automl

automl copied to clipboard

Hi, I am currently training EfficientDet on a custom dataset. Due to the change of the available GPU resources, I sometimes would like to adjust the batch size accordingly. For...

ChulanZhang

Steps to reproduce: 1. Clone this repository 2. Install requirements: ``` cd efficientdet pip install --upgrade pip pip install numpy pip install -r requirements.txt ``` 3. Download Pretrained EfficientDet Checkpoints...

awlego

pipeline.config parameters in detail & how can we fine tune this parameters?

1

I'm not getting any explanation about parameters on google so I'm looking for efficientdet/ efficientnet v2 pipeline.config parameters with their detailed explanation. & how can we do finetuning on these...

akashAD98

KeyError: 'trainable_weights'

1

Traceback (most recent call last): File "train_RGB_softmax.py", line 351, in tf.app.run() File "C:\Users\Administrator\AppData\Roaming\Python\Python37\site-packages\tensorflow\python\platform\app.py", line 125, in run _sys.exit(main(argv)) File "train_RGB_softmax.py", line 194, in main logits = build_model(images_batch) File "train_RGB_softmax.py", line...

fanchunpeng

Docker Hosted automl predict ignores score_threshold param

3

I have deployed the .pb file exported out of GCP as per below instructions https://cloud.google.com/vision/automl/object-detection/docs/containers-gcs-tutorial I am able to get the response but I am getting all predictions with all...

ayandebbarman

How to use efficientNet2 as backbone for efficientdet?

9

drahmad89

After training an efficientdet-lite0 or efficientdet-d0 model with quantization aware training (using `keras.train --hparams="model_optimizations.quantize={}" ...`), both the saved model and exported tflite model achieve 0 mAP when evaluated with keras.eval...

mswoff

Convert to edge tpu tflite failed. model: https://drive.google.com/file/d/1p7RHElZHlxFiKUdAxSLKOWBqQY5cL8OT/view?usp=sharing html: https://drive.google.com/file/d/1vRhW4Inu0pFG9PQAeV9L8esUuKj4UDLq/view?usp=sharing ```edgetpu_compiler -s -o ./model_tpu.tflite model.tflite``` ``` Edge TPU Compiler version 15.0.340273435 terminating with uncaught exception of type std::out_of_range: vector Internal...

fsx950223

I downloaded the model from TFHub (link below), imported it using an Import feature in Android Studio (New -> Others -> TF Lite Model), and deployed it successfully on Pixel...

patrick-ucr

Per-class AP for specific classes and new datasets

4

Hello, I've been trying to get per-class AP results for evaluation, but I can't figure out how to make it work when evaluating a new dataset or setting specific classes...

JaydenKing32