eqc-models 0.11.0__py3-none-any.whl → 0.12.0__py3-none-any.whl

{eqc_models-0.11.0.data → eqc_models-0.12.0.data}/platlib/eqc_models/ml/classifierqsvm.py

@@ -8,28 +8,224 @@ import json
 import warnings
 from functools import wraps
 import numpy as np
+from sklearn.preprocessing import MinMaxScaler
 
 from eqc_models.ml.classifierbase import ClassifierBase
 
 
 class QSVMClassifier(ClassifierBase):
     """An implementation of QSVM classifier that uses QCi's Dirac-3.
-    
+
     Parameters
     ----------
-    
+
     relaxation_schedule: Relaxation schedule used by Dirac-3; default:
     2.
+
+    num_samples: Number of samples used by Dirac-3; default: 1.
+
+    solver_access: Solver access type: cloud or direct; default: cloud.
+
+    ip_addr: IP address of the device when direct access is used; default: None.
+    
+    port: Port number of the device when direct access is used; default: None.    
     
+    lambda_coef: The penalty multipler
+
+    Examples
+    -----------
+
+    >>> from sklearn import datasets
+    >>> from sklearn.preprocessing import MinMaxScaler
+    >>> from sklearn.model_selection import train_test_split
+    >>> iris = datasets.load_iris()
+    >>> X = iris.data
+    >>> y = iris.target
+    >>> scaler = MinMaxScaler()
+    >>> X = scaler.fit_transform(X)
+    >>> for i in range(len(y)):
+    ...     if y[i] == 0:
+    ...         y[i] = -1
+    ...     elif y[i] == 2:
+    ...         y[i] = 1
+    >>> X_train, X_test, y_train, y_test = train_test_split(
+    ...     X,
+    ...     y,
+    ...     test_size=0.2,
+    ...     random_state=42,
+    ... )
+    >>> from eqc_models.ml.classifierqsvm import QSVMClassifier
+    >>> obj = QSVMClassifier(
+    ...     relaxation_schedule=2,
+    ...     num_samples=1,
+    ... )
+    >>> from contextlib import redirect_stdout
+    >>> import io
+    >>> f = io.StringIO()
+    >>> with redirect_stdout(f):
+    ...    obj = obj.fit(X_train, y_train)
+    ...    y_train_prd = obj.predict(X_train)
+    ...    y_test_prd = obj.predict(X_test)
+
+    """
+
+    def __init__(
+        self,
+        relaxation_schedule=1,
+        num_samples=1,
+        solver_access="cloud",
+        ip_addr=None,
+        port=None,            
+        lambda_coef=1.0,
+    ):
+        super(QSVMClassifier).__init__()
+
+        assert solver_access in ["cloud", "direct"]
+        
+        self.relaxation_schedule = relaxation_schedule
+        self.num_samples = num_samples
+        self.solver_access = solver_access
+        self.ip_addr = ip_addr
+        self.port = port        
+        self.lambda_coef = lambda_coef
+        self.fea_scaler = MinMaxScaler(feature_range=(-1, 1))
+
+    def fit(self, X, y):
+        """
+        Build a QSVM classifier from the training set (X, y).
+
+        Parameters
+        ----------
+        X : {array-like, sparse matrix} of shape (n_samples, n_features)
+        The training input samples.
+
+        y : array-like of shape (n_samples,)
+        The target values.
+
+        Returns
+        -------
+        Response of Dirac-3 in JSON format.
+        """
+
+        assert X.shape[0] == y.shape[0], "Inconsistent sizes!"
+
+        assert set(y) == {-1, 1}, "Target values should be in {-1, 1}"
+
+        X = self.fea_scaler.fit_transform(X)
+        
+        J, C, sum_constraint = self.get_hamiltonian(X, y)
+
+        assert J.shape[0] == J.shape[1], "Inconsistent hamiltonian size!"
+        assert J.shape[0] == C.shape[0], "Inconsistent hamiltonian size!"
+
+        self.set_model(J, C, sum_constraint)
+
+        sol, response = self.solve()
+
+        assert len(sol) == C.shape[0], "Inconsistent solution size!"
+
+        self.params = self.convert_sol_to_params(sol)
+
+        self.X_train = X
+        self.y_train = y
+
+        return response
+
+    def predict_raw(self, X: np.array):
+        """
+        Predict classes for X.
+
+        Parameters
+        ----------
+        X : {array-like, sparse matrix} of shape (n_samples, n_features)
+
+        Returns
+        -------
+        y : ndarray of shape (n_samples,)
+        The predicted classes.
+        """
+        n_records = X.shape[0]
+        X = self.fea_scaler.transform(X)        
+        X_tilde = np.concatenate((X, np.ones((n_records, 1))), axis=1)
+
+        y = np.einsum("i,ki->k", self.params, X_tilde)
+
+        return y
+
+    def predict(self, X: np.array):
+        """
+        Predict classes for X.
+
+        Parameters
+        ----------
+        X : {array-like, sparse matrix} of shape (n_samples, n_features)
+
+        Returns
+        -------
+        y : ndarray of shape (n_samples,)
+        The predicted classes.
+        """
+
+        y = self.predict_raw(X)
+        y = np.sign(y)
+
+        return y
+
+    def get_hamiltonian(
+        self,
+        X: np.array,
+        y: np.array,
+    ):
+        n_records = X.shape[0]
+        n_dims = X.shape[1]
+
+        J = np.zeros(shape=(1 + n_dims, 1 + n_dims), dtype=np.float32)
+        C = np.zeros(shape=(1 + n_dims,), dtype=np.float32)
+
+        X_tilde = np.concatenate((X, np.ones((n_records, 1))), axis=1)
+
+        J = self.lambda_coef * np.einsum(
+            "i,ik,il->kl", y**2, X_tilde, X_tilde
+        )
+
+        for k in range(n_dims):
+            J[k][k] += 0.5
+
+        C = -2.0 * self.lambda_coef * np.einsum("i,ik->k", y, X_tilde)
+
+        J = 0.5 * (J + J.transpose())
+        C = C.reshape((1 + n_dims, 1))
+
+        return J, C, 1.0
+
+    def convert_sol_to_params(self, sol):
+        return np.array(sol)
+
+
+class QSVMClassifierDual(ClassifierBase):
+    """An implementation of dual QSVM classifier that uses QCi's Dirac-3.
+
+    Parameters
+    ----------
+
+    relaxation_schedule: Relaxation schedule used by Dirac-3; default:
+    2.
+
     num_samples: Number of samples used by Dirac-3; default: 1.
+
+    solver_access: Solver access type: cloud or direct; default: cloud.
+
+    ip_addr: IP address of the device when direct access is used; default: None.
+    
+    port: Port number of the device when direct access is used; default: None.    
     
     upper_limit: Coefficient upper limit; a regularization parameter;
     default: 1.0.
-    
+
     gamma: Gaussian kernel parameter; default: 1.0.
-    
+
     eta: A penalty multiplier; default: 1.0.
-    
+
     zeta: A penalty multiplier; default: 1.0.
 
     Examples
@@ -54,8 +250,8 @@ class QSVMClassifier(ClassifierBase):
     ...     test_size=0.2,
     ...     random_state=42,
     ... )
-    >>> from eqc_models.ml.classifierqsvm import QSVMClassifier
-    >>> obj = QSVMClassifier(
+    >>> from eqc_models.ml.classifierqsvm import QSVMClassifierDual
+    >>> obj = QSVMClassifierDual(
     ...     relaxation_schedule=2,
     ...     num_samples=1,
     ...     upper_limit=1.0,
@@ -70,22 +266,30 @@ class QSVMClassifier(ClassifierBase):
     ...    obj = obj.fit(X_train, y_train)
     ...    y_train_prd = obj.predict(X_train)
     ...    y_test_prd = obj.predict(X_test)
-    
+
     """
-    
+
     def __init__(
         self,
         relaxation_schedule=2,
         num_samples=1,
+        solver_access="cloud",
+        ip_addr=None,
+        port=None,            
         upper_limit=1.0,
         gamma=1.0,
         eta=1.0,
         zeta=1.0,
-    ):        
-        super(QSVMClassifier).__init__()
+    ):
+        super(QSVMClassifierDual).__init__()
 
+        assert solver_access in ["cloud", "direct"]
+        
         self.relaxation_schedule = relaxation_schedule
         self.num_samples = num_samples
+        self.solver_access = solver_access
+        self.ip_addr = ip_addr
+        self.port = port        
         self.upper_limit = upper_limit
         self.gamma = gamma
         self.eta = eta
@@ -97,20 +301,20 @@ class QSVMClassifier(ClassifierBase):
     def fit(self, X, y):
         """
         Build a QSVM classifier from the training set (X, y).
-    
+
         Parameters
         ----------
         X : {array-like, sparse matrix} of shape (n_samples, n_features)
-        The training input samples. 
-    
+        The training input samples.
+
         y : array-like of shape (n_samples,)
         The target values.
-    
+
         Returns
         -------
         Response of Dirac-3 in JSON format.
         """
-                
+
         assert X.shape[0] == y.shape[0], "Inconsistent sizes!"
 
         assert set(y) == {-1, 1}, "Target values should be in {-1, 1}"
@@ -144,17 +348,17 @@ class QSVMClassifier(ClassifierBase):
     def predict(self, X: np.array):
         """
         Predict classes for X.
-        
+
         Parameters
         ----------
         X : {array-like, sparse matrix} of shape (n_samples, n_features)
-    
+
         Returns
         -------
         y : ndarray of shape (n_samples,)
         The predicted classes.
         """
-                
+
         assert self.X_train is not None, "Model not trained yet!"
         assert self.y_train is not None, "Model not trained yet!"
 

{eqc_models-0.11.0.data → eqc_models-0.12.0.data}/platlib/eqc_models/ml/clustering.py

@@ -140,14 +140,14 @@ class GraphClustering(ClusteringBase):
 
         self.set_model(J, C, sum_constraint)
 
-        sol = self.solve()
-
+        sol, response = self.solve()
+        
         assert len(sol) == C.shape[0], "Inconsistent solution size!"
         assert len(sol) == self.num_clusters * self.num_nodes
 
         self.labels = self.get_labels(sol)
 
-        return self
+        return response
 
     def fit_predict(self):
         """
@@ -300,14 +300,14 @@ class Clustering(ClusteringBase):
 
         self.set_model(J, C, sum_constraint)
 
-        sol = self.solve()
+        sol, response = self.solve()
 
         assert len(sol) == C.shape[0], "Inconsistent solution size!"
         assert len(sol) == self.num_clusters * num_items
 
         self.labels = self.get_labels(sol, num_items)
 
-        return self
+        return response
 
     def fit_predict(self, X: np.array):
         """

{eqc_models-0.11.0.data → eqc_models-0.12.0.data}/platlib/eqc_models/ml/clusteringbase.py

@@ -78,7 +78,7 @@ class ClusteringBase(QuadraticModel):
 
         print(response)
 
-        return sol
+        return sol, response
 
     def get_labels(self, sol):
         pass

{eqc_models-0.11.0.data → eqc_models-0.12.0.data}/platlib/eqc_models/ml/decomposition.py

@@ -21,17 +21,17 @@ from eqc_models.solvers.qciclient import (
 
 class DecompBase(QuadraticModel):
     """An Base class for decomposition algorithms.
-    
+
     Parameters
     ----------
-    
+
     relaxation_schedule: Relaxation schedule used by Dirac-3; default:
     2.
-    
+
     num_samples: Number of samples used by Dirac-3; default: 1.
 
     """
-    
+
     def __init__(
         self,
         relaxation_schedule=2,
@@ -163,8 +163,7 @@ class DecompBase(QuadraticModel):
 
 
 class PCA(DecompBase):
-    """
-    An implementation of Principal component analysis (PCA) that
+    """An implementation of Principal component analysis (PCA) that
     uses QCi's Dirac-3.
 
     Linear dimensionality reduction using Singular Value
@@ -173,18 +172,22 @@ class PCA(DecompBase):
 
     Parameters
     ----------
-    
+
     n_components: Number of components to keep; if n_components is not
     set all components are kept; default: None.
-    
+
     relaxation_schedule: Relaxation schedule used by Dirac-3; default:
     2.
-    
+
     num_samples: Number of samples used by Dirac-3; default: 1.
 
+    mode: Compute the largest or smallest principal components,
+    largest_components vs. smallest_components; default:
+    largest_components.
+
     Examples
     -----------
-    
+
     >>> from sklearn import datasets
     >>> iris = datasets.load_iris()
     >>> X = iris.data
@@ -202,18 +205,25 @@ class PCA(DecompBase):
     ...        num_samples=1,
     ...    )
     ...    X_pca = obj.fit_transform(X)
-    
+
     """
-    
+
     def __init__(
         self,
         n_components=None,
         relaxation_schedule=2,
         num_samples=1,
+        mode="largest_components",
     ):
         self.n_components = n_components
         self.relaxation_schedule = relaxation_schedule
         self.num_samples = num_samples
+
+        assert mode in ["largest_components", "smallest_components"], (
+            "Invalid value of mode <%s>" % mode
+        )
+
+        self.mode = mode
         self.X = None
         self.X_pca = None
 
@@ -224,12 +234,20 @@ class PCA(DecompBase):
         num_records = X.shape[0]
         num_features = X.shape[1]
 
-        J = -np.matmul(X.transpose(), X)
+        J = np.matmul(X.transpose(), X)
 
         assert J.shape[0] == num_features
         assert J.shape[1] == num_features
 
-        C = np.zeros((num_features, 1))
+        C = -np.sum(J, axis=1)
+
+        assert C.shape[0] == num_features
+
+        C = C.reshape((num_features, 1))
+
+        if self.mode == "largest_components":
+            J = -J
+            C = -C
 
         return J, C
 
@@ -239,12 +257,17 @@ class PCA(DecompBase):
         assert J.shape[0] == J.shape[1], "Inconsistent hamiltonian size!"
         assert J.shape[0] == C.shape[0], "Inconsistent hamiltonian size!"
 
+        sum_constraint = 0.5 * (1.0 + C.shape[0])
+        
         self._set_model(J, C, 1.0)
 
         sol, response = self._solve()
 
         assert len(sol) == C.shape[0], "Inconsistent solution size!"
 
+        sol = np.array(sol)
+        sol = 2.0 * sol - 1.0
+        
         fct = np.linalg.norm(sol)
         if fct > 0:
             fct = 1.0 / fct
@@ -282,7 +305,7 @@ class PCA(DecompBase):
 
         Returns
         -------
-        responses. 
+        responses.
         A dirct containing Dirac responses.
         """
 
@@ -310,7 +333,7 @@ class PCA(DecompBase):
             assert X.shape == self.X.shape, "Inconsistent size!"
 
             resp_hash["component_%d_response" % (i + 1)] = resp
-            
+
         self.X_pca = np.array(self.X_pca).transpose()
 
         assert self.X_pca.shape[0] == self.X.shape[0]

eqc_models-0.12.0.data/platlib/eqc_models/process/base.py

@@ -0,0 +1,18 @@
+import networkx as nx
+from eqc_models.base import ConstrainedPolynomialModel
+
+class ProcessModel:
+    """
+    Given a directed graph $G$ that describes a process flow, optimize the 
+    flow.
+
+    """
+    def __init__(self, G : nx.DiGraph):
+        self.G = G
+
+    @staticmethod
+    def process_constraints(G : nx.DiGraph):
+        """ Build process constraints from the graph """
+
+    def constraints(self) -> Tuple[np.ndarray]:
+        pass

eqc_models-0.12.0.data/platlib/eqc_models/process/mpc.py

@@ -0,0 +1,17 @@
+from eqc_models.process.base import ProcessModel
+
+class ModelPredictiveControl(ProcessModel):
+    """ Base class for implementing MPC optimization problems """
+
+    def __init__(self, G : nx.Graph, V : int=1, V_T:float=0.0):
+        self.processArgs(*args, **kwargs)
+        self.T = T
+        self.V_T = V_T
+        super(ModelPredictiveControl, self).__init__(G)
+        
+    def processArgs(self, *args, **kwargs):
+        """ Provide a method to capture arguments necessary for configuring an instance """
+        raise NotImplementedError("subclass must implement processArgs")
+
+    def constraints(self) -> Tuple[np.ndarray]:
+        pass

{eqc_models-0.11.0.data → eqc_models-0.12.0.data}/platlib/eqc_models/solvers/__init__.py

@@ -1,9 +1,5 @@
 # (C) Quantum Computing Inc., 2024.
-try:
-    from .eqcdirect import Dirac3DirectSolver, EqcDirectSolver
-except ImportError:
-    # eqc-direct is not available
-    Dirac3DirectSolver = None
+from .eqcdirect import Dirac3DirectSolver
 from .qciclient import (Dirac1CloudSolver, Dirac3CloudSolver, QciClientSolver,
                         Dirac3IntegerCloudSolver, Dirac3ContinuousCloudSolver)
 

eqc_models-0.12.0.data/platlib/eqc_models/solvers/eqcdirect.py

@@ -0,0 +1,71 @@
+# (C) Quantum Computing Inc., 2025.
+import logging
+import numpy as np
+from eqc_direct.client import EqcClient
+from eqc_models.base.base import ModelSolver
+from eqc_models.base.results import SolutionResults
+
+log = logging.getLogger(name=__name__)
+
+
+class Dirac3DirectSolver(ModelSolver):
+    ip_addr: str = "localhost"
+    port: str = "50051"
+    cert_file: str = None
+
+    def solve(
+        self,
+        model,
+        sum_constraint,
+        relaxation_schedule=1,
+        num_samples=1,
+        solution_precision=None,
+        mean_photon_number=None,
+        quantum_fluctuation_coefficient=None,
+    ):
+        polynomial = model.polynomial
+        coefficients = polynomial.coefficients
+        indices = polynomial.indices
+        num_variables = 0
+        for i in range(len(indices)):
+            if max(indices[i]) > num_variables:
+                num_variables = max(indices[i])
+
+        print("Num variables:", num_variables)
+
+        client = self.client
+        lock_id, startts, endts = client.wait_for_lock()
+
+        # assert lock["status_code"] == 0
+
+        # lock_id = lock["lock_id"]
+        response = None
+        try:
+            response = client.solve_sum_constrained(
+                coefficients,
+                indices,
+                num_variables,
+                num_samples=num_samples,
+                lock_id=lock_id,
+                relaxation_schedule=relaxation_schedule,
+                sum_constraint=sum_constraint,
+                solution_precision=solution_precision,
+                mean_photon_number=mean_photon_number,
+                quantum_fluctuation_coefficient=quantum_fluctuation_coefficient,
+            )
+        finally:
+            client.release_lock(lock_id=lock_id)
+
+        return response
+
+    def connect(self, ip_addr: str, port: str, cert_file: str = None):
+        self.ip_addr = ip_addr
+        self.port = port
+        self.cert_file = cert_file
+        client = self.client
+
+        return client.system_status()
+
+    @property
+    def client(self) -> EqcClient:
+        return EqcClient(self.ip_addr, self.port, cert_file=self.cert_file)

{eqc_models-0.11.0.data → eqc_models-0.12.0.data}/platlib/eqc_models/solvers/qciclient.py

@@ -519,7 +519,8 @@ class Dirac3CloudSolver(Dirac3Mixin, QciClientSolver):
             of photons that are present in a given quantum state.
             Modify this value to control the relaxation schedule more
             precisely than the four presets given in schedules 1
-            through 4. Allowed values are decimals between 0.1 and 2.
+            through 4. Allowed values are decimals between 0.000133333333333 and 
+            0.001.
         quantum_fluctuation_coefficient: int 
             an integer value which Sets the amount of loss introduced
             into the system for each loop during the measurement process.
@@ -622,7 +623,8 @@ class Dirac3IntegerCloudSolver(Dirac3Mixin, QciClientSolver):
             of photons that are present in a given quantum state.
             Modify this value to control the relaxation schedule more
             precisely than the four presets given in schedules 1
-            through 4. Allowed values are decimals between 0.1 and 2.
+            through 4. Allowed values are decimals between 0.000133333333333 
+            and 0.001.
         quantum_fluctuation_coefficient: int 
             an integer value which Sets the amount of loss introduced
             into the system for each loop during the measurement process.
@@ -716,7 +718,8 @@ class Dirac3ContinuousCloudSolver(Dirac3Mixin, QciClientSolver):
             of photons that are present in a given quantum state.
             Modify this value to control the relaxation schedule more
             precisely than the four presets given in schedules 1
-            through 4. Allowed values are decimals between 0.1 and 2.
+            through 4. Allowed values are decimals between 0.000133333333333 
+            and 0.001.
         quantum_fluctuation_coefficient: int 
             an integer value which Sets the amount of loss introduced
             into the system for each loop during the measurement process.

{eqc_models-0.11.0.dist-info → eqc_models-0.12.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: eqc-models
-Version: 0.11.0
+Version: 0.12.0
 Summary: Optimization and ML modeling package targeting EQC devices
 Author: Quantum Computing Inc.
 Author-email: support@quantumcomputinginc.com
@@ -16,8 +16,7 @@ Requires-Dist: pandas>=2.1.0
 Requires-Dist: scikit-learn>=1.2.1
 Requires-Dist: qci-client<5,>=4.3.0
 Requires-Dist: emucore-direct==1.0.6
-Provides-Extra: direct
-Requires-Dist: eqc-direct<2,>=1.3.0; extra == "direct"
+Requires-Dist: eqc-direct<3,>=2.0.1
 Provides-Extra: dev
 Requires-Dist: pytest<8,>=7.1.0; extra == "dev"
 Requires-Dist: pytest-cov; extra == "dev"