@danijar Can you please share a simple standalone code to reproduce the issue. Thanks!

@nbro I think the error is coming from the line `model_output` while creating the model. `model_output = tfp.layers.DistributionLambda(lambda t: tfd.Categorical(logits=t[0], validate_args=True))(x) ` The shape of output layer is ((), ())...

@nbro It is clearly showing the root-cause of the error. When I added `print(help(y_pred))` in the loss function, it showed `Tensor in module tensorflow.python.framework.ops object`. As a tensor object, it...