googlecolab / colabtools

Python libraries for Google Colaboratory
Apache License 2.0
2.2k stars 722 forks source link

Tensorflow Lite Tutorial: Notebook command fails with error: Object Detection with TensorFlow Lite Model Maker #2836

Closed bantar76 closed 2 years ago

bantar76 commented 2 years ago

Describe the current behavior I am following the tutorial using the jupyter notebook at the following link: https://colab.research.google.com/github/tensorflow/tensorflow/blob/master/tensorflow/lite/g3doc/models/modify/model_maker/object_detection.ipynb#scrollTo=kwlYdTcg63xy. I am trying to use training data in the Pascal VOC format as created by the labelImg tool. I modified the command at Step 2. Load the Dataset:

original command in notebook: train_data, validation_data, test_data = object_detector.DataLoader.from_csv('gs://cloud-ml-data/img/openimage/csv/salads_ml_use.csv')

my modified command per instructions: train_data, validation_data, test_data = object_detector.DataLoader.from_pascal_voc('./sample_data/train_data', './sample_data/validation_data', ["thing1", "thing2", "thing3", "thing4"], ['1','3','4','5','7','8','9'])

This is formatted per instructions found in the "Load your data with a different data format" section:

object_detector.DataLoader.from_pascal_voc(image_dir, annotations_dir, label_map={1: "person", 2: "notperson"})

The command fails with the following messages:

TypeError Traceback (most recent call last)

in () ----> 1 train_data, validation_data, test_data = object_detector.DataLoader.from_pascal_voc('./sample_data/train_data', './sample_data/validation_data', ["thing1", "thing2", "thing3", "thing4"], ['1','3','4','5','7','8','9']) TypeError: cannot unpack non-iterable DataLoader object **Describe the expected behavior** The command should complete successfully. **What web browser you are using** Chromium on Raspberry Pi Raspian OS **Additional context** Link to a minimal, public, self-contained notebook that reproduces this issue. - Share the file using your GitHub account using File > Save a copy as a GitHub Gist. - or Share Drive notebooks using the Share button then 'Get Shareable Link'. Notebook to repro the issue: https://colab.research.google.com/github/tensorflow/tensorflow/blob/master/tensorflow/lite/g3doc/models/modify/model_maker/object_detection.ipynb#scrollTo=kwlYdTcg63xy.
bantar76 commented 2 years ago

I found the appropriate syntax for the command. The tutorial page should be updated with the appropriate syntax.