check_n_epoch=10.
Lr0.001.
Cost=v1.
C1r=0.5.
C2cutoff=0.3.
Down = neighbors0_200_200.
Ratio = 0.125.
N=30.
O=Adam.
to be added
ratio_list=[1] (generated by Dr. Wang)
ratio_list=[0.2,0.4,0.6,0.8] path1_sample=1000 #upper path2_sample=path1_sample/ratio_list
genn_amt=1000.
beta1_list=np.random.uniform(low=9, high=11, size=(genn_amt,)) #upper side.
beta2_list=np.random.uniform(low=30, high=50, size=(genn_amt,)) #downside.
gamma1_list=np.random.uniform(low=30, high=50, size=(genn_amt,)).
gamma2_list=np.random.uniform(low=5, high=25, size=(genn_amt,)).
path1_pct=95.
path2_pct=95.
alpha1_list=np.random.uniform(low=20, high=35, size=(genn_amt,)).
alpha2_list=np.random.uniform(low=45, high=55, size=(genn_amt,)).
check_n_epoch = 5.
ratio=0.2.
Lr=0.001.
trace=R0.
corrcoef=R0.
C2cf=0.3.
Down=neighbors0_200_200.
Ratio=0.125.
N=30.
O=Adam.
ratio_list=[0.2,0.4,0.6,0.8].
path1_sample=1000 #upper.
path2_sample=path1_sample/ratio_list.
genn_amt=1000
alpha1_list=np.random.uniform(low=10, high=15, size=(genn_amt,)).
alpha2_list=np.random.uniform(low=30, high=40, size=(genn_amt,)).
beta1_list=np.random.uniform(low=0.9, high=1.1, size=(genn_amt,)).
beta2_list=np.random.uniform(low=0.9, high=1.1, size=(genn_amt,)).
gamma1_list=np.random.uniform(low=0.125, high=0.125, size=(genn_amt,)) NOTICE - CURRENT VERSION: low=0.125, high=0.125, high should be 0.2727272727272727.
gamma2_list=np.random.uniform(low=0.2727, high=0.2727, size=(genn_amt,)) NOTICE - CURRENT VERSION: low=0.2727, high=0.2727, high should be 0.4.
path1_pct_list=np.random.uniform(low=90, high=95, size=(genn_amt,)).
path2_pct_list=np.random.uniform(low=50, high=70, size=(genn_amt,)).
epoch=200.
check_n=None.
Lr=0.001.
C2cf=0.3.
Down=neighbors0_200_200.
Ratio=0.125.
N=30.
O=Adam.
traceR=0.0.
corrcoefR=0.0.
genn_amt=1000
start_s1_list=np.random.uniform(low=3, high=4, size=(genn_amt,)).
start_s2_list=np.random.uniform(low=3.5, high=4.5, size=(genn_amt,)).
start_u1_list=np.random.uniform(low=2, high=3, size=(genn_amt,)).
start_u2_list=np.random.uniform(low=0.5, high=1, size=(genn_amt,)).
beta1_list=np.random.uniform(low=1.1, high=1.2, size=(genn_amt,)).
beta2_list=np.random.uniform(low=0.8, high=0.9, size=(genn_amt,)).
gamma1_list=np.random.uniform(low=1.1, high=1.2, size=(genn_amt,)).
gamma2_list=np.random.uniform(low=0.8, high=0.9, size=(genn_amt,)).
epoch200 check_nNone Lr0.001 C2cf0.3 Downneighbors0_200_200 Ratio0.125 N30 OAdam traceR0.0 corrcoefR0.0
final selection: n=200, gene amount=1600.
random.seed(10).
embedding_downsampling, sampling_ixs, knn_embedding = downsampling_embedding(data_df,
para='neighbors',
target_amount=0,
step_i=60,
step_j=60,
n_neighbors=n_neighbors).