Comments (5)
Hi,
Thanks for reaching out!
However, when I downloading the test dataset, it automatically start downloading the file with training dataset
Apologies! This was a bug in the README, which we just fixed. The link should now point to the correct file.
Also, when I trying to run "python train.py", it is missing "dataset_spec.pb" file.
Could you please provide the entire error message?
Thanks
Tim
from human-scene-transformer.
Thanks for updating the test dataset, but the dataset itself is missing "labels" folder. So when we are running the jrdb_preprocess_test.py it can not correctly generate the dataset.
The error message we got from train.py is the following:
WARNING:tensorflow:From /home/cam/human-scene-transformer/human_scene_transformer/jrdb/input_fn.py:555: load (from tensorflow.python.data.experimental.ops.io) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.data.Dataset.load(...)
instead.
W1103 14:49:08.859732 140565449860928 deprecation.py:364] From /home/cam/human-scene-transformer/human_scene_transformer/jrdb/input_fn.py:555: load (from tensorflow.python.data.experimental.ops.io) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.data.Dataset.load(...)
instead.
Traceback (most recent call last):
File "train.py", line 141, in
app.run(main)
File "/home/cam/.local/lib/python3.8/site-packages/absl/app.py", line 308, in run
_run_main(main, args)
File "/home/cam/.local/lib/python3.8/site-packages/absl/app.py", line 254, in _run_main
sys.exit(main(argv))
File "train.py", line 102, in main
ds_train = jrdb_input_fn.get_train_dataset(dataset_params)
File "/home/cam/human-scene-transformer/human_scene_transformer/jrdb/input_fn.py", line 709, in get_train_dataset
ds_train = load_dataset(
File "/home/cam/human-scene-transformer/human_scene_transformer/jrdb/input_fn.py", line 614, in load_dataset
scene_dataset = _load_scene_with_features(
File "/home/cam/human-scene-transformer/human_scene_transformer/jrdb/input_fn.py", line 555, in _load_scene_with_features
raw_dataset = tf.data.experimental.load(
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/util/deprecation.py", line 371, in new_func
return func(*args, **kwargs)
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/data/experimental/ops/io.py", line 166, in load
return dataset_ops.Dataset.load(path, element_spec, compression, reader_func)
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/data/ops/dataset_ops.py", line 1811, in load
return load_op._load(
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/data/ops/load_op.py", line 28, in _load
return _LoadDataset(path, element_spec, compression, reader_func)
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/data/ops/load_op.py", line 46, in init
encoded_spec = f.read()
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/lib/io/file_io.py", line 116, in read
self._preread_check()
File "/home/cam/.local/lib/python3.8/site-packages/tensorflow/python/lib/io/file_io.py", line 77, in _preread_check
self._read_buf = _pywrap_file_io.BufferedInputStream(
tensorflow.python.framework.errors_impl.NotFoundError: <dataset_path>/bytes-cafe-2019-02-07_0/agents/position/dataset_spec.pb; No such file or directory
I'm guessing it may be raised because we're not having the correct training dataset?
Lastly, can you provide a pretrained model or a simple testing model to let us have a basic idea of knowing what's the format of output?
from human-scene-transformer.
Hi,
If the pre-processing and creation of the dataset does not succeed, the training procedure is expected to fail. Could you please post the error message you are getting during dataset generation?
Thanks for updating the test dataset, but the dataset itself is missing "labels" folder.
Is this for the JRDB train or test dataset? For the train dataset, the labels are expected to be part of the download. For the test dataset you will have to separately download the labels as described here [1].
Lastly, can you provide a pretrained model or a simple testing model to let us have a basic idea of knowing what's the format of output?
You can download the pre-trained checkpoints here [2].
Let me know if you have any other questions.
Best
Tim
[1] https://github.com/google-research/human-scene-transformer/tree/main/human_scene_transformer/data#get-the-leaderboard-test-set-tracks
[2] https://storage.googleapis.com/gresearch/human_scene_transformer/checkpoints.zip
from human-scene-transformer.
Hi Charlotte,
just checking in to see if this solved your problem?
Let me know if you have any other questions. Otherwise, feel free to close this issue.
Best
Tim
from human-scene-transformer.
Thank you for the response, we still getting some problems but its all about our dependencies. I think we're all good here. Thanks again
from human-scene-transformer.
Related Issues (3)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from human-scene-transformer.