Is it normal so much time training for Training Custom Object Detector??

[hobbyist]

I am following this tutorial: https://tensorflow-object-detection-api-...ining.html When I reach the training paragraph (Training the Model) and run this command:

python3 model_main_tf2.py --model_dir=models/my_ssd_resnet50_v1_fpn --pipeline_config_path=models/my_ssd_resnet50_v1_fpn/pipeline.config

I get mesages like this:

    Instructions for updating:
    `seed2` arg is deprecated.Use sample_distorted_bounding_box_v2 instead.
    W0128 18:24:29.575707 140532950755136 deprecation.py:341] From /usr/local/lib/python3.9/dist-packages/tensorflow/python/util/dispatch.py:1096: sample_distorted_bounding_box (from tensorflow.python.ops.image_ops_impl) is deprecated and will be removed in a future version.
    Instructions for updating:
    `seed2` arg is deprecated.Use sample_distorted_bounding_box_v2 instead.
    WARNING:tensorflow:From /usr/local/lib/python3.9/dist-packages/tensorflow/python/autograph/impl/api.py:465: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
    Instructions for updating:
    Use `tf.cast` instead.
    W0128 18:24:30.771597 140532950755136 deprecation.py:341] From /usr/local/lib/python3.9/dist-packages/tensorflow/python/autograph/impl/api.py:465: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
    Instructions for updating:
    Use `tf.cast` instead.
    2022-01-28 18:24:33.596621: W tensorflow/core/framework/dataset.cc:744] Input of GeneratorDatasetOp::Dataset will not be optimized because the dataset does not implement the AsGraphDefInternal() method needed to apply optimizations.

And as I see from htop command the threads still running. Is it normal? I mean the training is more than 48 hours... and is still training, it hasn't stop yet. I also tried this: https://tensorflow-object-detection-api-...rboard-sec in order to visualize what is happening on training but nothing was displayed... I include a printscreen below... Do you see something strange?

[hobbyist]

Ok, I have made the .pb file after many tries. I am following this tutorial: https://medium.com/analytics-vidhya/crea...cfff3e2114 However, I get this error:

     23 #Run the session using the tensors and feed the image to the session
     24 #img_in = cv2.resize(img_cv2, (224, 224))
---> 25     img_in = img_cv2[:, :, [2, 1, 0]]  # BGR2RGB
     26 outputs = sess.run([sess.graph.get_tensor_by_name('num_detections:0'),
     27              sess.graph.get_tensor_by_name('detection_scores:0'),

NameError: name 'img_cv2' is not defined

Any idea how to fix it?

[aserikova]

Did you solve problem with input of GeneratorDatasetOp::Dataset will not be optimized because the dataset does not implement the AsGraphDefInternal() method needed to apply optimizations.????