jak2_data.csv - Githubissues

wojiaolyj commented 5 years ago

can't find jak2_data.csv in the clone,How to get jak2_data.csv，

isayev commented 5 years ago

Hey, @wojiaolyj in the paper we used proprietary data, that cannot be shared. You could use JAK2 (CHEMBL ID 2971) dataset or any other target of your interest.

wojiaolyj commented 5 years ago

Do the following data and modules also need to be downloaded by myself? where to download？ 1./home/yf302/mariewelt/Notebooks/PyTorch/data/egfr_with_pubchem.smi 2.my_generator.load_model('/home/mariewelt/Notebooks/PyTorch/Model_checkpoints/generator/checkpoint_lstm') 3.egfr_predictor.load_model('/home/mariewelt/Notebooks/PyTorch/data/RF/EGFR_RF')

Mariewelt commented 5 years ago

Hi @wojiaolyj

main.py file was put to the repository by mistake, thanks for pointing this out! I will remove it now, as it's not doing anything meaningful. We share ipython notebooks JAK2-demo.ipynb and LogP-demo.ipynb, that demonstrate how to run our algorithm. Data for LogP-demo.ipynb are available in the data folder. Unfortunately, we can't share jak2 data, but as has already been said, you can download dataset for this target (or any other target of your choice) from CHEMBL database.

Mariewelt commented 5 years ago

@wojiaolyj I uploaded JAK2 dataset from CHEMBL to the data folder.

wojiaolyj commented 5 years ago

Thank you very much for the data you uploaded. I successfully installed Rdkit once, and then I made an error when I executed the program，I reinstalled the system. After installing Rdkit, I always showed that the follow two tests failed. I tried many methods were not solved. Is related to cuda and cudnn?

98% tests passed, 2 tests failed out of 131

Total Test time (real) = 244.60 sec

The following tests FAILED: 127 - pythonTestDirDbase (Failed) 130 - pythonTestDirChem (Failed) Errors while running CTest

gmseabra commented 5 years ago

I'm sorry for coming back to this issue, but is there a chance you guys could upload at least a sample file, just so we can see the full format expected?

Thanks!

wojiaolyj commented 5 years ago

This problem occurs when I run the program JAK2-demo, my cuda version is cuda9.0.

hidden_size = 1500 stack_width = 1500 stack_depth = 200 lr = 0.01 my_generator = StackAugmentedGRU(input_size=gen_data.n_characters, hidden_size=hidden_size, output_size=gen_data.n_characters, stack_width=stack_width, stack_depth=stack_depth, use_cuda=use_cuda, n_layers=1, optimizer='Adadelta', lr=lr)

if use_cuda: my_generator = my_generator.cuda()

RuntimeError Traceback (most recent call last)

in () 12 13 if use_cuda: ---> 14 my_generator = my_generator.cuda() ~/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py in cuda(self, device) 256 Module: self 257 """ --> 258 return self._apply(lambda t: t.cuda(device)) 259 260 def cpu(self): ~/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py in _apply(self, fn) 183 def _apply(self, fn): 184 for module in self.children(): --> 185 module._apply(fn) 186 187 for param in self._parameters.values(): ~/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py in _apply(self, fn) 189 # Tensors stored in modules are graph leaves, and we don't 190 # want to create copy nodes, so we have to unpack the data. --> 191 param.data = fn(param.data) 192 if param._grad is not None: 193 param._grad.data = fn(param._grad.data) ~/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py in (t) 256 Module: self 257 """ --> 258 return self._apply(lambda t: t.cuda(device)) 259 260 def cpu(self): RuntimeError: CUDA error: unknown error

isayev / ReLeaSE

jak2_data.csv #7

if use_cuda: my_generator = my_generator.cuda()