Fine-tuning inception_v3 failed with tf0.12 in windows

[ligangcc92]

I run the command as the document saied:

python train_image_classifier.py --train_dir="D:\PythonProjects\flowers-model\inception_v3" --dataset_dir="D:\PythonProjects\train_data" --dataset_name=flowers --dataset_split_name=train --model_name=inception_v3 --checkpoint_path="D:\PythonProjects\pretrained_checkpoints\inception_v3.ckpt" --checkpoint_exclude_scopes=InceptionV3/Logits,InceptionV3/AuxLogits/Logits --trainable_scopes=InceptionV3/Logits,InceptionV3/AuxLogits/Logits

and get some error:

InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [5] rhs shape= [1001] [[Node: save/Assign_8 = Assign[T=DT_FLOAT, _class=["loc:@InceptionV3/AuxLogits/Conv2d_2b_1x1/biases"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/gpu:0"](InceptionV3/AuxLogits/Conv2d_2b_1x1/biases, save/RestoreV2_8/_5)]]

what's wrong with that?

[michaelisard]

This kind of usage question is better asked on Stackoverflow.