jaechanglim / cvae

github for "Molecular generative model based on conditional variational autoencoder for de novo molecular design"

github for "Molecular generative model based on conditional variational autoencoder for de novo molecular design"

https://jcheminf.biomedcentral.com/articles/10.1186/s13321-018-0286-7

https://arxiv.org/abs/1806.05805

1. Fisrt, you need to prepare file containing smiles. each line of the file is smiles of a molecule (please see smiles.txt in this repo)
2. Calculate properties of molecules using following command

python cal_prop.py --input_filename=smiles.txt --output_filename=smiles_prop.txt

3. You can train a model using following command

python -u train.py --prop_file=smiles_prop.txt --save_dir=./save

4. After trining a model, you can generate molecules with desirable properties. Suppose you want to generate molecules whoe MW=300, LogP=4, and TPSA=100, then the command is following:

python sample.py --prop_file=smiles_prop.txt --save_file=save/model_3.ckpt-3 --target_prop='300 4 100' --result_filename=result.txt

The result will be written in result.txt

cvae's Issues

What's the version of tensorflow ?

I used tf1.13 trigger a error:

Traceback (most recent call last):
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1334, in _do_call
return fn(*args)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1319, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1407, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.InternalError: Blas GEMM launch failed : a.shape=(64, 512), b.shape=(512, 200), m=64, n=200, k=512
[[{{node MatMul_1}}]]
[[{{node Mean}}]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "train.py", line 71, in
cost = model.train(x, y, l, c)
File "/home/ziang/CVAE/model.py", line 143, in train
_, r_loss, l_loss = self.sess.run([self.opt, self.reconstr_loss, self.latent_loss], feed_dict = {self.X :x, self.Y:y, self.L : l, self.C : c})
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 929, in run
run_metadata_ptr)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1152, in _run
feed_dict_tensor, options, run_metadata)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1328, in _do_run
run_metadata)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1348, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InternalError: Blas GEMM launch failed : a.shape=(64, 512), b.shape=(512, 200), m=64, n=200, k=512
[[node MatMul_1 (defined at /home/ziang/CVAE/model.py:105) ]]
[[node Mean (defined at /home/ziang/CVAE/model.py:75) ]]

Caused by op 'MatMul_1', defined at:
File "train.py", line 52, in
args
File "/home/ziang/CVAE/model.py", line 22, in init
self._create_network()
File "/home/ziang/CVAE/model.py", line 66, in _create_network
self.latent_vector, self.mean, self.log_sigma = self.encode()
File "/home/ziang/CVAE/model.py", line 105, in encode
log_sigma = tf.matmul(h, self.weights['out_log_sigma'])+self.biases['out_log_sigma']
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/ops/math_ops.py", line 2455, in matmul
a, b, transpose_a=transpose_a, transpose_b=transpose_b, name=name)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/ops/gen_math_ops.py", line 5333, in mat_mul
name=name)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
op_def=op_def)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
return func(*args, **kwargs)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 3300, in create_op
op_def=op_def)
File "/gxr/ziang/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1801, in init
self._traceback = tf_stack.extract_stack()

InternalError (see above for traceback): Blas GEMM launch failed : a.shape=(64, 512), b.shape=(512, 200), m=64, n=200, k=512
[[node MatMul_1 (defined at /home/ziang/CVAE/model.py:105) ]]
[[node Mean (defined at /home/ziang/CVAE/model.py:75) ]]

About tensorflow

Hello, I noticed that many of the functions in your model.py involving the tensorflow library do not match the latest version of tensorflow. Do you have any updated code?

questions abount "Scaffold-based molecular design with a graph generative model"

I have read th paper and think that the project is particularly interesting. So I want to study the algorithm in the paper, but I found that the project of the repository on the github cannot be accessed-https://github.com/jaechanglim/GGM. Can you resume the access of the repository or provide a copy of the code? Thank you very much!
Best wishes.