VR-4665: Require requirements param for Python blob (VertaAI#952)

* Require requirements param

* Update tests

* Update notebooks

* Update webinar notebooks

* Update everything else

Author: convoliution

client/verta/tests/test_versioning/test_environment.py

@@ -95,27 +95,28 @@ def test_parse_pip_freeze(self):
 
 class TestPython:
     def test_py_ver(self):
-        env = verta.environment.Python()
+        env = verta.environment.Python(requirements=[])
 
         assert env._msg.python.version.major == sys.version_info.major
         assert env._msg.python.version.minor == sys.version_info.minor
         assert env._msg.python.version.patch == sys.version_info.micro
 
     def test_env_vars(self):
         env_vars = os.environ.keys()
-        env = verta.environment.Python(env_vars=env_vars)
+        env = verta.environment.Python(requirements=[], env_vars=env_vars)
 
         assert env._msg.environment_variables
 
     def test_commit(self, commit):
-        env = verta.environment.Python()
+        env = verta.environment.Python(requirements=[])
 
         commit.update('env', env)
         commit.save(message="banana")
         assert commit.get('env')
 
     def test_reqs_from_env(self):
-        env = verta.environment.Python()
+        reqs = verta.environment.Python.read_pip_environment()
+        env = verta.environment.Python(requirements=reqs)
         assert env._msg.python.requirements
 
     def test_reqs(self, requirements_file):
@@ -130,13 +131,13 @@ def test_reqs_without_versions(self, requirements_file_without_versions):
 
     def test_constraints_from_file(self, requirements_file):
         reqs = verta.environment.Python.read_pip_file(requirements_file.name)
-        env = verta.environment.Python(constraints=reqs)
+        env = verta.environment.Python(requirements=[], constraints=reqs)
         assert env._msg.python.constraints
 
     def test_constraints_from_file_no_versions_error(self, requirements_file_without_versions):
         reqs = verta.environment.Python.read_pip_file(requirements_file_without_versions.name)
         with pytest.raises(ValueError):
-            verta.environment.Python(constraints=reqs)
+            verta.environment.Python(requirements=[], constraints=reqs)
 
     def test_inject_verta_cloudpickle(self):
         env = verta.environment.Python(requirements=[])
@@ -154,7 +155,7 @@ def test_reqs_no_unsupported_lines(self, requirements_file_with_unsupported_line
         assert requirements == {"verta", "cloudpickle"}
 
     def test_no_autocapture(self):
-        env_ver = verta.environment.Python(_autocapture=False)
+        env_ver = verta.environment.Python(requirements=[], _autocapture=False)
 
         # protobuf message is empty
         assert not json_format.MessageToDict(

client/verta/tests/test_versioning/test_repository.py

@@ -192,7 +192,8 @@ def test_revert(self, repository):
 
     def test_log_to_run(self, experiment_run, commit):
         blob1 = verta.dataset.Path(__file__)
-        blob2 = verta.environment.Python()
+        reqs = verta.environment.Python.read_pip_environment()
+        blob2 = verta.environment.Python(reqs)
         path1 = "data/1"
         path2 = "env/1"
 

client/verta/verta/environment/_python.py

@@ -22,9 +22,8 @@ class Python(_environment._Environment):
 
     Parameters
     ----------
-    requirements : list of str, optional
-        List of PyPI package names. If not provided, all packages currently installed through
-        pip will be captured.
+    requirements : list of str
+        List of PyPI package names.
     constraints : list of str, optional
         List of PyPI package names with version specifiers. If not provided, nothing will be
         captured.
@@ -38,21 +37,22 @@ class Python(_environment._Environment):
     .. code-block:: python
 
         from verta.environment import Python
-        env1 = Python(requirements=Python.read_pip_file("../requirements.txt"))
-        env2 = Python(
+        env1 = Python(requirements=Python.read_pip_environment())
+        env2 = Python(requirements=Python.read_pip_file("../requirements.txt"))
+        env3 = Python(
             requirements=["tensorflow"],
             env_vars=["CUDA_VISIBLE_DEVICES"],
         )
 
     """
-    def __init__(self, requirements=None, constraints=None, env_vars=None, _autocapture=True):
+    def __init__(self, requirements, constraints=None, env_vars=None, _autocapture=True):
         super(Python, self).__init__(env_vars, _autocapture)
 
         if _autocapture:
             self._capture_python_version()
-        if requirements is not None or _autocapture:
+        if requirements or _autocapture:
             self._capture_requirements(requirements)
-        if constraints is not None:
+        if constraints:
             self._capture_constraints(constraints)
 
     def __repr__(self):
@@ -105,7 +105,7 @@ def __repr__(self):
 
     @classmethod
     def _from_proto(cls, blob_msg):
-        obj = cls(_autocapture=False)
+        obj = cls(requirements=[], _autocapture=False)
         obj._msg.CopyFrom(blob_msg.environment)
 
         return obj
@@ -175,11 +175,8 @@ def _capture_python_version(self):
         self._msg.python.version.patch = sys.version_info.micro
 
     def _capture_requirements(self, requirements):
-        if requirements is None:
-            # TODO: support conda
-            req_specs = self.read_pip_environment()
-        elif (isinstance(requirements, list)
-              and all(isinstance(req, six.string_types) for req in requirements)):
+        if (isinstance(requirements, list)
+                and all(isinstance(req, six.string_types) for req in requirements)):
             req_specs = copy.copy(requirements)
             _pip_requirements_utils.process_requirements(req_specs)
         else:

client/workflows/examples/versioning/01 Simple Sentiment Classfier.ipynb

@@ -303,7 +303,7 @@
     "code_ver = Notebook()  # Notebook & git environment\n",
     "config_ver = Hyperparameters({'n_iter': 20})\n",
     "dataset_ver = S3(\"s3://{}/{}\".format(S3_BUCKET, S3_KEY))\n",
-    "env_ver = Python()  # pip environment and Python version"
+    "env_ver = Python(Python.read_pip_environment())  # pip environment and Python version"
    ]
   },
   {

client/workflows/examples/versioning/02 Reverting Training Data.ipynb

@@ -249,7 +249,7 @@
     "code_ver = Notebook()  # Notebook & git environment\n",
     "config_ver = Hyperparameters({'n_iter': 20})\n",
     "dataset_ver = S3(\"s3://{}/{}\".format(S3_BUCKET, S3_KEY))\n",
-    "env_ver = Python()  # pip environment and Python version"
+    "env_ver = Python(Python.read_pip_environment())  # pip environment and Python version"
    ]
   },
   {

client/workflows/examples/versioning/03 Merging Multilingual Support.ipynb

@@ -309,7 +309,7 @@
     "    \"s3://{}/{}\".format(S3_BUCKET, EN_S3_KEY),\n",
     "    \"s3://{}/{}\".format(S3_BUCKET, DE_S3_KEY),\n",
     "])\n",
-    "env_ver = Python()  # pip environment and Python version"
+    "env_ver = Python(Python.read_pip_environment())  # pip environment and Python version"
    ]
   },
   {

demos/webinar-2020-4-1/census-s3-oss-versioning.ipynb

@@ -257,7 +257,7 @@
     "    config_ver = Hyperparameters(hyperparams)\n",
     "    dataset_ver_train = S3(\"s3://{}/{}\".format(\"verta-starter\", \"census-train.csv\"))\n",
     "    dataset_ver_test = S3(\"s3://{}/{}\".format(\"verta-starter\", \"census-test.csv\"))\n",
-    "    env_ver = Python()\n",
+    "    env_ver = Python(Python.read_pip_environment())\n",
     "\n",
     "    commit.update(\"notebooks/census-s3-example\", code_ver)\n",
     "    commit.update(\"config/hyperparams\", config_ver)\n",

demos/webinar-2020-5-6/02-mdb_versioned/01-train/01 Basic NLP.ipynb

@@ -303,7 +303,7 @@
     "code_ver = Notebook()  # Notebook & git environment\n",
     "config_ver = Hyperparameters({'n_iter': 20})\n",
     "dataset_ver = S3(\"s3://{}/{}\".format(S3_BUCKET, S3_KEY))\n",
-    "env_ver = Python()  # pip environment and Python version"
+    "env_ver = Python(Python.read_pip_environment())  # pip environment and Python version"
    ]
   },
   {

demos/webinar-2020-5-6/02-mdb_versioned/01-train/02 Positive Data NLP.ipynb

@@ -249,7 +249,7 @@
     "code_ver = Notebook()  # Notebook & git environment\n",
     "config_ver = Hyperparameters({'n_iter': 20})\n",
     "dataset_ver = S3(\"s3://{}/{}\".format(S3_BUCKET, S3_KEY))\n",
-    "env_ver = Python()  # pip environment and Python version"
+    "env_ver = Python(Python.read_pip_environment())  # pip environment and Python version"
    ]
   },
   {

demos/webinar-2020-5-6/02-mdb_versioned/01-train/03 Multilingual NLP.ipynb

@@ -309,7 +309,7 @@
     "    \"s3://{}/{}\".format(S3_BUCKET, EN_S3_KEY),\n",
     "    \"s3://{}/{}\".format(S3_BUCKET, DE_S3_KEY),\n",
     "])\n",
-    "env_ver = Python()  # pip environment and Python version"
+    "env_ver = Python(Python.read_pip_environment())  # pip environment and Python version"
    ]
   },
   {