Merge pull request #12 from Learnware-LAMDA/dev_client

[FIX] fix bugs in LearnwareClient

examples/dataset_pfs_workflow/pfs/pfs_cross_transfer.py

@@ -85,7 +85,9 @@ def get_split_errs(algo):
                 split = train_xs.shape[0] - proportion_list[tmp]
 
                 model.fit(
-                    train_xs[split:,],
+                    train_xs[
+                        split:,
+                    ],
                     train_ys[split:],
                     eval_set=[(val_xs, val_ys)],
                     early_stopping_rounds=50,

learnware/client/learnware_client.py

@@ -341,11 +341,8 @@ class LearnwareClient:
 
             if load_model:
                 learnware_obj.instantiate_model()
-                pass
 
             return learnware_obj
-            pass
-        pass
 
     def system(self, command):
         retcd = os.system(command)
@@ -392,9 +389,9 @@ class LearnwareClient:
                     package_utils.filter_nonexist_pip_packages_file(requirements_path, requirements_path_filter)
 
                     if conda_env is not None:
-                        self.system(f"conda create --name {conda_env}")
+                        self.system(f"conda create -y --name {conda_env} python=3.8")
                         self.system(
-                            f"conda run --no-capture-output python3 -m pip install -r {requirements_path_filter}"
+                            f"conda run --name {conda_env} --no-capture-output python3 -m pip install -r {requirements_path_filter}"
                         )
                     else:
                         self.system(f"python3 -m pip install -r {requirements_path_filter}")

learnware/client/utils.py

@@ -61,12 +61,14 @@ def install_environment(zip_path, conda_env):
                     requirements_file=requirements_path, output_file=requirements_path_filter
                 )
                 logger.info(f"create empty conda env [{conda_env}]")
-                system_execute(args=["conda", "create", "--name", f"{conda_env}", "python=3.8"])
+                system_execute(args=["conda", "create", "-y", "--name", f"{conda_env}", "python=3.8"])
                 logger.info(f"install pip requirements for conda env [{conda_env}]")
                 system_execute(
                     args=[
                         "conda",
                         "run",
+                        "-n",
+                        f"{conda_env}",
                         "--no-capture-output",
                         "python3",
                         "-m",
@@ -80,4 +82,17 @@ def install_environment(zip_path, conda_env):
                 raise Exception("Environment.yaml or requirements.txt not found in the learnware zip file.")
 
     logger.info(f"install learnware package for conda env [{conda_env}]")
-    system_execute(args=["conda", "run", "--no-capture-output", "python3", "-m", "pip", "install", "learnware"])
+    system_execute(
+        args=[
+            "conda",
+            "run",
+            "-n",
+            f"{conda_env}",
+            "--no-capture-output",
+            "python3",
+            "-m",
+            "pip",
+            "install",
+            "learnware",
+        ]
+    )

tests/test_client/test_download.py

@@ -0,0 +1,68 @@
+import os
+import zipfile
+import numpy as np
+
+import learnware
+from learnware.learnware import get_learnware_from_dirpath
+from learnware.client import LearnwareClient
+from learnware.client.container import ModelEnvContainer, LearnwaresContainer
+from learnware.learnware.reuse import AveragingReuser
+
+
+def test_container(zip_paths):
+    semantic_specification = dict()
+    semantic_specification["Data"] = {"Type": "Class", "Values": ["Text"]}
+    semantic_specification["Task"] = {"Type": "Class", "Values": ["Ranking"]}
+    semantic_specification["Library"] = {"Type": "Class", "Values": ["Scikit-learn"]}
+    semantic_specification["Scenario"] = {"Type": "Tag", "Values": "Financial"}
+    semantic_specification["Name"] = {"Type": "String", "Values": "test"}
+    semantic_specification["Description"] = {"Type": "String", "Values": "test"}
+
+    learnware_list = []
+    for id, zip_path in enumerate(zip_paths):
+        dir_path = zip_path[:-4]
+        with zipfile.ZipFile(zip_path, "r") as z_file:
+            z_file.extractall(dir_path)
+
+        learnware = get_learnware_from_dirpath(f"test_id{id}", semantic_specification, dir_path)
+        learnware_list.append(learnware)
+
+    with LearnwaresContainer(learnware_list, zip_paths) as env_container:
+        learnware_list = env_container.get_learnware_list_with_container()
+        reuser = AveragingReuser(learnware_list, mode="vote_by_label")
+        input_array = np.random.random(size=(20, 13))
+        print(reuser.predict(input_array))
+
+        for idx, learnware in enumerate(learnware_list):
+            print(f"learnware_{idx}", learnware.predict(input_array))
+
+
+def test_load(zip_paths):
+    learnware_list = [client.load_learnware(file, load_model=False) for file in zip_paths]
+
+    with LearnwaresContainer(learnware_list, zip_paths) as env_container:
+        learnware_list = env_container.get_learnware_list_with_container()
+        reuser = AveragingReuser(learnware_list, mode="vote_by_label")
+        input_array = np.random.random(size=(20, 13))
+        print(reuser.predict(input_array))
+
+        for idx, learnware in enumerate(learnware_list):
+            print(f"learnware_{idx}", learnware.predict(input_array))
+
+
+if __name__ == "__main__":
+    email = "liujd@lamda.nju.edu.cn"
+    token = "f7e647146a314c6e8b4e2e1079c4bca4"
+
+    client = LearnwareClient()
+    client.login(email, token)
+
+    learnware_ids = ["00000084", "00000154", "00000155"]
+    zip_paths = ["1.zip", "2.zip", "3.zip"]
+    root = os.path.dirname(__file__)
+    for i in range(len(learnware_ids)):
+        zip_paths[i] = os.path.join(root, zip_paths[i])
+        client.download_learnware(learnware_ids[i], zip_paths[i])
+
+    test_container(zip_paths)
+    # test_load(zip_paths)

tests/test_client/test_reuse.py

@@ -0,0 +1,42 @@
+import zipfile
+import numpy as np
+
+from learnware.learnware import get_learnware_from_dirpath, Learnware
+from learnware.market import EasyMarket
+from learnware.client.container import ModelEnvContainer, LearnwaresContainer
+from learnware.learnware.reuse import AveragingReuser
+
+if __name__ == "__main__":
+    semantic_specification = dict()
+    semantic_specification["Data"] = {"Type": "Class", "Values": ["Text"]}
+    semantic_specification["Task"] = {"Type": "Class", "Values": ["Ranking"]}
+    semantic_specification["Library"] = {"Type": "Class", "Values": ["Scikit-learn"]}
+    semantic_specification["Scenario"] = {"Type": "Tag", "Values": "Financial"}
+    semantic_specification["Name"] = {"Type": "String", "Values": "test"}
+    semantic_specification["Description"] = {"Type": "String", "Values": "test"}
+
+    zip_paths = [
+        "/home/bixd/workspace/learnware/Learnware/tests/test_learnware_client/rf_tic.zip",
+        "/home/bixd/workspace/learnware/Learnware/tests/test_learnware_client/svc_tic.zip",
+    ]
+    dir_paths = [
+        "/home/bixd/workspace/learnware/Learnware/tests/test_learnware_client/rf_tic",
+        "/home/bixd/workspace/learnware/Learnware/tests/test_learnware_client/svc_tic",
+    ]
+
+    learnware_list = []
+    for id, (zip_path, dir_path) in enumerate(zip(zip_paths, dir_paths)):
+        with zipfile.ZipFile(zip_path, "r") as z_file:
+            z_file.extractall(dir_path)
+
+        learnware = get_learnware_from_dirpath(f"test_id{id}", semantic_specification, dir_path)
+        learnware_list.append(learnware)
+
+    with LearnwaresContainer(learnware_list, zip_paths) as env_container:
+
+        learnware_list = env_container.get_learnware_list_with_container()
+        reuser = AveragingReuser(learnware_list, mode="vote")
+        input_array = np.random.randint(0, 3, size=(20, 9))
+        print(reuser.predict(input_array).argmax(axis=1))
+        for id, ind_learner in enumerate(learnware_list):
+            print(f"learner_{id}", reuser.predict(input_array).argmax(axis=1))