liyu95 / supportnet Goto Github PK
View Code? Open in Web Editor NEWSupportNet: solving catastrophic forgetting in class incremental learning with support data
SupportNet: solving catastrophic forgetting in class incremental learning with support data
Hello, can you upload the breakhis and hela10 dataset in the src_image_data folder?
#load Pfam dataset and encoding them into array functional_domain_encoding_list=Pfam.Pfam_from_pickle_file_encoding( '/home/liy0f/ec_project/data_and_feature/Pfam_name_list_new_data.pickle', '/home/liy0f/ec_project/data_and_feature/Pfam_model_names_list.pickle')
Hello,about the two pickles,can you tell me how to get these and what datasets they belong to in your paper, can you provide the two pickles for me ? thank you very much.
Can I use other data sets in the code that you provide? Do I need to modify the data format?
Hi,
I noticed that the fisher information is calculated by the test data. See line 345 in src_image_data/train_procedures.py
fisher_info_val=previous_network.compute_fisher_information(X_test_converted,session)
But I think we should calculate it using the training data. Did I misunderstand something or is there any reason to use the test data? Thanks.
Best!
Hello,Because I don’t know how to deal with the data set when I run the experiment, can you provide the data set used in the code, thank you very much.
When the program is running to the end, there is a state of insufficient memory. My server's memory is 128G. Is it not enough for this program?
ok,I don't know why the last issues is closed, there is a question.
Traceback (most recent call last):
File "supportnet_ec.py", line 94, in
lambda: tf.constant(0.5, dtype=tf.float32))
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/util/ deprecation.py", line 432, in new_func
return func(*args, **kwargs)
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/ops/c ontrol_flow_ops.py", line 2040, in cond
orig_res_t, res_t = context_t.BuildCondBranch(true_fn)
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/ops/c ontrol_flow_ops.py", line 1890, in BuildCondBranch
original_result = fn()
File "supportnet_ec.py", line 93, in
tf.gather(final_feature, support_ind))/tf.cast(tf.size(support_ind),dtype=tf .float32),
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/ops/m ath_ops.py", line 874, in r_binary_op_wrapper
x = ops.convert_to_tensor(x, dtype=y.dtype.base_dtype, name="x")
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/frame work/ops.py", line 1011, in convert_to_tensor
as_ref=False)
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/frame work/ops.py", line 1107, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/frame work/constant_op.py", line 217, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/frame work/constant_op.py", line 196, in constant
value, dtype=dtype, shape=shape, verify_shape=verify_shape))
File "/home/xiaolin/.local/lib/python2.7/site-packages/tensorflow/python/frame work/tensor_util.py", line 424, in make_tensor_proto
raise ValueError("None values not supported.")
ValueError: None values not supported.
Now, I am searching the solution about the problem.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.