-
Thank you for your reimplementation first! However, I found that the training speed of multi gpu mode is much slower than single gpu:
single GPU: 1.01s / iter
2 GPUs: 3.84s / iter
-
## Environment information (required)
### Diagnostics
Diagnostics output
``````
--- check: autoidentify
INFO: diagnose_tensorboard.py version e43767ef2b648d0d5d57c00f38ccbd38390e38da
-…
-
Hi, I am trying to load images by first encoding them from a buffer (np.frombuffer) and then decoding using ops.ImageDecoder. However, it gives me the error of 'failed: Encoded image has ambiguous for…
-
Here is my code.
```
with pipe:
jpegs, labels = fn.readers.caffe2(path=path,
shard_id=rank,
num…
-
1. According to user guide, both the min_filter_ and mag_filter of resize operator is DALIInterpType.INTERP_LINEAR;
2. But in the source code of file "operators/image/resize/resampling_attr.cpp".
…
-
hi! I have a server for client, main code like below:
```
class ExternalInputIterator(object):
def __init__(self, batch_size):
self.task_queue = []
self.batch_size = batch_siz…
-
Related to Model/Framework(s) or something else (describe)
Examples:
DALI version - 1.8.0
Is your feature request related to a problem? Please describe.
Need to process more flac files i…
-
When I use the nvidia-dali 0.10.0, "RuntimeError: [/opt/dali/dali/pipeline/operators/reader/video_reader_op.h:51] [/opt/dali/dali/pipeline/operators/reader/loader/video_loader.cc:59] ]
-
Hi its me again, I've been implement dali frame process with below code, but when I test 1000 cases compare whit origin torch process, there are 24 results is different. then we compare the features…
-
I am using the NVIDIA Dali pipeline to get an output as a TensorGPU object,
I like to share those TensorGPU objects (nvidia.dali.backend_impl.TensorGPU) among the other processes,
I following the st…