dspin 1.5.0__tar.gz → 1.5.2__tar.gz

{dspin-1.5.0/dspin.egg-info → dspin-1.5.2}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.1
 Name: dspin
-Version: 1.5.0
+Version: 1.5.2
 Summary: Regulatory network models from single-cell perturbation profiling
 Home-page: https://github.com/JialongJiang/DSPIN
 Author: Jialong Jiang
@@ -25,18 +25,6 @@ Requires-Dist: scanpy
 Requires-Dist: tqdm
 Requires-Dist: igraph
 Requires-Dist: leidenalg
-Dynamic: author
-Dynamic: author-email
-Dynamic: classifier
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: keywords
-Dynamic: license
-Dynamic: license-file
-Dynamic: requires-dist
-Dynamic: requires-python
-Dynamic: summary
 
 # D-SPIN: Constructing regulatory network models from scRNA-seq perturbation profiling  
 

{dspin-1.5.0 → dspin-1.5.2}/dspin/compute.py

@@ -852,11 +852,12 @@ def samp_moments(j_mat: np.ndarray,
 
 
 @numba.njit()
-def samp_mcmc_states(j_mat: np.ndarray, 
+def sample_states_mcmc(j_mat: np.ndarray, 
                     h_vec: np.ndarray, 
                     sample_size: int, 
                     mixing_time: int, 
-                    samp_gap: int) -> np.ndarray:
+                    samp_gap: int,
+                    seed: int = None) -> np.ndarray:
     """
     Sample states for the Markov Chain Monte Carlo (MCMC).
 
@@ -872,6 +873,8 @@ def samp_mcmc_states(j_mat: np.ndarray,
         Mixing time for the MCMC.
     samp_gap : int
         Gap between samples.
+    seed : int, optional
+        Seed for the random number generator. Default is None.
 
     Returns
     -------
@@ -891,8 +894,10 @@ def samp_mcmc_states(j_mat: np.ndarray,
     # Cast & make contiguous (Numba-safe)
     j_mat = np.ascontiguousarray(j_mat.astype(np.float64))
     # Accept both (n,) and (n,1) h_vec; flatten to 1D
-    h_vec = h_vec.reshape(h_vec.size)
-    h_vec = np.ascontiguousarray(h_vec.astype(np.float64))
+    h_vec = np.ascontiguousarray(h_vec.astype(np.float64)).reshape(- 1)
+
+    if seed is not None:
+        np.random.seed(seed)
 
     # Spins in {-1,0,1}
     cur_spin = (np.random.randint(0, 3, num_spin) - 1).astype(np.float64)
@@ -1272,7 +1277,7 @@ def learn_network_adam(raw_data: Any,
     rec_jgrad_sum_norm = np.inf * np.ones(num_epoch)
     mjj, vjj = np.zeros(cur_j.shape), np.zeros(cur_j.shape)
     mhh, vhh = np.zeros(cur_h.shape), np.zeros(cur_h.shape)
-    log_adam_grad = {name: deque(maxlen=backtrack_gap)
+    log_adam_grad = {name: deque(maxlen=backtrack_gap + 1)
                      for name in ["mjj", "vjj", "mhh", "vhh"]}
     
     apply_regularization(np.zeros((num_spin, num_spin, num_round)), np.zeros((num_spin, num_round)), cur_j, cur_h, train_dat, print_regu=True)
@@ -1322,7 +1327,6 @@ def learn_network_adam(raw_data: Any,
             pbar.set_postfix({"Network Gradient": f"{rec_jgrad_sum_norm[counter - 1]:.6f}"})
 
         # Handle backtracking
-        
         if_backtrack = False
 
         if (counter > backtrack_gap) and (rec_jgrad_sum_norm[counter - 1] > 1.5 * rec_jgrad_sum_norm[counter - 1 - backtrack_gap]):
@@ -1339,9 +1343,12 @@ def learn_network_adam(raw_data: Any,
         if if_backtrack:
             print('Backtracking at epoch %d' % counter)
             backtrack_counter += 1
+            cur_j = rec_jmat_all[counter - 1 - backtrack_gap, :, :]
+            cur_h = rec_hvec_all[counter - 1 - backtrack_gap, :, :]
+
             mjj, vjj, mhh, vhh = [log_adam_grad[key][0]
                                   for key in ['mjj', 'vjj', 'mhh', 'vhh']]
-            counter = counter - backtrack_gap
+            counter = counter - backtrack_gap + 1
             stepsz = stepsz / 4
             if backtrack_counter > backtrack_tol:
                 print(
@@ -1354,8 +1361,9 @@ def learn_network_adam(raw_data: Any,
     pbar.close()
 
     # Retrieve parameters corresponding to the minimum gradient norm found during training
-    trace_epoch = 50 + num_epoch - counter
-    pos = num_epoch - trace_epoch + np.argmin(rec_jgrad_sum_norm[- trace_epoch: ])
+    trace_range = np.sort(np.where(rec_jgrad_sum_norm < np.inf)[0])[- 50 :]
+    pos = trace_range[np.argmin(rec_jgrad_sum_norm[trace_range])]
+
     # print(pos)
     cur_h = rec_hvec_all[pos, :, :]
     cur_j = rec_jmat_all[pos, :, :]
@@ -1420,7 +1428,7 @@ def learn_program_regulators(gene_states: List[np.ndarray],
     rec_selfj_all = np.zeros((num_epoch, num_program))
     rec_selfh_all = np.zeros((num_epoch, num_program))
     
-    rec_grad = np.zeros(num_epoch)
+    rec_grad = np.ones(num_epoch) * np.inf
 
     m_interaction, v_interaction = np.zeros_like(cur_interaction), np.zeros_like(cur_interaction)
     m_selfj, v_selfj = np.zeros_like(cur_selfj), np.zeros_like(cur_selfj)
@@ -1440,7 +1448,7 @@ def learn_program_regulators(gene_states: List[np.ndarray],
     rec_selfj_grad = np.zeros((num_round, num_program))
     rec_selfh_grad = np.zeros((num_round, num_program))
 
-    log_adam_grad = {name: deque(maxlen=backtrack_gap)
+    log_adam_grad = {name: deque(maxlen=backtrack_gap + 1)
                      for name in ["m_interaction", "v_interaction", "m_selfj", "v_selfj", "m_selfh", "v_selfh"]}
 
     tqdm._instances.clear()
@@ -1539,8 +1547,9 @@ def learn_program_regulators(gene_states: List[np.ndarray],
     pbar.close()
 
     # Retrieve parameters corresponding to the minimum gradient norm found during training
-    trace_epoch = 50 + num_epoch - counter
-    pos = num_epoch - trace_epoch + np.argmin(rec_grad[- trace_epoch: ])
+    trace_range = np.sort(np.where(rec_grad < np.inf)[0])[- 50 :]
+    pos = trace_range[np.argmin(rec_grad[trace_range])]
+
     cur_interaction = rec_interaction_all[pos, :, :]
     cur_selfj = rec_selfj_all[pos, :]
     cur_selfh = rec_selfh_all[pos, :]

{dspin-1.5.0 → dspin-1.5.2/dspin.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.1
 Name: dspin
-Version: 1.5.0
+Version: 1.5.2
 Summary: Regulatory network models from single-cell perturbation profiling
 Home-page: https://github.com/JialongJiang/DSPIN
 Author: Jialong Jiang
@@ -25,18 +25,6 @@ Requires-Dist: scanpy
 Requires-Dist: tqdm
 Requires-Dist: igraph
 Requires-Dist: leidenalg
-Dynamic: author
-Dynamic: author-email
-Dynamic: classifier
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: keywords
-Dynamic: license
-Dynamic: license-file
-Dynamic: requires-dist
-Dynamic: requires-python
-Dynamic: summary
 
 # D-SPIN: Constructing regulatory network models from scRNA-seq perturbation profiling  
 

{dspin-1.5.0 → dspin-1.5.2}/setup.py

@@ -21,7 +21,7 @@ install_requires = [
 # ]
 
 # Managing your version (adjust accordingly)
-version = "1.5.0"  # You can manage versioning in a more sophisticated way if needed
+version = "1.5.2"  # You can manage versioning in a more sophisticated way if needed
 
 # Long description (usually a README file)
 readme = open("README_PyPI.md").read()  # Adjust the file name if necessary