deeplab-training
deeplab-training copied to clipboard
heaversm
Training your own Deeplab Model in Tensorflow
this command doesnt work anymore google removed rg_meta `python3 utilities/scrapeImages.py --search "worker" --num_images 1 --directory "scrape/google"`
**I set the code in 'train-sqr.sh' for my datasets:** " ##Set up the working environment. CURRENT_DIR=$(pwd) WORK_DIR="${CURRENT_DIR}/deeplab" DATASET_DIR="datasets" ##Set up the working directories. PQR_FOLDER="20201206_test" EXP_FOLDER="exp/train_on_trainval_set" INIT_FOLDER="${WORK_DIR}/${DATASET_DIR}/${PQR_FOLDER}/${EXP_FOLDER}/init_models" TRAIN_LOGDIR="${WORK_DIR}/${DATASET_DIR}/${PQR_FOLDER}/${EXP_FOLDER}/train" DATASET="${WORK_DIR}/${DATASET_DIR}/${PQR_FOLDER}/tfrecord" mkdir -p...
``` File "C:/Users/ASDUser/Deeplab/models/research/deeplab/train.py", line 12, in from tensorflow.data import quantize as contrib_quantize ModuleNotFoundError: No module named 'tensorflow.contrib' ``` Hi I am following your step by step. All are success, unfortunately...
Running the file 'convert_rgb_to_index.py' in the dataset folder gives the following error message: File "convert_rgb_to_index.py", line 36, in arr = arr[:,:,0:3] IndexError: too many indices for array
When I run either the `eval` or `vis` script I am given the error `ensorflow.python.framework.errors_impl.InvalidArgumentError: padded_shape[0]=86 is not divisible by block_shape[0]=6` and `tensorflow.python.framework.errors_impl.InvalidArgumentError: padded_shape[0]=86 is not divisible by block_shape[0]=6` respectively....
Hey @heaversm Thanks for the detailed explanation. Since there is no validation graph, how to find if the deep lab model is overfitting or not?
Running _train-pqr.sh_ doesn't cause any errors, the model is saved. But running _eval-pqr.sh_ gives the error below: ``` I0422 13:24:17.127754 140272342116160 evaluation.py:450] Starting evaluation at 2020-04-22-11:24:17 < 2020-04-22 13:24:20.225281: W...
after running `sh train-pqr.sh ` am getting the error ``` ValueError: Shapes must be equal rank, but are 0 and 1 From merging shape 0 with other shapes. for 'pad_to_bounding_box/stack_4'...
Hello, I followed your tutorial but I'm getting the following error: `ERROR:tensorflow:================================== Object was never used (type ): If you want to mark it as used call its "mark_used()" method....
