ttgump / scdeepcluster Goto Github PK

Hi,

Could you check and see whether the jupyter noebook is working fine? I tried to open it with different laptops and on none of them could get jupyter notebook loaded.

Thanks a lot.

hello, I have some question about the code.
the first question:
'''
zinb = ZINB(pi, theta=disp, ridge_lambda=self.ridge, debug=self.debug)
self.loss = zinb.loss
'''
the model contains the above loss function but I find that the loss function of ZINB class needs two parameters(y_true and y_pred), you may not pass arguments to it.
the second question:
'''
loss = self.model.train_on_batch(x = [x_counts[index * batch_size:(index + 1) * batch_size],
sf[index * batch_size : (index + 1) * batch_size]],
y = [p[index * batch_size:(index + 1) * batch_size],
raw_counts[index * batch_size:(index + 1) * batch_size]])
'''
what's the action of y? what can it calculate?
the third question:
the preprocessing progress transform the count data to normalized data, but you may pass the normalized data to the neural network, why not the count data based on your model?
thank you!

What is the difference between scDeepCluster.py,scDeepCluster_latent.py and scDeepCluster_estimate_k.py?

Hi Tiantain,

Is scDeepCluster available for HPC for example SDSC or TACC?

Thanks.

Shicheng

I have noticed the issue #6, so I have created the same environment as you mentioned in readme.md

After running the tutorial you provided in own machine,

the ARI,NMI is very low and did't achieved perfect result. It seems strange that scDeepCluster is unstable for this simulation data, And I noticed that the tutorial.ipynb you have running is not the version you provided in readme.md, the warning information in the tutorial.ipynb seems you are using python3.7, which is not compatible with the python 3.6.3 you mentioned in main page, can you give me some suggestions? Thank you!

hi, I 've downloaded and repeated your code successfully, then I got three clustering results called NMI,CA and ARI.but how can I find clustering information of every cell such as Fig. 3: Benchmark results on four real scRNA-seq datasets with true labels-b
fig

Hi Tian Tian:

I read the paper "Clustering single-cell RNA-seq data with a model-based deep learning approach" on nature machine intelligence, I can not repeat a few results in your paper, such as the result of Mouse ES cells dataset on Fig.3 in the paper, the CA, NMI and ARI is 0.98, 0.97, 0.97 in the paper, but I repeat is ACC: 0.7674, NMI: 0.6714, ARI: 0.5398.
In another hand, for the splatter dataset, the CA and ARI are 0.75, 0.55 when setting sigma to 0.2 in Supplementary Figure 3. The results I repeat are 0.4420 and 0.0390. so, maybe I need to set some parameters but I ignored, can you give me some suggestions? Thank you!

best wishes

when I run your code, I find that keras runs on cpu, not my gpu. I search for solution on the internet but get nothing. Thus can you tell me some of your opinion?

Hello, I set up the environment according to your installation requirements, but it has not been successful. I encountered the following problems.