Rename cli command

README.md

@@ -33,9 +33,9 @@ To run DeepView.Profile, you need:
 ### Installation from source
 ```bash
 git clone https://github.com/CentML/DeepView.Profile
-cd skyline
+cd DeepView.Profile
 poetry install
-poetry run skyline --help
+poetry run deepview --help
 ```
 
 ### Installation from PyPi
@@ -44,19 +44,19 @@ poetry run skyline --help
 
 Installing with [Poetry](https://python-poetry.org/)
 ```zsh
-poetry add skyline-profiler
-poetry run skyline --help
+poetry add deepview-profiler
+poetry run deepview --help
 ```
 
 Installing with [Pipenv](https://pipenv.pypa.io/en/latest/)
 ```zsh
-pipenv install skyline-profiler
-pipenv run skyline --help
+pipenv install deepview-profiler
+pipenv run deepview --help
 ```
 
 Installing with [Pip](https://packaging.python.org/en/latest/tutorials/installing-packages/#use-pip-for-installing)
 ```zsh
-python3 -m pip install skyline-profiler
+python3 -m pip install deepview-profiler
 python3 skyline
 ```
 
@@ -68,26 +68,26 @@ Once your entry point file is ready, there are two ways to profile interactive p
 
 ### Interactive Profiling
 ```zsh
-poetry run skyline interactive
+poetry run deepview interactive
 ```
 
 ### Standalone Profiling
 Standalone profiling is useful when you just want access to DeepView.Profile's profiling functionality. DeepView.Profile will save the profiling results (called a "report") into a [SQLite database file](https://www.sqlite.org/) that you can then query yourself. We describe the database schema for DeepView.Profile's run time and memory reports in the [Run Time Report Format](docs/run-time-report.md) and [Memory Report Format](docs/memory-report.md) pages respectively.
 
-To have DeepView.Profile perform run time profiling, you use the `skyline time`
+To have DeepView.Profile perform run time profiling, you use the `deepview time`
 subcommand. In addition to the entry point file, you also need to specify the
 file where you want DeepView.Profile to save the run time profiling report using the
 `--output` or `-o` flag.
 
 ```zsh
-poetry run skyline time entry_point.py --output my_output_file.sqlite
+poetry run deepview time entry_point.py --output my_output_file.sqlite
 ```
 
 Launching memory profiling is almost the same as launching run time profiling.
-You just need to use `skyline memory` instead of `skyline time`.
+You just need to use `deepview memory` instead of `deepview time`.
 
 ```zsh
-poetry run skyline memory entry_point.py --output my_output_file.sqlite
+poetry run deepview memory entry_point.py --output my_output_file.sqlite
 ```
 
 <h2 id="dev-setup">Development Environment Setup</h2>

docs/remote.md

@@ -24,18 +24,18 @@ To install the DeepView.Explore plugin on the server, take the following steps.
 
 ## Starting a Remote Profiling Session
 
-### Starting the Skyline Profiler
-DeepView.Profile needs to running on the server to enable the plugin. You can connect to the server via SSH and start DeepView.Profile by running the `skyline interactive` command as usual.
+### Starting the DeepView.Profiler
+DeepView.Profile needs to running on the server to enable the plugin. You can connect to the server via SSH and start DeepView.Profile by running the `deepview interactive` command as usual.
 
 ```zsh
-poetry run skyline interactive
+poetry run deepview interactive
 ```
 
 If you want to use a different port, you can use the `--port` flag to tell the profiler to listen on a different port.
 
 ```zsh
-poetry run skyline interactive --port portNumber
+poetry run deepview interactive --port portNumber
 ```
 
 ### Starting DeepView.Explore
-Launch VSCode and open DeepView.Explore by running the Skyline command in the command palette (Ctrl-Shift-P on Linux/Windows, ⌘-Shift-P on macOS). Select your project root and begin profiling.
+Launch VSCode and open DeepView.Explore by running the deepview command in the command palette (Ctrl-Shift-P on Linux/Windows, ⌘-Shift-P on macOS). Select your project root and begin profiling.

pyproject.toml

@@ -1,12 +1,12 @@
 [tool.poetry]
-name = "skyline-profiler"
+name = "deepview-profiler"
 version = "0.12.1"
 description = "Interactive performance profiling and debugging tool for PyTorch neural networks."
 authors = ["Geoffrey Yu <gxyu@cs.toronto.edu>"]
 maintainers = ["Akbar Nurlybayev <akbar@centml.ai>", "Yubo Gao <ybgao@centml.ai>"]
 license = "Apache-2.0"
 readme = "README.md"
-repository = "https://github.com/CentML/skyline"
+repository = "https://github.com/CentML/DeepView.Profile"
 keywords = ["pytorch", "neural networks", "debugger", "profiler"]
 classifiers = [
     "Development Status :: 4 - Beta",
@@ -22,7 +22,7 @@ packages = [
 include = [ "pyproject.toml" ]
 
 [tool.poetry.scripts]
-skyline = "skyline.skyline:main"
+deepview = "skyline.skyline:main"
 
 [tool.poetry.dependencies]
 python = "^3.7"

skyline/analysis/request_manager.py

@@ -142,7 +142,7 @@ def _handle_analysis_request(self, analysis_request, context):
                 self._send_analysis_error,
                 AnalysisError(
                     'An unexpected error occurred when analyzing your model. '
-                    'Please file a bug report and then restart Skyline.'
+                    'Please file a bug report and then restart Deepview.'
                 ),
                 context,
             )
@@ -190,7 +190,7 @@ def _send_habitat_response(self, habitat_resp, context):
             self._message_sender.send_habitat_response(habitat_resp, context)
         except:
             logger.exception(
-                'Exception occurred when sending a habitat response.')
+                'Exception occurred when sending a DeepView.Predict response.')
 
     def _send_energy_response(self, energy_resp, context):
         # Called from the main executor. Do not call directly!

skyline/analysis/runner.py

@@ -15,7 +15,7 @@ def analyze_project(project_root, entry_point, nvml):
     yield session.measure_throughput()
     torch.cuda.empty_cache()
 
-    print("analyze_project: running habitat_predict()")
+    print("analyze_project: running deepview_predict()")
     yield session.habitat_predict()
     torch.cuda.empty_cache()
 

skyline/analysis/session.py

@@ -31,7 +31,7 @@
 
     habitat_found = True
 except ImportError:
-    logger.debug("Habitat not found, GPU predictions not available")
+    logger.debug("Deepview.Predict not found, GPU predictions not available")
     habitat_found = False
 
 
@@ -211,10 +211,10 @@ def habitat_compute_threshold(self, runnable, context):
     def habitat_predict(self):
         resp = pm.HabitatResponse()
         if not habitat_found: 
-            logger.debug("Skipping Habitat predictions, returning empty response.")
+            logger.debug("Skipping deepview predictions, returning empty response.")
             return resp
 
-        print("habitat_predict: begin")
+        print("deepview_predict: begin")
         DEVICES = [
             habitat.Device.P100,
             habitat.Device.P4000,
@@ -244,14 +244,14 @@ def habitat_predict(self):
                 source_device = device
         pynvml.nvmlShutdown()
         if not source_device:
-            logger.debug("Skipping Habitat predictions, source not in list of supported GPUs.")
+            logger.debug("Skipping Deepview predictions, source not in list of supported GPUs.")
             src = pm.HabitatDevicePrediction()
             src.device_name = 'unavailable'
             src.runtime_ms = -1
             resp.predictions.append(src)
             return resp
 
-        print("habitat_predict: detected source device", source_device.name)
+        print("deepview_predict: detected source device", source_device.name)
 
         # get model
         model = self._model_provider()
@@ -284,7 +284,7 @@ def runnable():
         with tracker.track():
             iteration(*inputs)
 
-        print("habitat_predict: tracing on origin device")
+        print("deepview_predict: tracing on origin device")
         trace = tracker.get_tracked_trace()
 
         src = pm.HabitatDevicePrediction()
@@ -293,7 +293,7 @@ def runnable():
         resp.predictions.append(src)
 
         for device in DEVICES:
-            print("habitat_predict: predicting for", device)
+            print("deepview_predict: predicting for", device)
             predicted_trace = trace.to_device(device)
 
             pred = pm.HabitatDevicePrediction()

skyline/analysis/static.py

@@ -39,6 +39,6 @@ def visit_FunctionDef(self, node):
         if self.function_node is not None:
             # Return early if we've already found the provider
             return
-        if node.name != 'skyline_input_provider':
+        if node.name != 'deepview_input_provider':
             return
         self.function_node = node

skyline/commands/interactive.py

@@ -14,7 +14,7 @@
 def register_command(subparsers):
     parser = subparsers.add_parser(
         "interactive",
-        help="Start a new Skyline interactive profiling session.",
+        help="Start a new Deepview interactive profiling session.",
     )
     parser.add_argument(
         "--host",
@@ -63,7 +63,7 @@ def signal_handler(signal, frame):
     with SkylineServer(args.host, args.port) as server:
         _, port = server.listening_on
         logger.info(
-            "Skyline interactive profiling session started! "
+            "Deepview interactive profiling session started! "
             "Listening on port %d.",
             port,
         )

skyline/commands/measurements.py

@@ -19,7 +19,7 @@ def register_command(subparsers):
     )
     parser.add_argument(
         "entry_point",
-        help="The entry point file in this project that contains the Skyline "
+        help="The entry point file in this project that contains the Deepview "
              "provider functions.",
     )
     parser.add_argument(
@@ -79,7 +79,7 @@ def actual_main(args):
                 'samples_per_second',
                 'memory_usage_bytes',
             ])
-            project_root = os.cwd()
+            project_root = os.getcwd()
             for batch_size in args.batch_sizes:
                 for trial in range(args.trials):
                     session = AnalysisSession.new_from(

skyline/commands/memory.py

@@ -48,7 +48,7 @@ def actual_main(args):
         sys.exit(1)
 
     try:
-        project_root = os.cwd()
+        project_root = os.getcwd()
         session = AnalysisSession.new_from(
             project_root, args.entry_point)
         session.generate_memory_usage_report(

skyline/commands/prediction_models.py

@@ -15,11 +15,11 @@
 def register_command(subparsers):
     parser = subparsers.add_parser(
         "prediction-models",
-        help="Evaluate Skyline's prediction accuracy.",
+        help="Evaluate Deepview's prediction accuracy.",
     )
     parser.add_argument(
         "entry_point",
-        help="The entry point file in this project that contains the Skyline "
+        help="The entry point file in this project that contains the Deepview "
              "provider functions.",
     )
     parser.add_argument(
@@ -75,7 +75,7 @@ def actual_main(args):
                 'memory_usage_bytes_slope',
                 'memory_usage_bytes_bias',
             ])
-            project_root = os.cwd()
+            project_root = os.getcwd()
             for batch_size in args.batch_sizes:
                 session = AnalysisSession.new_from(
                     project_root, args.entry_point)

skyline/commands/time.py

@@ -18,7 +18,7 @@ def register_command(subparsers):
     )
     parser.add_argument(
         "entry_point",
-        help="The entry point file in this project that contains the Skyline "
+        help="The entry point file in this project that contains the Deepview "
              "provider functions.",
     )
     parser.add_argument(
@@ -49,7 +49,7 @@ def actual_main(args):
         sys.exit(1)
 
     try:
-        project_root = os.cwd()
+        project_root = os.getcwd()
         session = AnalysisSession.new_from(
             project_root, args.entry_point)
         session.generate_run_time_breakdown_report(

skyline/db/database.py

@@ -3,7 +3,7 @@
 import sqlite3
 
 BASE_DIR = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-DB_PATH = os.path.join(BASE_DIR, "skyline.sqlite3")
+DB_PATH = os.path.join(BASE_DIR, "deepview.sqlite3")
 
 
 class DatabaseInterface:

skyline/energy/measurer.py

@@ -20,7 +20,8 @@ def measurer_init(self):
             self.last_cpu = np.array(energy[0::2])
             self.last_dram = np.array(energy[1::2])
         except Exception as e:
-            print("Warning. Failed to get CPU energy")
+            print("Warning. Failed to get CPU energy. You need to set the right permissions for pyRAPL")
+            print("eg. $ sudo chmod -R a+r /sys/class/powercap/intel-rapl")
 
     def measurer_measure(self):
         # Get energy consumed so far (since last CPU reset)

skyline/error_printing.py

@@ -3,7 +3,7 @@
 
 def print_analysis_error(error, file=sys.stderr):
     print(
-        "Skyline encountered an error when profiling your model:",
+        "Deepview encountered an error when profiling your model:",
         file=file,
     )
     print("->", str(error), file=file)

skyline/initialization.py

@@ -50,9 +50,9 @@ def _validate_dependencies():
         return True
     except ImportError as ex:
         logger.error(
-            "Skyline could not find the '%s' module, which is a required "
+            "Deepview could not find the '%s' module, which is a required "
             "dependency. Please make sure all the required dependencies are "
-            "installed before launching Skyline. If you use a package "
+            "installed before launching Deepview. If you use a package "
             "manager, these dependencies will be automatically installed for "
             "you.",
             ex.name,
@@ -64,7 +64,7 @@ def _validate_gpu():
     import torch
     if not torch.cuda.is_available():
         logger.error(
-            "Skyline did not detect a GPU on this machine. Skyline only "
+            "Deepview did not detect a GPU on this machine. Deepview only "
             "profiles deep learning workloads on GPUs."
         )
         return False

skyline/io/connection_acceptor.py

@@ -40,7 +40,7 @@ def start(self):
         self._sentinel.start()
         self._acceptor.start()
         logger.debug(
-            "Skyline is listening for connections on (%s:%d).",
+            "Deepview is listening for connections on (%s:%d).",
             self._host,
             self._port,
         )
@@ -51,7 +51,7 @@ def stop(self):
         self._server_socket.close()
         self._sentinel.stop()
         logging.debug(
-            "Skyline has stopped listening for connections on (%s:%d).",
+            "Deepview has stopped listening for connections on (%s:%d).",
             self._host,
             self._port,
         )
@@ -80,4 +80,4 @@ def _accept_connections(self):
                 self._handler_function(socket, address)
         except:
             logging.exception(
-                "Skyline has unexpectedly stopped accepting connections.")
+                "Deepview has unexpectedly stopped accepting connections.")

skyline/server.py

@@ -47,7 +47,7 @@ def __exit__(self, exc_type, exc_value, traceback):
     def start(self):
         self._analysis_request_manager.start()
         self._connection_acceptor.start()
-        logger.debug("Skyline server has started.")
+        logger.debug("Deepview server has started.")
 
     def stop(self):
         def shutdown():
@@ -57,7 +57,7 @@ def shutdown():
         self._analysis_request_manager.stop()
         self._main_executor.submit(shutdown).result()
         self._main_executor.shutdown()
-        logger.debug("Skyline server has shut down.")
+        logger.debug("Deepview server has shut down.")
 
     @property
     def listening_on(self):
@@ -90,7 +90,8 @@ def _on_connection_closed(self, address):
 
     def _submit_work(self, func, *args, **kwargs):
         print("submit_work", func)
-        print("submit_work args:", args)
+        #print("submit_work args:", args)
+        logger.debug("submit_work args:", args)
         print("submit_work kwargs:", kwargs)
         # Do not call directly - called by another thread to submit work
         # onto the main executor