Merge pull request #19 from LAMDA-NJU/check_model_rkme_dim

[MNT] check model input with rkme dimension

learnware/market/easy.py

@@ -7,6 +7,7 @@ import numpy as np
 import pandas as pd
 from cvxopt import solvers, matrix
 from typing import Tuple, Any, List, Union, Dict
+import traceback
 
 from .base import BaseMarket, BaseUserInfo
 from .database_ops import DatabaseOperations
@@ -88,11 +89,15 @@ class EasyMarket(BaseMarket):
             - The NOPREDICTION_LEARNWARE denotes the learnware pass the check but cannot make prediction due to some env dependency
             - The NOPREDICTION_LEARNWARE denotes the leanrware pass the check and can make prediction
         """
+
+        semantic_spec = learnware.get_specification().get_semantic_spec()
+
         try:
             # check model instantiation
             learnware.instantiate_model()
 
         except Exception as e:
+            traceback.print_exc()
             logger.warning(f"The learnware [{learnware.id}] is instantiated failed! Due to {e}")
             return cls.NONUSABLE_LEARNWARE
 
@@ -100,7 +105,21 @@ class EasyMarket(BaseMarket):
             learnware_model = learnware.get_model()
 
             # check input shape
-            inputs = np.random.randn(10, *learnware_model.input_shape)
+            if semantic_spec['Data']['Values'][0] == 'Table':
+                input_shape = (semantic_spec['Input']['Dimension'], )
+            else:
+                input_shape = learnware_model.input_shape
+                pass
+
+            # check rkme dimension
+            stat_spec = learnware.get_specification().get_stat_spec_by_name("RKMEStatSpecification")
+            if stat_spec is not None:
+                if stat_spec.get_z().shape[1:] != input_shape:
+                    logger.warning(f"The learnware [{learnware.id}] input dimension mismatch with stat specification")
+                    return cls.NONUSABLE_LEARNWARE
+                pass
+
+            inputs = np.random.randn(10, *input_shape)
             outputs = learnware.predict(inputs)
 
             # check output type
@@ -111,11 +130,23 @@ class EasyMarket(BaseMarket):
                 return cls.NONUSABLE_LEARNWARE
 
             # check output shape
-            if outputs.shape[1:] != learnware_model.output_shape:
-                logger.warning(f"The learnware [{learnware.id}] input and output dimention is error")
-                return cls.NONUSABLE_LEARNWARE
+            if outputs.ndim == 1:
+                outputs = outputs.reshape(-1, 1)
+                pass
+            
+            if semantic_spec['Task']['Values'][0] in ('Classification', 'Regression', 'Feature Extraction'):
+                output_dim = int(semantic_spec['Output']['Dimension'])
+                if outputs[0].shape[0] != output_dim:
+                    logger.warning(f"The learnware [{learnware.id}] input and output dimention is error")
+                    return cls.NONUSABLE_LEARNWARE
+                pass
+            else:
+                if outputs.shape[1:] != learnware_model.output_shape:
+                    logger.warning(f"The learnware [{learnware.id}] input and output dimention is error")
+                    return cls.NONUSABLE_LEARNWARE
 
         except Exception as e:
+            logger.exception
             logger.warning(f"The learnware [{learnware.id}] prediction is not avaliable! Due to {repr(e)}")
             return cls.NONUSABLE_LEARNWARE