automl

automl copied to clipboard

location of inference results (bbox coordinates) - where?

5

I am running an inference (runmode=saved_model_infer, with d7x version) on my own images. They are located in "testdata" dir. The inference is successful, with the annotated images written to "serve_image_out"...

romanon5

I have a dataset that has huge discrepancies in number of elements in different categories. Sometimes in evaluation set there are no representatives of some classes. What I found is...

aljed

MultiWorkerMirrorStrategy for distributed training not working in gpus

1

Hi I am using `MultiWorkerMirrorStrategy` and `tf.estimator.train_and_evaluate` for distributed training with 3 epoch. Please find below the information: ``` GPU: 4 x NVIDIA Tesla V100 Datasets: COCOA Model: Efficientdet-d5 Tensorflow:...

ankur47

Referring to: https://colab.research.google.com/github/google/automl/blob/master/efficientnetv2/tfhub.ipynb. `get_hub_url_and_isize()` already returns the image size corresponding to a given model variant. But the following code is not using it: ```py hub_url, image_size = get_hub_url_and_isize(model_name, ckpt_type, hub_type)...

sayakpaul

issues of multi-Gpus training

28

I training on custom data set based on Window 10. I don't know what happened in the training. Anyone can tell me about this? ----------------------------------------------------- 2021-03-30 22:53:06.333435: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully...

Ronald-Kray

Version automl/efficientdet/keras/ tag: 1.2 I have two trainings one after another. First training:  config.yaml contains:  Second training starts from first training pretrained_ckpt.  config.yaml contains:  epoch_learning_rate tensorboard...

lejekjr

EarlyStopping

1

Is EarlyStopping available for keras backend training?

lejekjr

Now I'm finetuning efficientdet with my custom dataset. My dataset includes 171,120 training data and 11,558 validation data. I know that usually num_examples_per_epoch is same with the amount of data....

horangjang77

Loss Decreases during Training, but AP and AR is -1

2

I am trying to fine-tune the model on the MOT15 dataset found [here](https://motchallenge.net/data/MOT15/) I have added a custom script called `create_mot15_tfrecord.py` to covert the MOT15 ground truth data to TF...

BhandarkarPawan

Hi, I try to train EfficientDet-d7x on my custom dataset. Here is my config.yaml:  Here is my command:  Here is my evaluation result during training:  The overall...

puffdrum