diff --git a/CHANGELOG.md b/CHANGELOG.md index f7f5241..4f5057c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -3,6 +3,14 @@ Observes [Semantic Versioning](https://semver.org/spec/v2.0.0.html) standard and [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) convention. +## [0.2.11] - 2023-01-04 + ++ Fix - docstring typo ++ Fix - `dj.config()` setup moved to `tutorial_pipeline.py` instead of `__init__.py` ++ Bugfix - Resolved AttributeError from the latest update of the YAML dependency ++ Update - Flowchart images to increase consistency with other DataJoint Elements ++ Update - Elements installed directly from GitHub instead of PyPI + ## [0.2.10] - 2023-11-20 + Fix - Revert fixing of networkx version in setup diff --git a/element_deeplabcut/__init__.py b/element_deeplabcut/__init__.py index 7f2535c..8b13789 100644 --- a/element_deeplabcut/__init__.py +++ b/element_deeplabcut/__init__.py @@ -1,21 +1 @@ -import os -import datajoint as dj -if "custom" not in dj.config: - dj.config["custom"] = {} - -# overwrite dj.config['custom'] values with environment variables if available - -dj.config["custom"]["database.prefix"] = os.getenv( - "DATABASE_PREFIX", dj.config["custom"].get("database.prefix", "") -) - -dj.config["custom"]["dlc_root_data_dir"] = os.getenv( - "DLC_ROOT_DATA_DIR", dj.config["custom"].get("dlc_root_data_dir", "") -) - -dj.config["custom"]["dlc_processed_data_dir"] = os.getenv( - "DLC_PROCESSED_DATA_DIR", dj.config["custom"].get("dlc_processed_data_dir", "") -) - -db_prefix = dj.config["custom"].get("database.prefix", "") diff --git a/element_deeplabcut/model.py b/element_deeplabcut/model.py index 1aa3902..8a6e47f 100644 --- a/element_deeplabcut/model.py +++ b/element_deeplabcut/model.py @@ -7,7 +7,7 @@ import os import cv2 import csv -import yaml +from ruamel.yaml import YAML import inspect import importlib import numpy as np @@ -239,8 +239,9 @@ def extract_new_body_parts(cls, dlc_config: dict, verbose: bool = True): ".yaml", ), f"dlc_config is neither dict nor filepath\n Check: {dlc_config_fp}" if dlc_config_fp.suffix in (".yml", ".yaml"): + yaml = YAML(typ="safe", pure=True) with open(dlc_config_fp, "rb") as f: - dlc_config = yaml.safe_load(f) + dlc_config = yaml.load(f) # -- Check and insert new BodyPart -- assert "bodyparts" in dlc_config, f"Found no bodyparts section in {dlc_config}" tracked_body_parts = cls.fetch("body_part") @@ -307,7 +308,7 @@ class Model(dj.Manual): config_template (longblob): Dictionary of the config for analyze_videos(). project_path ( varchar(255) ): DLC's project_path in config relative to root. model_prefix ( varchar(32) ): Optional. Prefix for model files. - model_description ( varchar(1000) ): Optional. User-entered description. + model_description ( varchar(300) ): Optional. User-entered description. TrainingParamSet (foreign key): Optional. Training parameters primary key. Note: @@ -381,8 +382,9 @@ def insert_new_model( "dlc_config is not a filepath" + f"\n Check: {dlc_config_fp}" ) if dlc_config_fp.suffix in (".yml", ".yaml"): + yaml = YAML(typ="safe", pure=True) with open(dlc_config_fp, "rb") as f: - dlc_config = yaml.safe_load(f) + dlc_config = yaml.load(f) if isinstance(params, dict): dlc_config.update(params) diff --git a/element_deeplabcut/readers/dlc_reader.py b/element_deeplabcut/readers/dlc_reader.py index a7f6a32..d726454 100644 --- a/element_deeplabcut/readers/dlc_reader.py +++ b/element_deeplabcut/readers/dlc_reader.py @@ -4,7 +4,7 @@ import pandas as pd from pathlib import Path import pickle -import ruamel.yaml as yaml +from ruamel.yaml import YAML from element_interface.utils import find_root_directory, dict_to_uuid from .. import model from ..model import get_dlc_root_data_dir @@ -145,7 +145,8 @@ def yml(self): """json-structured config.yaml file contents""" if self._yml is None: with open(self.yml_path, "rb") as f: - self._yml = yaml.safe_load(f) + yaml = YAML(typ="safe", pure=True) + self._yml = yaml.load(f) return self._yml @property diff --git a/element_deeplabcut/version.py b/element_deeplabcut/version.py index 1719e56..1ce9d0b 100644 --- a/element_deeplabcut/version.py +++ b/element_deeplabcut/version.py @@ -1,4 +1,4 @@ """ Package metadata """ -__version__ = "0.2.10" +__version__ = "0.2.11" diff --git a/images/flowchart.drawio b/images/flowchart.drawio index 450f89f..2bb0137 100644 --- a/images/flowchart.drawio +++ b/images/flowchart.drawio @@ -1 +1,81 @@ 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 \ No newline at end of file + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/images/flowchart.svg b/images/flowchart.svg index ed3f22c..4b5c726 100644 --- a/images/flowchart.svg +++ b/images/flowchart.svg @@ -1,3 +1,4 @@ + -
Synchronize data modalities & exploratory analysis
Synchronize data...
Visualize



 
Visualize...

 Export & publish

 
 
Export & publish...
Create project &
label training data 
in DeepLabCut
Create project &...
Enter metadata
into pipeline
Enter metadata...
Run model
training and 
evaluation
Run model...
Film new videos &
import metadata
Film new videos &...
Pair videos with
models & run
pose estimation
Pair videos with...
Text is not SVG - cannot display
\ No newline at end of file +
Synchronize data modalities & exploratory analysis
Synchronize data...
Visualize



 
Visualize...

 Export & publish

 
 
Export & publish...
Create project &
label training data 
in DeepLabCut
Create project &...
Enter metadata
into pipeline
Enter metadata...
Run model
training and 
evaluation
Run model...
Film new videos &
import metadata
Film new videos &...
Pair videos with
models & run
pose estimation
Pair videos with...
Text is not SVG - cannot display
\ No newline at end of file diff --git a/notebooks/tutorial.ipynb b/notebooks/tutorial.ipynb index b63193c..cf7840a 100644 --- a/notebooks/tutorial.ipynb +++ b/notebooks/tutorial.ipynb @@ -11,15 +11,14 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### **Open-source Data Pipeline for Markerless Pose Estimation in Neurophysiology**\n" + "#### **Open-source Data Pipeline for Markerless Pose Estimation in Neurophysiology**\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "This tutorial aims to provide a comprehensive understanding of the open-source data pipeline by `Element-DeepLabCut`." + "Welcome to the tutorial for the DataJoint Element for pose estimation analysis. This tutorial aims to provide a comprehensive understanding of the open-source data pipeline by `element-deeplabcut`." ] }, { @@ -33,7 +32,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The package is designed to integrate the **model training** and **pose estimation analyses** into a data pipeline and streamline model and video management using DataJoint. " + "The package is designed to seamlessly integrate the **model training** and **pose estimation analyses** into a data pipeline and streamline model and video management using DataJoint. " ] }, { @@ -54,7 +53,23 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "#### **Key Components and Objectives**\n" + "### Prerequisites\n", + "\n", + "Please see the [datajoint tutorials GitHub\n", + "repository](https://github.com/datajoint/datajoint-tutorials/tree/main) before\n", + "proceeding.\n", + "\n", + "A basic understanding of the following DataJoint concepts will be beneficial to your\n", + "understanding of this tutorial: \n", + "1. The `Imported` and `Computed` tables types in `datajoint-python`.\n", + "2. The functionality of the `.populate()` method. " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### **Tutorial Overview**" ] }, { @@ -62,17 +77,12 @@ "metadata": {}, "source": [ "\n", - "**- Setup**\n", - "\n", - "**- Designing the DataJoint Pipeline**\n", - "\n", - "**- Step 1: Register an Existing Model in the DataJoint Pipeline**\n", - "\n", - "**- Step 2: Insert Example Data into Subject and Session tables**\n", - "\n", - "**- Step 3: Run the DeepLabCut Inference Task**\n", - "\n", - "**- Step 4: Visualize the Results**" + "+ Setup\n", + "+ *Activate* the DataJoint pipeline\n", + "+ *Register* an existing model in the DataJoint pipeline\n", + "+ *Insert* example data into subject and session tables\n", + "+ Run the inference task\n", + "+ Visualize the results" ] }, { @@ -86,13 +96,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "This tutorial examines the **behavior of a freely moving mouse** in an open-field environment.\n", - "\n", - "The goal is to extract pose estimations of the animal's **head and tail base** from video footage. \n", + "This tutorial examines the behavior of a freely moving mouse in an open-field environment. The goal is to extract pose estimations of the animal's head and tail base from video footage. The resulting x and y coordinates offer valuable insights into the animal's movements, postures, and interactions within the environment. \n", "\n", - "The **resulting x and y coordinates** offer valuable insights into the **animal's movements, postures, and interactions** within the environment. \n", - "\n", - "The results of this Element example can be **combined with other modalities** to create a complete customizable data pipeline for your specific lab or study. For instance, you can combine `element-deeplabcut` with `element-calcium-imaging` and `element-array-ephys` to characterize the neural activity.\n", + "The results of this Element example can be combined with **other modalities** to create a complete customizable data pipeline for your specific lab or study. For instance, you can combine `element-deeplabcut` with `element-calcium-imaging` and `element-array-ephys` to characterize the neural activity along with markless pose estimation during behavior.\n", "\n", "#### Steps to Run the Element-DeepLabCut\n", "\n", @@ -143,7 +149,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "This codespace provides a local database private to you for experimentation. Let's connect to the database server:" + "If the tutorial is run in Codespaces, a private, local database server is created and\n", + "made available for you. This is where we will insert and store our processed results.\n", + "Let's connect to the database server." ] }, { @@ -155,8 +163,8 @@ "name": "stderr", "output_type": "stream", "text": [ - "[2023-10-19 19:26:24,608][INFO]: Connecting root@fakeservices.datajoint.io:3306\n", - "[2023-10-19 19:26:24,616][INFO]: Connected root@fakeservices.datajoint.io:3306\n" + "[2023-12-22 21:34:03,116][INFO]: Connecting root@fakeservices.datajoint.io:3306\n", + "[2023-12-22 21:34:03,194][INFO]: Connected root@fakeservices.datajoint.io:3306\n" ] }, { @@ -178,14 +186,15 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### **Design the DataJoint Pipeline**" + "### **Activate the DataJoint pipeline**" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "This tutorial presumes that the `element-deeplabcut` has been pre-configured and instantiated, with the database linked downstream to pre-existing `subject` and `session` tables. \n", + "This tutorial presumes that the `element-deeplabcut` has been pre-configured and instantiated, with the database linked downstream to pre-existing `subject` and `session` tables. Please refer to the\n", + "[`tutorial_pipeline.py`](./tutorial_pipeline.py) for the source code.\n", "\n", "Now, we will proceed to import the essential schemas required to construct this data pipeline, with particular attention to the primary components: `train` and `model`." ] @@ -199,7 +208,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "[2023-10-19 19:26:26,055][WARNING]: lab.Project and related tables will be removed in a future version of Element Lab. Please use the project schema.\n" + "[2023-12-22 21:34:04,823][WARNING]: lab.Project and related tables will be removed in a future version of Element Lab. Please use the project schema.\n" ] } ], @@ -211,7 +220,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We can represent a diagram of some of the upstream and downstream dependencies connected to these `model` and `train` schemas:" + "We can represent the tables in the `model` and `train` schemas as well as some of the upstream dependencies to `session` and `subject` schemas as a diagram." ] }, { @@ -222,768 +231,552 @@ { "data": { "image/svg+xml": [ - "\n", - "\n", + "\n", + "\n", "%3\n", - "\n", - "\n", + "\n", + "\n", "\n", - "subject.Line\n", - "\n", - "\n", - "subject.Line\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject.Line\n", - "\n", - "\n", - "subject.Subject.Line\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "subject.Line->subject.Subject.Line\n", - "\n", - "\n", - "\n", - "\n", - "subject.Line.Allele\n", - "\n", - "\n", - "subject.Line.Allele\n", + "session.Session.Attribute\n", + "\n", + "\n", + "session.Session.Attribute\n", "\n", "\n", "\n", - "\n", - "\n", - "subject.Line->subject.Line.Allele\n", - "\n", - "\n", - "\n", + "\n", "\n", - "subject.Subject.Source\n", - "\n", - "\n", - "subject.Subject.Source\n", + "subject.Subject.Protocol\n", + "\n", + "\n", + "subject.Subject.Protocol\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "lab.Project\n", - "\n", - "\n", - "lab.Project\n", + "subject.Subject.Strain\n", + "\n", + "\n", + "subject.Subject.Strain\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "session.ProjectSession\n", - "\n", - "\n", - "session.ProjectSession\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "lab.Project->session.ProjectSession\n", - "\n", - "\n", - "\n", - "\n", - "lab.ProjectPublication\n", - "\n", - "\n", - "lab.ProjectPublication\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "lab.Project->lab.ProjectPublication\n", - "\n", - "\n", - "\n", - "\n", - "lab.ProjectSourceCode\n", - "\n", - "\n", - "lab.ProjectSourceCode\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "lab.Project->lab.ProjectSourceCode\n", - "\n", - "\n", - "\n", - "\n", - "lab.ProjectKeywords\n", - "\n", - "\n", - "lab.ProjectKeywords\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "lab.Project->lab.ProjectKeywords\n", - "\n", - "\n", - "\n", - "\n", - "lab.ProjectUser\n", - "\n", - "\n", - "lab.ProjectUser\n", + "subject.Subject.Line\n", + "\n", + "\n", + "subject.Subject.Line\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Project->lab.ProjectUser\n", - "\n", - "\n", - "\n", + "\n", "\n", - "model.Model.BodyPart\n", - "\n", - "\n", - "model.Model.BodyPart\n", + "subject.Subject.Species\n", + "\n", + "\n", + "subject.Subject.Species\n", "\n", "\n", "\n", - "\n", - "\n", - "model.PoseEstimation.BodyPartPosition\n", - "\n", - "\n", - "model.PoseEstimation.BodyPartPosition\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "model.Model.BodyPart->model.PoseEstimation.BodyPartPosition\n", - "\n", - "\n", - "\n", + "\n", "\n", - "model.PoseEstimationTask\n", - "\n", - "\n", - "model.PoseEstimationTask\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "model.PoseEstimation\n", - "\n", - "\n", - "model.PoseEstimation\n", + "session.ProjectSession\n", + "\n", + "\n", + "session.ProjectSession\n", "\n", "\n", "\n", - "\n", - "\n", - "model.PoseEstimationTask->model.PoseEstimation\n", - "\n", - "\n", - "\n", + "\n", "\n", - "session.SessionDirectory\n", - "\n", - "\n", - "session.SessionDirectory\n", + "subject.Strain\n", + "\n", + "\n", + "subject.Strain\n", "\n", "\n", "\n", - "\n", + "\n", + "\n", + "subject.Strain->subject.Subject.Strain\n", + "\n", + "\n", + "\n", "\n", - "lab.Lab.Organization\n", - "\n", - "\n", - "lab.Lab.Organization\n", + "model.PoseEstimation\n", + "\n", + "\n", + "model.PoseEstimation\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "lab.LabMembership\n", - "\n", - "\n", - "lab.LabMembership\n", + "model.PoseEstimation.BodyPartPosition\n", + "\n", + "\n", + "model.PoseEstimation.BodyPartPosition\n", "\n", "\n", "\n", - "\n", + "\n", + "\n", + "model.PoseEstimation->model.PoseEstimation.BodyPartPosition\n", + "\n", + "\n", + "\n", "\n", - "session.Session.Attribute\n", - "\n", - "\n", - "session.Session.Attribute\n", + "subject.Species\n", + "\n", + "\n", + "subject.Species\n", "\n", "\n", "\n", - "\n", + "\n", + "\n", + "subject.Species->subject.Subject.Species\n", + "\n", + "\n", + "\n", "\n", - "lab.User\n", - "\n", - "\n", - "lab.User\n", + "subject.Subject.Source\n", + "\n", + "\n", + "subject.Subject.Source\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.User->lab.LabMembership\n", - "\n", - "\n", - "\n", - "\n", - "session.SessionExperimenter\n", - "\n", - "\n", - "session.SessionExperimenter\n", + "\n", + "\n", + "model.Model.BodyPart\n", + "\n", + "\n", + "model.Model.BodyPart\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.User->session.SessionExperimenter\n", - "\n", + "\n", + "\n", + "model.Model.BodyPart->model.PoseEstimation.BodyPartPosition\n", + "\n", "\n", - "\n", - "\n", - "subject.Subject.User\n", - "\n", - "\n", - "subject.Subject.User\n", + "\n", + "\n", + "subject.SubjectCull\n", + "\n", + "\n", + "subject.SubjectCull\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.User->subject.Subject.User\n", - "\n", - "\n", - "\n", - "\n", - "lab.User->lab.ProjectUser\n", - "\n", + "\n", + "\n", + "model.ModelEvaluation\n", + "\n", + "\n", + "model.ModelEvaluation\n", + "\n", "\n", - "\n", - "\n", - "model.PoseEstimation->model.PoseEstimation.BodyPartPosition\n", - "\n", "\n", "\n", - "\n", + "\n", "model.VideoRecording.File\n", - "\n", - "\n", - "model.VideoRecording.File\n", + "\n", + "\n", + "model.VideoRecording.File\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Device\n", - "\n", - "\n", - "lab.Device\n", + "\n", + "\n", + "model.Model\n", + "\n", + "\n", + "model.Model\n", "\n", "\n", "\n", - "\n", - "\n", - "model.VideoRecording\n", - "\n", - "\n", - "model.VideoRecording\n", + "\n", + "\n", + "model.Model->model.Model.BodyPart\n", + "\n", + "\n", + "\n", + "\n", + "model.Model->model.ModelEvaluation\n", + "\n", + "\n", + "\n", + "\n", + "model.PoseEstimationTask\n", + "\n", + "\n", + "model.PoseEstimationTask\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Device->model.VideoRecording\n", - "\n", + "\n", + "\n", + "model.Model->model.PoseEstimationTask\n", + "\n", "\n", "\n", - "\n", + "\n", "train.TrainingParamSet\n", - "\n", - "\n", - "train.TrainingParamSet\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "model.Model\n", - "\n", - "\n", - "model.Model\n", + "\n", + "\n", + "train.TrainingParamSet\n", "\n", "\n", "\n", "\n", - "\n", + "\n", "train.TrainingParamSet->model.Model\n", - "\n", + "\n", "\n", "\n", - "\n", + "\n", "train.TrainingTask\n", - "\n", - "\n", - "train.TrainingTask\n", + "\n", + "\n", + "train.TrainingTask\n", "\n", "\n", "\n", "\n", - "\n", + "\n", "train.TrainingParamSet->train.TrainingTask\n", - "\n", + "\n", "\n", - "\n", - "\n", - "subject.Subject.Protocol\n", - "\n", - "\n", - "subject.Subject.Protocol\n", - "\n", - "\n", - "\n", - "\n", + "\n", "\n", - "session.Session\n", - "\n", - "\n", - "session.Session\n", + "subject.Line\n", + "\n", + "\n", + "subject.Line\n", "\n", "\n", "\n", - "\n", - "\n", - "session.Session->session.ProjectSession\n", - "\n", + "\n", + "\n", + "subject.Line->subject.Subject.Line\n", + "\n", "\n", - "\n", - "\n", - "session.Session->session.SessionDirectory\n", - "\n", + "\n", + "\n", + "subject.Line.Allele\n", + "\n", + "\n", + "subject.Line.Allele\n", + "\n", "\n", - "\n", - "\n", - "session.Session->session.Session.Attribute\n", - "\n", "\n", - "\n", - "\n", - "session.Session->session.SessionExperimenter\n", - "\n", + "\n", + "\n", + "subject.Line->subject.Line.Allele\n", + "\n", "\n", - "\n", - "\n", - "session.SessionNote\n", - "\n", - "\n", - "session.SessionNote\n", + "\n", + "\n", + "train.ModelTraining\n", + "\n", + "\n", + "train.ModelTraining\n", "\n", "\n", "\n", - "\n", - "\n", - "session.Session->session.SessionNote\n", - "\n", - "\n", - "\n", - "\n", - "session.Session->model.VideoRecording\n", - "\n", - "\n", "\n", - "\n", + "\n", "subject.Subject\n", - "\n", - "\n", - "subject.Subject\n", + "\n", + "\n", + "subject.Subject\n", "\n", "\n", "\n", - "\n", - "\n", - "subject.Subject->subject.Subject.Source\n", - "\n", - "\n", "\n", - "\n", + "\n", "subject.Subject->subject.Subject.Protocol\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject->session.Session\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject->subject.Subject.Line\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject.Lab\n", - "\n", - "\n", - "subject.Subject.Lab\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject->subject.Subject.Lab\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject->subject.Subject.User\n", - "\n", - "\n", - "\n", - "\n", - "subject.Subject.Strain\n", - "\n", - "\n", - "subject.Subject.Strain\n", - "\n", - "\n", + "\n", "\n", "\n", - "\n", + "\n", "subject.Subject->subject.Subject.Strain\n", - "\n", + "\n", "\n", - "\n", - "\n", - "subject.SubjectDeath\n", - "\n", - "\n", - "subject.SubjectDeath\n", - "\n", + "\n", + "\n", + "subject.Subject->subject.Subject.Line\n", + "\n", "\n", + "\n", + "\n", + "subject.Subject->subject.Subject.Species\n", + "\n", "\n", - "\n", - "\n", - "subject.Subject->subject.SubjectDeath\n", - "\n", + "\n", + "\n", + "subject.Subject->subject.Subject.Source\n", + "\n", "\n", "\n", - "\n", + "\n", "subject.Zygosity\n", - "\n", - "\n", - "subject.Zygosity\n", + "\n", + "\n", + "subject.Zygosity\n", "\n", "\n", "\n", "\n", - "\n", + "\n", "subject.Subject->subject.Zygosity\n", - "\n", + "\n", "\n", - "\n", - "\n", - "lab.Lab\n", - "\n", - "\n", - "lab.Lab\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "lab.Lab->lab.Lab.Organization\n", - "\n", - "\n", - "\n", - "\n", - "lab.Lab->lab.LabMembership\n", - "\n", - "\n", - "\n", - "\n", - "lab.Lab->subject.Subject.Lab\n", - "\n", - "\n", - "\n", - "\n", - "lab.Location\n", - "\n", - "\n", - "lab.Location\n", + "\n", + "\n", + "subject.SubjectDeath\n", + "\n", + "\n", + "subject.SubjectDeath\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Lab->lab.Location\n", - "\n", + "\n", + "\n", + "subject.Subject->subject.SubjectDeath\n", + "\n", "\n", - "\n", - "\n", - "lab.UserRole\n", - "\n", - "\n", - "lab.UserRole\n", + "\n", + "\n", + "subject.Subject.Lab\n", + "\n", + "\n", + "subject.Subject.Lab\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.UserRole->lab.LabMembership\n", - "\n", + "\n", + "\n", + "subject.Subject->subject.Subject.Lab\n", + "\n", "\n", - "\n", - "\n", - "subject.Strain\n", - "\n", - "\n", - "subject.Strain\n", + "\n", + "\n", + "session.Session\n", + "\n", + "\n", + "session.Session\n", "\n", "\n", "\n", - "\n", - "\n", - "subject.Strain->subject.Subject.Strain\n", - "\n", + "\n", + "\n", + "subject.Subject->session.Session\n", + "\n", "\n", - "\n", - "\n", - "lab.Organization\n", - "\n", - "\n", - "lab.Organization\n", + "\n", + "\n", + "subject.Subject.User\n", + "\n", + "\n", + "subject.Subject.User\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Organization->lab.Lab.Organization\n", - "\n", - "\n", - "\n", - "\n", - "model.RecordingInfo\n", - "\n", - "\n", - "model.RecordingInfo\n", - "\n", - "\n", + "\n", + "\n", + "subject.Subject->subject.Subject.User\n", + "\n", "\n", "\n", - "\n", + "\n", "subject.Allele.Source\n", - "\n", - "\n", - "subject.Allele.Source\n", + "\n", + "\n", + "subject.Allele.Source\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.ProtocolType\n", - "\n", - "\n", - "lab.ProtocolType\n", + "\n", + "\n", + "session.SessionNote\n", + "\n", + "\n", + "session.SessionNote\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Protocol\n", - "\n", - "\n", - "lab.Protocol\n", - "\n", + "\n", + "\n", + "subject.SubjectDeath->subject.SubjectCull\n", + "\n", "\n", + "\n", + "\n", + "train.VideoSet.File\n", + "\n", + "\n", + "train.VideoSet.File\n", + "\n", "\n", - "\n", - "\n", - "lab.ProtocolType->lab.Protocol\n", - "\n", "\n", "\n", - "\n", + "\n", "model.BodyPart\n", - "\n", - "\n", - "model.BodyPart\n", + "\n", + "\n", + "model.BodyPart\n", "\n", "\n", "\n", "\n", - "\n", + "\n", "model.BodyPart->model.Model.BodyPart\n", - "\n", + "\n", "\n", - "\n", - "\n", - "train.VideoSet.File\n", - "\n", - "\n", - "train.VideoSet.File\n", + "\n", + "\n", + "session.SessionDirectory\n", + "\n", + "\n", + "session.SessionDirectory\n", "\n", "\n", "\n", - "\n", - "\n", - "model.Model->model.Model.BodyPart\n", - "\n", - "\n", - "\n", - "\n", - "model.Model->model.PoseEstimationTask\n", - "\n", + "\n", + "\n", + "session.Session->session.Session.Attribute\n", + "\n", "\n", - "\n", - "\n", - "model.ModelEvaluation\n", - "\n", - "\n", - "model.ModelEvaluation\n", - "\n", + "\n", + "\n", + "session.Session->session.ProjectSession\n", + "\n", "\n", + "\n", + "\n", + "session.Session->session.SessionNote\n", + "\n", "\n", - "\n", - "\n", - "model.Model->model.ModelEvaluation\n", - "\n", + "\n", + "\n", + "session.Session->session.SessionDirectory\n", + "\n", "\n", - "\n", - "\n", - "train.VideoSet\n", - "\n", - "\n", - "train.VideoSet\n", + "\n", + "\n", + "model.VideoRecording\n", + "\n", + "\n", + "model.VideoRecording\n", "\n", "\n", "\n", - "\n", - "\n", - "train.VideoSet->train.VideoSet.File\n", - "\n", + "\n", + "\n", + "session.Session->model.VideoRecording\n", + "\n", "\n", - "\n", - "\n", - "train.VideoSet->train.TrainingTask\n", - "\n", - "\n", - "\n", - "\n", - "lab.Source\n", - "\n", - "\n", - "lab.Source\n", + "\n", + "\n", + "session.SessionExperimenter\n", + "\n", + "\n", + "session.SessionExperimenter\n", "\n", "\n", "\n", - "\n", - "\n", - "lab.Source->subject.Subject.Source\n", - "\n", - "\n", - "\n", - "\n", - "lab.Source->subject.Allele.Source\n", - "\n", + "\n", + "\n", + "session.Session->session.SessionExperimenter\n", + "\n", "\n", "\n", - "\n", + "\n", "subject.Allele\n", - "\n", - "\n", - "subject.Allele\n", + "\n", + "\n", + "subject.Allele\n", "\n", "\n", "\n", - "\n", - "\n", - "subject.Allele->subject.Line.Allele\n", - "\n", - "\n", "\n", - "\n", + "\n", "subject.Allele->subject.Allele.Source\n", - "\n", + "\n", "\n", "\n", - "\n", + "\n", "subject.Allele->subject.Zygosity\n", - "\n", + "\n", "\n", - "\n", - "\n", - "model.VideoRecording->model.PoseEstimationTask\n", - "\n", + "\n", + "\n", + "subject.Allele->subject.Line.Allele\n", + "\n", + "\n", + "\n", + "\n", + "train.VideoSet\n", + "\n", + "\n", + "train.VideoSet\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "train.VideoSet->train.VideoSet.File\n", + "\n", + "\n", + "\n", + "\n", + "train.VideoSet->train.TrainingTask\n", + "\n", "\n", "\n", - "\n", + "\n", "model.VideoRecording->model.VideoRecording.File\n", - "\n", + "\n", + "\n", + "\n", + "\n", + "model.RecordingInfo\n", + "\n", + "\n", + "model.RecordingInfo\n", + "\n", + "\n", "\n", "\n", - "\n", + "\n", "model.VideoRecording->model.RecordingInfo\n", - "\n", - "\n", - "\n", - "\n", - "train.ModelTraining\n", - "\n", - "\n", - "train.ModelTraining\n", - "\n", + "\n", "\n", + "\n", + "\n", + "model.VideoRecording->model.PoseEstimationTask\n", + "\n", "\n", "\n", - "\n", + "\n", "train.TrainingTask->train.ModelTraining\n", - "\n", - "\n", - "\n", - "\n", - "subject.SubjectCull\n", - "\n", - "\n", - "subject.SubjectCull\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "subject.SubjectDeath->subject.SubjectCull\n", - "\n", + "\n", "\n", - "\n", - "\n", - "lab.Protocol->subject.Subject.Protocol\n", - "\n", + "\n", + "\n", + "model.PoseEstimationTask->model.PoseEstimation\n", + "\n", "\n", "\n", "" ], "text/plain": [ - "" + "" ] }, "execution_count": 5, @@ -992,22 +785,14 @@ } ], "source": [ - "(\n", - " dj.Diagram(subject)\n", - " + dj.Diagram(lab)\n", - " + dj.Diagram(session)\n", - " + dj.Diagram(model)\n", - " + dj.Diagram(train)\n", - ")" + "(dj.Diagram(subject) + dj.Diagram(session) + dj.Diagram(model) + dj.Diagram(train))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "As evident, this data pipeline is fairly comprehensive, encompassing several tables associated with different DeepLabCut components like model, train, and evaluation. A few tables, such as `Subject` or `Lab`, while integral to the pipeline, fall outside the scope of the scope of element-deeplabcut tutorial as they are upstream. \n", - "\n", - "Our focus in this tutorial will be primarily on the two core schemas:" + "As evident from the diagram, this data pipeline encompasses several tables associated with different pose estimation components like model, train, and evaluation. A few tables, such as `subject.Subject` or `session.Session`, while important for a complete pipeline, fall outside the scope of the `element-deeplabcut` tutorial, and will therefore, not be explored extensively here. The primary focus of this tutorial will be on the `train` and `model` schemas." ] }, { @@ -1018,225 +803,225 @@ { "data": { "image/svg+xml": [ - "\n", + "\n", "\n", "%3\n", - "\n", - "\n", + "\n", + "\n", "\n", - "train.TrainingParamSet\n", - "\n", - "\n", - "train.TrainingParamSet\n", + "model.Model.BodyPart\n", + "\n", + "\n", + "model.Model.BodyPart\n", "\n", "\n", "\n", - "\n", - "\n", - "train.TrainingTask\n", - "\n", - "\n", - "train.TrainingTask\n", + "\n", + "\n", + "model.PoseEstimation.BodyPartPosition\n", + "\n", + "\n", + "model.PoseEstimation.BodyPartPosition\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "train.TrainingParamSet->train.TrainingTask\n", - "\n", + "model.Model.BodyPart->model.PoseEstimation.BodyPartPosition\n", + "\n", "\n", - "\n", - "\n", - "model.Model\n", - "\n", - "\n", - "model.Model\n", + "\n", + "\n", + "train.VideoSet\n", + "\n", + "\n", + "train.VideoSet\n", "\n", "\n", "\n", - "\n", - "\n", - "train.TrainingParamSet->model.Model\n", - "\n", - "\n", - "\n", - "\n", - "model.VideoRecording\n", - "\n", - "\n", - "model.VideoRecording\n", + "\n", + "\n", + "train.VideoSet.File\n", + "\n", + "\n", + "train.VideoSet.File\n", "\n", "\n", "\n", - "\n", - "\n", - "model.PoseEstimationTask\n", - "\n", - "\n", - "model.PoseEstimationTask\n", + "\n", + "\n", + "train.VideoSet->train.VideoSet.File\n", + "\n", + "\n", + "\n", + "\n", + "train.TrainingTask\n", + "\n", + "\n", + "train.TrainingTask\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "model.VideoRecording->model.PoseEstimationTask\n", - "\n", + "train.VideoSet->train.TrainingTask\n", + "\n", + "\n", + "\n", + "\n", + "model.ModelEvaluation\n", + "\n", + "\n", + "model.ModelEvaluation\n", + "\n", + "\n", "\n", "\n", - "\n", + "\n", "model.VideoRecording.File\n", - "\n", - "\n", - "model.VideoRecording.File\n", + "\n", + "\n", + "model.VideoRecording.File\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "model.VideoRecording\n", + "\n", + "\n", + "model.VideoRecording\n", "\n", "\n", "\n", "\n", "\n", "model.VideoRecording->model.VideoRecording.File\n", - "\n", + "\n", "\n", "\n", - "\n", + "\n", "model.RecordingInfo\n", - "\n", - "\n", - "model.RecordingInfo\n", + "\n", + "\n", + "model.RecordingInfo\n", "\n", "\n", "\n", "\n", "\n", "model.VideoRecording->model.RecordingInfo\n", - "\n", + "\n", "\n", - "\n", - "\n", - "model.Model.BodyPart\n", - "\n", - "\n", - "model.Model.BodyPart\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "model.PoseEstimation.BodyPartPosition\n", - "\n", - "\n", - "model.PoseEstimation.BodyPartPosition\n", + "\n", + "\n", + "model.PoseEstimationTask\n", + "\n", + "\n", + "model.PoseEstimationTask\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "model.Model.BodyPart->model.PoseEstimation.BodyPartPosition\n", - "\n", + "model.VideoRecording->model.PoseEstimationTask\n", + "\n", "\n", - "\n", - "\n", - "model.PoseEstimation\n", - "\n", - "\n", - "model.PoseEstimation\n", + "\n", + "\n", + "model.BodyPart\n", + "\n", + "\n", + "model.BodyPart\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "model.PoseEstimationTask->model.PoseEstimation\n", - "\n", + "model.BodyPart->model.Model.BodyPart\n", + "\n", "\n", - "\n", - "\n", - "model.BodyPart\n", - "\n", - "\n", - "model.BodyPart\n", + "\n", + "\n", + "model.Model\n", + "\n", + "\n", + "model.Model\n", "\n", "\n", "\n", - "\n", + "\n", "\n", - "model.BodyPart->model.Model.BodyPart\n", - "\n", + "model.Model->model.Model.BodyPart\n", + "\n", + "\n", + "\n", + "\n", + "model.Model->model.ModelEvaluation\n", + "\n", + "\n", + "\n", + "\n", + "model.Model->model.PoseEstimationTask\n", + "\n", "\n", "\n", - "\n", + "\n", "train.ModelTraining\n", - "\n", - "\n", - "train.ModelTraining\n", + "\n", + "\n", + "train.ModelTraining\n", "\n", "\n", "\n", - "\n", - "\n", - "train.TrainingTask->train.ModelTraining\n", - "\n", - "\n", - "\n", - "\n", - "train.VideoSet.File\n", - "\n", - "\n", - "train.VideoSet.File\n", + "\n", + "\n", + "train.TrainingParamSet\n", + "\n", + "\n", + "train.TrainingParamSet\n", "\n", "\n", "\n", - "\n", - "\n", - "model.Model->model.Model.BodyPart\n", - "\n", - "\n", - "\n", + "\n", "\n", - "model.Model->model.PoseEstimationTask\n", - "\n", + "train.TrainingParamSet->model.Model\n", + "\n", "\n", - "\n", - "\n", - "model.ModelEvaluation\n", - "\n", - "\n", - "model.ModelEvaluation\n", - "\n", + "\n", + "\n", + "train.TrainingParamSet->train.TrainingTask\n", + "\n", "\n", + "\n", + "\n", + "model.PoseEstimation\n", + "\n", + "\n", + "model.PoseEstimation\n", + "\n", "\n", - "\n", - "\n", - "model.Model->model.ModelEvaluation\n", - "\n", "\n", "\n", "\n", "model.PoseEstimation->model.PoseEstimation.BodyPartPosition\n", - "\n", - "\n", - "\n", - "\n", - "train.VideoSet\n", - "\n", - "\n", - "train.VideoSet\n", - "\n", - "\n", + "\n", "\n", - "\n", + "\n", "\n", - "train.VideoSet->train.TrainingTask\n", - "\n", + "train.TrainingTask->train.ModelTraining\n", + "\n", "\n", - "\n", + "\n", "\n", - "train.VideoSet->train.VideoSet.File\n", - "\n", + "model.PoseEstimationTask->model.PoseEstimation\n", + "\n", "\n", "\n", "" ], "text/plain": [ - "" + "" ] }, "execution_count": 6, @@ -1252,14 +1037,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "This diagram represents an example of the `element-deeplabcut` pipeline." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### **Step 1 - Register an Existing Model in the DataJoint Pipeline**" + "### **Register an existing model in the DataJoint pipeline**" ] }, { @@ -1277,9 +1055,7 @@ "metadata": {}, "outputs": [], "source": [ - "config_file_rel = (\n", - " \"./example_data/inbox/from_top_tracking-DataJoint-2023-10-11/config.yaml\"\n", - ")" + "config_file_rel = \"from_top_tracking-DataJoint-2023-10-11/config.yaml\"" ] }, { @@ -1300,21 +1076,21 @@ "name": "stderr", "output_type": "stream", "text": [ - "2023-10-19 19:26:49.961931: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA\n", + "2023-12-22 21:34:44.474084: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA\n", "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", - "2023-10-19 19:26:50.069069: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/site-packages/cv2/../../lib64:/lib:/opt/conda/lib\n", - "2023-10-19 19:26:50.069107: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.\n", - "2023-10-19 19:26:50.091643: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n", - "2023-10-19 19:26:51.042839: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/site-packages/cv2/../../lib64:/lib:/opt/conda/lib\n", - "2023-10-19 19:26:51.043018: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/site-packages/cv2/../../lib64:/lib:/opt/conda/lib\n", - "2023-10-19 19:26:51.043033: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n" + "2023-12-22 21:34:45.132521: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/site-packages/cv2/../../lib64:/lib:/opt/conda/lib\n", + "2023-12-22 21:34:45.132563: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.\n", + "2023-12-22 21:34:45.177758: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n", + "2023-12-22 21:34:46.381666: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/site-packages/cv2/../../lib64:/lib:/opt/conda/lib\n", + "2023-12-22 21:34:46.381891: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/site-packages/cv2/../../lib64:/lib:/opt/conda/lib\n", + "2023-12-22 21:34:46.381911: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "Loading DLC 2.3.7...\n", + "Loading DLC 2.3.8...\n", "DLC loaded in light mode; you cannot use any GUI (labeling, relabeling and standalone GUI)\n", "--- DLC Model specification to be inserted ---\n", "\tmodel_name: from_top_tracking_model_test\n", @@ -1390,21 +1166,23 @@ "\n", " \n", " \n", "
\n", - " \n", + "
\n", "
\n", "

model_name

\n", " User-friendly model name\n", @@ -1527,14 +1305,14 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### **Step 2 - Insert Subject, Session, and Behavior Videos**" + "### **Insert example data into subject and session tables**" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "Now, let's delve into the `Subject` and `Session` tables and include some example data." + "Now, let's delve into the `subject.Subject` and `session.Session` tables and include some example data." ] }, { @@ -1548,21 +1326,23 @@ "\n", " \n", "