-
Notifications
You must be signed in to change notification settings - Fork 605
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Compatibility with Tensorflow slim #140
Comments
Hi, |
from nvidia.dali.pipeline import Pipeline slim = tf.contrib.slim lmdb_folder = "/data/ilsvrc12_train_lmdb" tfrecord = "/data/imagenet/train-00001-of-01024" N = 4 # number of GPUs if not os.path.exists("idx_files"): if not os.path.isfile(tfrecord_idx): class CommonPipeline(Pipeline):
class CaffeReadPipeline(CommonPipeline):
class TFRecordPipeline(CommonPipeline):
def get_batch_test_dali(batch_size, pipe_type):
pipe_types = [[CaffeReadPipeline, tf.int32, (0, 999)], [TFRecordPipeline, tf.int64, (1, 1000)]] for pipe_name in pipe_types:
|
Hi,
|
One more comment, after all it looks like that defining labels shape cause additional problems in some networks. We will back out that change, you can address your problem calling:
For the output form the daliop. |
There seems to be a problem with compatibility with functions used in Tensorflow slim.
Ex code)
[images, labels] = get_batch_test_dali(FLAGS.batch_size)
batch_queue = slim.prefetch_queue.prefetch_queue(
[images, labels], capacity=2 * deploy_config.num_clones)
(Output)
...
File "/usr/local/lib/python3.5/dist-packages/tensorflow/contrib/slim/python/slim/data/prefetch_queue.py", line 78, in prefetch_queue
dtypes = [t.dtype for t in tensor_list]
File "/usr/local/lib/python3.5/dist-packages/tensorflow/contrib/slim/python/slim/data/prefetch_queue.py", line 78, in
dtypes = [t.dtype for t in tensor_list]
AttributeError: 'list' object has no attribute 'dtype'
Did you go over this part?
The text was updated successfully, but these errors were encountered: