def from_parameters(params: Parameters) -> "WorkflowBuilder":
wb = WorkflowBuilder(
name=params.string("workflow_name", default="Workflow"),
created_by=params.string("workflow_created",
default="Default Constructor"),
workflow_directory=params.creatable_directory(
"workflow_directory"),
default_site=params.string("site"),
conda_script_generator=CondaJobScriptGenerator.from_parameters(
params),
docker_script_generator=DockerJobScriptGenerator.from_parameters(
params),
namespace=params.string("namespace"),
default_resource_request=ResourceRequest.from_parameters(params),
data_configuration=params.string("data_configuration",
default="sharedfs"),
experiment_name=params.string("experiment_name", default=""),
)
if params.boolean("include_nas", default=True):
add_local_nas_to_sites(
wb._sites_catalog,
params # pylint: disable=protected-access
)
if params.boolean("include_saga", default=True):
add_saga_cluster_to_sites(
wb._sites_catalog,
params # pylint: disable=protected-access
)
configure_saga_properities(
wb._properties,
params # pylint: disable=protected-access
)
return wb
def main(params: Parameters):
adam_root = params.existing_directory("adam_root")
m9_experiments_dir = adam_root / "parameters" / "experiments" / "m9"
param_files: List[Path] = []
if params.boolean("include_objects", default=True):
param_files.append(m9_experiments_dir / "objects.params")
if params.boolean("include_attributes", default=True):
param_files.append(m9_experiments_dir / "attributes.params")
if params.boolean("include_relations", default=True):
param_files.append(m9_experiments_dir / "relations.params")
if params.boolean("include_events", default=True):
param_files.append(m9_experiments_dir / "events.params")
# This activates a special "debug" curriculum,
# which is meant to be edited in the code by a developer to do fine-grained debugging.
if params.boolean("include_debug", default=False):
param_files.append(m9_experiments_dir / "debug.params")
# If any of the param files don't exist, bail out earlier instead of making the user
# wait for the error.
for param_file in param_files:
if not param_file.exists():
raise RuntimeError(
f"Expected param file {param_file} does not exist")
for param_file in param_files:
logging.info("Running %s", param_file)
experiment_params = YAMLParametersLoader().load(param_file)
log_experiment_entry_point(experiment_params)
def pre_observer(
self,
*,
params: Parameters = Parameters.empty(),
experiment_group_dir: Optional[Path] = None,
) -> "DescriptionObserver": # type: ignore
track_accuracy = params.boolean("include_acc_observer", default=False)
log_accuracy = params.boolean("accuracy_to_txt", default=False)
log_accuracy_path = params.string(
"accuracy_logging_path",
default=f"{experiment_group_dir}/accuracy_pre_out.txt"
if experiment_group_dir else "accuracy_pre_out.txt",
)
track_precision_recall = params.boolean("include_pr_observer",
default=False)
log_precision_recall = params.boolean("log_pr", default=False)
log_precision_recall_path = params.string(
"pr_log_path",
default=f"{experiment_group_dir}/pr_post_out.txt"
if experiment_group_dir else "pr_post_out.txt",
)
return HTMLLoggerPreObserver(
name="Pre-observer",
html_logger=self,
candidate_accuracy_observer=CandidateAccuracyObserver(
name="Pre-observer-acc",
accuracy_to_txt=log_accuracy,
txt_path=log_accuracy_path,
) if track_accuracy else None,
precision_recall_observer=PrecisionRecallObserver(
name="Pre-observer-pr",
make_report=log_precision_recall,
txt_path=log_precision_recall_path,
) if track_precision_recall else None,
)
def main(params: Parameters) -> None:
root_output_directory = params.creatable_directory("output_directory")
curriculum_string = params.string("curriculum",
valid_options=STR_TO_CURRICULUM.keys(),
default="phase1")
language_mode = params.enum("language_mode",
LanguageMode,
default=LanguageMode.ENGLISH)
language_string = str(language_mode).split(".")[-1].lower()
num_samples = params.optional_positive_integer("num_samples")
num_noise_objects = params.optional_positive_integer("num_noise_objects")
phase1_curriculum_dir = root_output_directory / language_string / curriculum_string
phase1_curriculum_dir.mkdir(parents=True, exist_ok=True)
# We lazily instantiate the curriculum so we don't need to worry
# about any of them we don't actually use.
curriculum_to_render = STR_TO_CURRICULUM[curriculum_string](
num_samples, num_noise_objects,
phase2_language_generator(language_mode))
sort_by_utterance_length_flag = params.boolean("sort_by_utterance",
default=False)
if sort_by_utterance_length_flag:
random_seed = params.integer("random_seed", default=1)
CurriculumToHtmlDumper().dump_to_html_as_sorted_by_utterance_length(
curriculum_to_render,
output_directory=phase1_curriculum_dir,
title="GAILA Phase 1 Curriculum Sorted by Utterance Length",
curriculum_string=curriculum_string,
random_seed=random_seed,
)
else:
CurriculumToHtmlDumper().dump_to_html(
curriculum_to_render,
output_directory=phase1_curriculum_dir,
title="GAILA Phase 1 Curriculum",
)
def test_observer(
self,
*,
params: Parameters = Parameters.empty(),
experiment_group_dir: Optional[Path] = None,
) -> "DescriptionObserver": # type: ignore
# these are the params to use for writing accuracy to a text file at every iteration (e.g. to graph later)
track_accuracy = params.boolean("include_acc_observer", default=True)
log_accuracy = params.boolean("accuracy_to_txt", default=False)
log_accuracy_path = params.string(
"accuracy_logging_path",
default=f"{experiment_group_dir}/accuracy_test_out.txt"
if experiment_group_dir else "accuracy_test_out.txt",
)
track_precision_recall = params.boolean("include_pr_observer",
default=False)
log_precision_recall = params.boolean("log_pr", default=False)
log_precision_recall_path = params.string(
"pr_log_path",
default=f"{experiment_group_dir}/pr_test_out.txt"
if experiment_group_dir else "pr_test_out.txt",
)
accuracy_observer = None
precision_recall_observer = None
if track_accuracy:
accuracy_observer = CandidateAccuracyObserver(
name="Test-observer-acc",
accuracy_to_txt=log_accuracy,
txt_path=log_accuracy_path,
)
if track_precision_recall:
precision_recall_observer = PrecisionRecallObserver(
name="Test-observer-pr",
make_report=log_precision_recall,
txt_path=log_precision_recall_path,
)
return HTMLLoggerPostObserver(
name="t-observer",
html_logger=self,
candidate_accuracy_observer=accuracy_observer,
precision_recall_observer=precision_recall_observer,
test_mode=True,
)
def create_logger(params: Parameters) -> "LearningProgressHtmlLogger":
output_dir = params.creatable_directory("experiment_group_dir")
experiment_name = params.string("experiment")
include_links_to_images = params.optional_boolean("include_image_links")
num_pretty_descriptions = params.positive_integer(
"num_pretty_descriptions", default=3
)
sort_by_length = params.boolean(
"sort_learner_descriptions_by_length", default=False
)
logging_dir = output_dir / experiment_name
logging_dir.mkdir(parents=True, exist_ok=True)
output_html_path = str(logging_dir / "index.html")
if include_links_to_images is None:
include_links_to_images = False
logging.info("Experiment will be logged to %s", output_html_path)
with open(output_html_path, "w") as outfile:
html_dumper = CurriculumToHtmlDumper()
outfile.write(f"<head>\n\t<style>{CSS}\n\t</style>\n</head>")
outfile.write(f"\n<body>\n\t<h1>{experiment_name}</h1>")
# A JavaScript function to allow toggling perception information
outfile.write(
"""
<script>
function myFunction(id) {
var x = document.getElementById(id);
if (x.style.display === "none") {
x.style.display = "block";
} else {
x.style.display = "none";
}
}
</script>
"""
)
return LearningProgressHtmlLogger(
outfile_dir=output_html_path,
html_dumper=html_dumper,
include_links_to_images=include_links_to_images,
num_pretty_descriptions=num_pretty_descriptions,
sort_by_length=sort_by_length,
)
def log_experiment_entry_point(params: Parameters) -> None:
experiment_name = params.string("experiment")
debug_log_dir = params.optional_creatable_directory("debug_log_directory")
graph_logger: Optional[HypothesisLogger]
if debug_log_dir:
logging.info("Debug graphs will be written to %s", debug_log_dir)
graph_logger = HypothesisLogger(debug_log_dir,
enable_graph_rendering=True)
else:
graph_logger = None
logger = LearningProgressHtmlLogger.create_logger(params)
language_mode = params.enum("language_mode",
LanguageMode,
default=LanguageMode.ENGLISH)
(training_instance_groups,
test_instance_groups) = curriculum_from_params(params, language_mode)
execute_experiment(
Experiment(
name=experiment_name,
training_stages=training_instance_groups,
learner_factory=learner_factory_from_params(
params, graph_logger, language_mode),
pre_example_training_observers=[
logger.pre_observer(),
CandidateAccuracyObserver("pre-acc-observer"),
],
post_example_training_observers=[logger.post_observer()],
test_instance_groups=test_instance_groups,
test_observers=[logger.test_observer()],
sequence_chooser=RandomChooser.for_seed(0),
),
log_path=params.optional_creatable_directory("hypothesis_log_dir"),
log_hypotheses_every_n_examples=params.integer(
"log_hypothesis_every_n_steps", default=250),
log_learner_state=params.boolean("log_learner_state", default=True),
learner_logging_path=params.optional_creatable_directory(
"experiment_group_dir"),
starting_point=params.integer("starting_point", default=-1),
point_to_log=params.integer("point_to_log", default=0),
load_learner_state=params.optional_existing_file("learner_state_path"),
)
def build_pursuit_curriculum(
num_samples: Optional[int],
num_noise_objects: Optional[int],
language_generator: LanguageGenerator[
HighLevelSemanticsSituation, LinearizedDependencyTree
],
*,
pursuit_curriculum_params: Parameters = Parameters.empty(),
) -> Sequence[Phase1InstanceGroup]:
num_instances = pursuit_curriculum_params.integer(
"num_instances", default=num_samples if num_samples else 10
)
num_noise_instances = pursuit_curriculum_params.integer(
"num_noise_instances", default=num_noise_objects if num_noise_objects else 2
)
num_objects_in_instance = pursuit_curriculum_params.integer(
"num_objects_in_instance", default=3
)
add_gaze = pursuit_curriculum_params.boolean("add_gaze", default=False)
prob_given = pursuit_curriculum_params.floating_point("prob_given", default=1.0)
prob_not_given = pursuit_curriculum_params.floating_point(
"prob_not_given", default=0.0
)
rng = random.Random()
rng.seed(0)
gaze_perciever = GazePerceivedNoisily(
rng=rng,
prob_gaze_perceived_given_gaze=prob_given,
prob_gaze_perceived_given_not_gaze=prob_not_given,
)
perception_generator = HighLevelSemanticsSituationToDevelopmentalPrimitivePerceptionGenerator(
ontology=GAILA_PHASE_2_ONTOLOGY, gaze_strategy=gaze_perciever
)
return [
make_simple_pursuit_curriculum(
target_objects=M6_CURRICULUM_ALL_OBJECTS,
num_instances=num_instances,
num_objects_in_instance=num_objects_in_instance,
num_noise_instances=num_noise_instances,
language_generator=language_generator,
add_gaze=add_gaze,
perception_generator=perception_generator,
)
]
def main(params: Parameters):
conda_script_generator = CondaJobScriptGenerator.from_parameters(params)
entry_point = params.string("entry_point")
work_dir = params.optional_creatable_directory(
"working_directory") or Path(os.getcwd())
stdout_file = params.string("log_file") or work_dir / "___stdout.log"
shell_script = conda_script_generator.generate_shell_script(
entry_point_name=entry_point,
param_file=params.existing_file("job_param_file"),
working_directory=work_dir,
stdout_file=stdout_file,
)
params.creatable_file("conda_script_path").write_text( # type: ignore
shell_script, encoding="utf-8")
if params.boolean("echo_template", default=False):
print(shell_script)
def _explicit_split(source: KeyValueSource[str, bytes], params: Parameters):
explicit_split_namespace = params.namespace(_EXPLICIT_SPLIT_PARAM)
# We track these so we can ensure the split is a complete partition of the input,
# if the user so desires.
keys_copied = []
for split_namespace in explicit_split_namespace.sub_namespaces():
keys_for_split = file_lines_to_set(
split_namespace.existing_file("keys_file"))
with KeyValueSink.zip_bytes_sink(
split_namespace.creatable_file("output_file")) as split_sink:
for key in keys_for_split:
source_value = source.get(key)
if source_value is not None:
split_sink.put(key, source_value)
keys_copied.append(key)
else:
error_message = (
f"For split specified in {split_namespace.namespace_prefix}, "
f"requested key value {key} not found in {source}.")
available_keys = source.keys()
if available_keys is not None:
error_message = (
f"{error_message} Here are a few" # type: ignore
f"available keys: {str_list_limited(source.keys(), 10)}"
)
raise RuntimeError(error_message)
if params.boolean("must_be_exhaustive", default=True):
keys_not_copied = immutableset(source.keys()) - set(keys_copied)
if keys_not_copied:
raise RuntimeError(
f"Expected the split to be a partition, but "
f"{len(keys_not_copied)} were not included in any output split, "
f"including {str_list_limited(keys_not_copied, 10)}. "
f"If you did not intend the split to be exhaustive, "
f"please specify set parameter must_be_exhaustive to False")
def main(cluster_params: Parameters, job_param_file: Path) -> None:
runner = SlurmPythonRunner.from_parameters(cluster_params)
job_params = YAMLParametersLoader().load(job_param_file)
entry_point = job_params.string("entry_point")
memory = MemoryAmount.parse(job_params.string("memory"))
runner.run_entry_point(
entry_point_name=entry_point,
param_file=job_param_file,
partition=cluster_params.string("partition"),
working_directory=job_params.optional_creatable_directory(
"working_directory") or Path(os.getcwd()),
num_gpus=job_params.integer("num_gpus",
default=0,
valid_range=Range.at_least(0)),
num_cpus=job_params.integer("num_cpus",
default=1,
valid_range=Range.at_least(1)),
job_name=job_params.string("job_name", default=entry_point),
memory_request=memory,
echo_template=cluster_params.boolean("echo_template", default=False),
slurm_script_path=job_params.optional_creatable_file(
"slurm_script_path"),
)
def log_experiment_entry_point(params: Parameters) -> None:
experiment_name = params.string("experiment")
debug_log_dir = params.optional_creatable_directory("debug_log_directory")
graph_logger: Optional[HypothesisLogger]
if debug_log_dir:
logging.info("Debug graphs will be written to %s", debug_log_dir)
graph_logger = HypothesisLogger(debug_log_dir,
enable_graph_rendering=True)
else:
graph_logger = None
logger = LearningProgressHtmlLogger.create_logger(params)
language_mode = params.enum("language_mode",
LanguageMode,
default=LanguageMode.ENGLISH)
curriculum_repository_path = params.optional_existing_directory(
"load_from_curriculum_repository")
if curriculum_repository_path:
curriculum = read_experiment_curriculum(curriculum_repository_path,
params, language_mode)
(training_instance_groups, test_instance_groups) = (
curriculum.train_curriculum,
curriculum.test_curriculum,
)
else:
(training_instance_groups,
test_instance_groups) = curriculum_from_params(params, language_mode)
experiment_group_dir = params.optional_creatable_directory(
"experiment_group_dir")
resume_from_last_logged_state = params.boolean(
"resume_from_latest_logged_state", default=False)
# Check if we have explicit observer states to load
observers_state = params.optional_existing_file("observers_state_path")
test_observer = [] # type: ignore
pre_observer = [] # type: ignore
post_observer = [] # type: ignore
if resume_from_last_logged_state and observers_state:
raise RuntimeError(
f"Can not resume from last logged state and provide explicit observer state paths"
)
if resume_from_last_logged_state:
if not experiment_group_dir:
raise RuntimeError(
"experiment_group_dir must be specified when resume_from_last_logged_state is true."
)
# Try to Load Observers
for _, observers_state_path in observer_states_by_most_recent(
cast(Path, experiment_group_dir) / "observer_state",
"observers_state_at_"):
try:
with observers_state_path.open("rb") as f:
observers_holder = pickle.load(f)
pre_observer = observers_holder.pre_observers
post_observer = observers_holder.post_observers
test_observer = observers_holder.test_observers
except OSError:
logging.warning(
"Unable to open observer state at %s; skipping.",
str(observers_state_path),
)
except pickle.UnpicklingError:
logging.warning(
"Couldn't unpickle observer state at %s; skipping.",
str(observers_state_path),
)
if not pre_observer and not post_observer and not test_observer:
logging.warning("Reverting to default observers.")
pre_observer = [
logger.pre_observer( # type: ignore
params=params.namespace_or_empty("pre_observer"),
experiment_group_dir=experiment_group_dir,
)
]
post_observer = [
logger.post_observer( # type: ignore
params=params.namespace_or_empty("post_observer"),
experiment_group_dir=experiment_group_dir,
)
]
test_observer = [
logger.test_observer( # type: ignore
params=params.namespace_or_empty("test_observer"),
experiment_group_dir=experiment_group_dir,
)
]
elif observers_state:
try:
with observers_state.open("rb") as f:
observers_holder = pickle.load(f)
pre_observer = observers_holder.pre_observers
post_observer = observers_holder.post_observers
test_observer = observers_holder.test_observers
except OSError:
logging.warning("Unable to open observer state at %s; skipping.",
str(observers_state))
except pickle.UnpicklingError:
logging.warning(
"Couldn't unpickle observer state at %s; skipping.",
str(observers_state))
else:
pre_observer = [
logger.pre_observer( # type: ignore
params=params.namespace_or_empty("pre_observer"),
experiment_group_dir=experiment_group_dir,
)
]
post_observer = [
logger.post_observer( # type: ignore
params=params.namespace_or_empty("post_observer"),
experiment_group_dir=experiment_group_dir,
)
]
test_observer = [
logger.test_observer( # type: ignore
params=params.namespace_or_empty("test_observer"),
experiment_group_dir=experiment_group_dir,
)
]
execute_experiment(
Experiment(
name=experiment_name,
training_stages=training_instance_groups,
learner_factory=learner_factory_from_params(
params, graph_logger, language_mode),
pre_example_training_observers=pre_observer,
post_example_training_observers=post_observer,
test_instance_groups=test_instance_groups,
test_observers=test_observer,
sequence_chooser=RandomChooser.for_seed(0),
),
log_path=params.optional_creatable_directory("hypothesis_log_dir"),
log_hypotheses_every_n_examples=params.integer(
"log_hypothesis_every_n_steps", default=250),
log_learner_state=params.boolean("log_learner_state", default=True),
learner_logging_path=experiment_group_dir,
starting_point=params.integer("starting_point", default=0),
point_to_log=params.integer("point_to_log", default=0),
load_learner_state=params.optional_existing_file("learner_state_path"),
resume_from_latest_logged_state=resume_from_last_logged_state,
debug_learner_pickling=params.boolean("debug_learner_pickling",
default=False),
)
def main(
params: Parameters,
scenes_iterable_input: Optional[Iterable[Phase1InstanceGroup]] = None,
output_directory: Optional[Path] = None,
visualizer: Optional[SituationVisualizer] = None,
) -> None:
language_mode = params.enum("language_mode",
LanguageMode,
default=LanguageMode.ENGLISH)
if scenes_iterable_input is None:
scenes_iterable: Iterable[Phase1InstanceGroup] = [
make_curriculum(None, None,
phase2_language_generator(language_mode))
]
else:
scenes_iterable = scenes_iterable_input
num_iterations = params.positive_integer("iterations")
steps_before_vis = params.positive_integer("steps_before_vis")
specific_scene = params.optional_positive_integer("scene")
automatically_save_renderings = params.boolean(
"automatically_save_renderings", default=False)
if "experiment_group_dir" in params:
rendering_filename_generator = from_experiment_filename_generator
else:
rendering_filename_generator = default_filename_generator
screenshot_dir = output_directory
random.seed(params.integer("seed"))
np.random.seed(params.integer("seed"))
if params.string("debug_bounding_boxes", default="off") == "on":
debug_bounding_boxes = True
else:
debug_bounding_boxes = False
if params.string("gaze_arrows", default="off") == "on":
gaze_arrows = True
else:
gaze_arrows = False
# go through curriculum scenes and output geometry types
if visualizer is None:
viz = SituationVisualizer()
else:
viz = visualizer
viz.clear_scene()
model_scales = viz.get_model_scales()
for object_type, multiplier in OBJECT_SCALE_MULTIPLIER_MAP.items():
if object_type in model_scales:
v3 = model_scales[object_type]
new_v3 = (v3[0] * multiplier, v3[1] * multiplier,
v3[2] * multiplier)
model_scales[object_type] = new_v3
else:
model_scales[object_type] = (multiplier, multiplier, multiplier)
for model_name, scale in model_scales.items():
logging.info("SCALE: %s -> %s", model_name, scale.__str__())
# used to start a frame from where the previous one left off
previous_model_positions: Optional[PositionsMap] = None
for scene_number, scene_elements in enumerate(
SceneCreator.create_scenes(scenes_iterable)):
# If a scene number is provided in the params file, only render that scene
if specific_scene and scene_number < specific_scene:
continue
if specific_scene and scene_number > specific_scene:
break
scene_filename = rendering_filename_generator(scene_number,
scene_elements)
if scene_filename in _FILENAMES_USED:
continue
_FILENAMES_USED.add(scene_filename)
print(f"SCENE {scene_number}")
viz.set_title(" ".join(token for token in scene_elements.tokens) +
" (" + str(scene_elements.current_frame + 1) + "/" +
str(scene_elements.total_frames) + ")")
# if this is a new scene, forget the positions from the last scene
if scene_elements.current_frame == 0:
previous_model_positions = None
if automatically_save_renderings:
# if in auto mode and scene contains an excluded vocab word, skip it
skip_scene = False
for token in scene_elements.tokens:
if token in EXCLUDED_VOCAB:
skip_scene = True
if skip_scene:
continue
# for debugging purposes:
# SceneCreator.graph_for_each(scene_elements.object_graph, print_obj_names)
# bind visualizer and properties to top level rendering function:
bound_render_obj = partial(render_obj, viz,
scene_elements.property_map,
previous_model_positions)
# bind visualizer and properties to nested obj rendering function
bound_render_nested_obj = partial(render_obj_nested, viz,
scene_elements.property_map,
previous_model_positions)
# render each object in graph
SceneCreator.graph_for_each_top_level(scene_elements.object_graph,
bound_render_obj,
bound_render_nested_obj)
# apply scale to top level nodes in scene
for node in scene_elements.object_graph:
if (node.name not in OBJECT_NAMES_TO_EXCLUDE and
node.name.split("_")[0] in OBJECT_SCALE_MULTIPLIER_MAP):
viz.multiply_scale(
node.name,
OBJECT_SCALE_MULTIPLIER_MAP[node.name.split("_")[0]])
# find the Region relations that refer to separate objects:
# (e.g. the cookie is in the region of the hand (of the person), not the leg-segment in in the region of the torso).
inter_object_in_region_map: DefaultDict[
ObjectPerception,
List[Region[ObjectPerception]]] = defaultdict(list)
for top_level_node in scene_elements.object_graph:
if top_level_node.perceived_obj in scene_elements.in_region_map:
inter_object_in_region_map[
top_level_node.
perceived_obj] = scene_elements.in_region_map[
top_level_node.perceived_obj]
# print(inter_object_in_region_map)
# we want to assemble a lookup of the offsets (position) of each object's subobjects.
sub_object_offsets = {}
for node_name, node in viz.geo_nodes.items():
child_node_to_offset = {}
recurse_list: List[NodePath] = node.children
while recurse_list:
next_batch: List[NodePath] = []
for child in recurse_list:
next_batch += child.children
# make sure this is a sub-object
if child.hasMat() and child.parent.name != node_name:
# child has non-identity transformation matrix applied to it (transform differs from parent)
# TODO: we could re-export all of the models in such a way to eliminate this extra layer
# in the scene graph
child_node_to_offset[
child.parent.name] = child.get_pos()
recurse_list = next_batch
sub_object_offsets[node_name] = child_node_to_offset
# handle skipping scene
if not automatically_save_renderings:
viz.run_for_seconds(1)
skip_command = input("type 's' and hit ENTER to skip this scene")
if skip_command == "s":
viz.clear_scene()
viz.run_for_seconds(0.25)
continue
handle_to_in_region_map = {
object_perception.debug_handle: region_list
for object_perception, region_list in
inter_object_in_region_map.items()
}
frozen_objects = objects_to_freeze(
handle_to_in_region_map,
scene_elements.situation,
scene_elements.situation_object_to_handle,
)
if scene_elements.interpolated_scene_moving_items:
# freeze everything not included in the interpolated scene
frozen_objects = (immutableset([
key.debug_handle
for key in scene_elements.in_region_map.keys()
]) - scene_elements.interpolated_scene_moving_items)
# now that every object has been instantiated into the scene,
# they need to be re-positioned.
repositioned_map = None
for repositioned_map in _solve_top_level_positions(
top_level_objects=immutableset([
node.perceived_obj for node in scene_elements.object_graph
if node.name not in OBJECT_NAMES_TO_EXCLUDE
]),
sub_object_offsets=sub_object_offsets,
in_region_map=inter_object_in_region_map,
model_scales=model_scales,
frozen_objects=frozen_objects,
iterations=num_iterations,
yield_steps=steps_before_vis,
previous_positions=previous_model_positions,
):
viz.clear_debug_nodes()
viz.clear_gaze_arrows()
if not automatically_save_renderings:
viz.run_for_seconds(0.25)
viz.set_positions(repositioned_map)
if debug_bounding_boxes:
for name in repositioned_map.name_to_position:
viz.add_debug_bounding_box(
name,
repositioned_map.name_to_position[name],
repositioned_map.name_to_scale[name],
)
if gaze_arrows:
for handle, props in scene_elements.property_map.items():
for prop in props:
if isinstance(
prop,
OntologyNode) and prop.handle == "gazed-at":
viz.add_gaze_arrow(
handle,
repositioned_map.name_to_position[handle],
repositioned_map.name_to_scale[handle],
)
# the visualizer seems to need about a second to render an update
if not automatically_save_renderings:
viz.run_for_seconds(1)
# viz.print_scene_graph()
previous_model_positions = None
# only store previous positions when continuing to next frame / scene
previous_model_positions = repositioned_map
viz.run_for_seconds(1)
screenshot(
automatically_save_renderings=automatically_save_renderings,
filename=scene_filename,
screenshot_dir=screenshot_dir,
viz=viz,
)
viz.clear_scene()
viz.run_for_seconds(0.25)
def main(params: Parameters):
adam_root = params.existing_directory("adam_root")
m13_experiments_dir = adam_root / "parameters" / "experiments" / "m13"
use_pegasus = params.boolean("use_pegasus", default=False)
if use_pegasus:
initialize_vista_pegasus_wrapper(params)
param_files: List[Path] = []
if params.boolean("include_objects", default=True):
param_files.append(m13_experiments_dir / "objects.params")
if params.boolean("include_imprecise_size", default=True):
param_files.append(m13_experiments_dir / "imprecise_size.params")
if params.boolean("include_imprecise_temporal", default=True):
param_files.append(m13_experiments_dir / "imprecise_temporal.params")
if params.boolean("include_subtle_verb", default=True):
param_files.append(m13_experiments_dir / "subtle_verb.params")
if params.boolean("include_object_restrictions", default=True):
param_files.append(m13_experiments_dir / "object_restrictions.params")
if params.boolean("include_functionally_defined_objects", default=True):
param_files.append(m13_experiments_dir / "functionally_defined_objects.params")
if params.boolean("include_relations", default=True):
param_files.append(m13_experiments_dir / "relations.params")
if params.boolean("include_generics", default=True):
param_files.append(m13_experiments_dir / "generics.params")
if params.boolean("include_verbs_with_dynamic_prepositions", default=True):
param_files.append(
m13_experiments_dir / "events_with_dynamic_prepositions.params"
)
if params.boolean("include_m9_complete", default=False):
param_files.append(m13_experiments_dir / "m9_complete.params")
if params.boolean("include_m13_complete", default=False):
param_files.append(m13_experiments_dir / "m13_complete.params")
if params.boolean("include_m13_shuffled", default=False):
param_files.append(m13_experiments_dir / "m13_shuffled.params")
# This activates a special "debug" curriculum,
# which is meant to be edited in the code by a developer to do fine-grained debugging.
if params.boolean("include_debug", default=False):
param_files.append(m13_experiments_dir / "debug.params")
# If any of the param files don't exist, bail out earlier instead of making the user
# wait for the error.
for param_file in param_files:
if not param_file.exists():
raise RuntimeError(f"Expected param file {param_file} does not exist")
for param_file in param_files:
logging.info("Running %s", param_file)
experiment_params = YAMLParametersLoader().load(param_file)
if not use_pegasus:
log_experiment_entry_point(experiment_params)
else:
experiment_name = Locator(experiment_params.string("experiment"))
experiment_params = experiment_params.unify(
{
"experiment_group_dir": directory_for(experiment_name) / "output",
"hypothesis_log_dir": directory_for(experiment_name) / "hypotheses",
# State pickles will go under experiment_name/learner_state
"learner_logging_path": directory_for(experiment_name),
"log_learner_state": True,
"resume_from_latest_logged_state": True,
"log_hypothesis_every_n_steps": params.integer(
"save_state_every_n_steps"
),
"debug_learner_pickling": params.boolean(
"debug_learner_pickling", default=False
),
}
)
run_python_on_parameters(
experiment_name, log_experiment_script, experiment_params, depends_on=[]
)
if use_pegasus:
write_workflow_description()
def main(params: Parameters):
adam_root = params.existing_directory("adam_root")
m13_experiments_dir = adam_root / "parameters" / "experiments" / "m13"
param_files: List[Path] = []
if params.boolean("include_objects", default=True):
param_files.append(m13_experiments_dir / "objects.params")
if params.boolean("include_imprecise_size", default=True):
param_files.append(m13_experiments_dir / "imprecise_size.params")
if params.boolean("include_imprecise_temporal", default=True):
param_files.append(m13_experiments_dir / "imprecise_temporal.params")
if params.boolean("include_subtle_verb", default=True):
param_files.append(m13_experiments_dir / "subtle_verb.params")
if params.boolean("include_object_restrictions", default=True):
param_files.append(m13_experiments_dir / "object_restrictions.params")
if params.boolean("include_functionally_defined_objects", default=True):
param_files.append(m13_experiments_dir /
"functionally_defined_objects.params")
if params.boolean("include_relations", default=True):
param_files.append(m13_experiments_dir / "relations.params")
if params.boolean("include_generics", default=True):
param_files.append(m13_experiments_dir / "generics.params")
if params.boolean("include_verbs_with_dynamic_prepositions", default=True):
param_files.append(m13_experiments_dir /
"events_with_dynamic_prepositions.params")
if params.boolean("include_m9_complete", default=False):
param_files.append(m13_experiments_dir / "m9_complete.params")
if params.boolean("include_m13_complete", default=False):
param_files.append(m13_experiments_dir / "m13_complete.params")
if params.boolean("include_m13_shuffled", default=False):
param_files.append(m13_experiments_dir / "m13_shuffled.params")
# This activates a special "debug" curriculum,
# which is meant to be edited in the code by a developer to do fine-grained debugging.
if params.boolean("include_debug", default=False):
param_files.append(m13_experiments_dir / "debug.params")
# If any of the param files don't exist, bail out earlier instead of making the user
# wait for the error.
for param_file in param_files:
if not param_file.exists():
raise RuntimeError(
f"Expected param file {param_file} does not exist")
for param_file in param_files:
logging.info("Running %s", param_file)
experiment_params = YAMLParametersLoader().load(param_file)
log_experiment_entry_point(experiment_params)
def integrated_pursuit_learner_experiment_test(
num_samples: Optional[int],
num_noise_objects: Optional[int],
language_generator: LanguageGenerator[
HighLevelSemanticsSituation, LinearizedDependencyTree
],
*,
params: Parameters = Parameters.empty(),
) -> Sequence[Phase1InstanceGroup]:
# pylint: disable=unused-argument
# Random Number Generator for Curriculum Use
rng = random.Random()
rng.seed(params.integer("random_seed", default=1))
# Random Chooser for Curriculum Generation
chooser = RandomChooser.for_seed(params.integer("chooser_seed", default=1))
if num_samples is None:
num_samples = 5
target_objects = [
standard_object(node.handle, node)
for node in INTEGRATED_EXPERIMENT_CURRICULUM_OBJECTS
]
target_color_objects = [
standard_object(f"{node.handle}_{color.handle}", node, added_properties=[color])
for node in INTEGRATED_EXPERIMENT_CURRICULUM_OBJECTS
for color in INTEGRATED_EXPERIMENT_COLORS
if node not in [ZUP, SPAD, DAYGIN, MAWG, TOMBUR, GLIM]
]
ordered_curriculum = [
_single_object_described_curriculum(
num_samples,
target_objects,
immutableset(immutableset()),
add_noise=False,
chooser=chooser,
block_multiple_of_same_type=True,
language_generator=language_generator,
min_samples=num_samples,
)
]
if params.boolean("include_attributes", default=True):
ordered_curriculum.append(
_single_attribute_described_curriculum(
num_samples,
target_color_objects,
immutableset(immutableset()),
add_noise=False,
chooser=chooser,
block_multiple_of_same_type=True,
language_generator=language_generator,
min_samples=num_samples,
)
)
if params.boolean("include_relations", default=True):
ordered_curriculum.append(
_prepositional_relation_described_curriculum(
num_samples,
immutableset(immutableset()),
add_noise=False,
chooser=chooser,
block_multiple_of_same_type=True,
language_generator=language_generator,
)
)
# Convert the 'from situation instances' into explicit instances this allows for
# 1) Less computation time on the learner experiment to generate the perception graphs
# 2) Allows us to shuffle the output order which we otherwise can't do
explicit_instances = [
instance for sit in ordered_curriculum for instance in sit.instances()
]
return [
ExplicitWithSituationInstanceGroup(
name="m18-integrated-learners-experiment-test",
instances=tuple(shuffle_curriculum(explicit_instances, rng=rng))
if params.boolean("shuffled", default=False)
else tuple(explicit_instances),
)
]
def integrated_experiment_entry_point(params: Parameters) -> None:
initialize_vista_pegasus_wrapper(params)
baseline_parameters = params.namespace("integrated_learners_experiment")
pursuit_resource_request_params = params.namespace(
"pursuit_resource_request")
# This code is commented out but may be used in the near future to add language ablation
# Capabilities to this curriculum.
# get the minimum and maximum accuracy of the language with the situation
# min_language_accuracy = params.floating_point("min_language_accuracy", default=0.1)
# max_language_accuracy = params.floating_point("max_language_accuracy", default=0.5)
# num_language_accuracy_increment = params.integer(
# "num_language_accuracy_increment", default=5
# )
# values_for_accuracy = np.linspace(
# min_language_accuracy, max_language_accuracy, num_language_accuracy_increment
# )
# Get if attributes or relations should be included
include_attributes = params.boolean("include_attributes", default=True)
include_relations = params.boolean("include_relations", default=True)
limit_jobs_for_category(
"pursuit_job_limit",
params.integer("num_pursuit_learners_active", default=8))
curriculum_repository_path = params.creatable_directory(
"curriculum_repository_path")
# Job to build desired curriculum(s) which our learners use
curriculum_dependencies = immutableset((
CURRICULUM_NAME_FORMAT.format(
noise=add_noise,
shuffled=shuffle,
relations=include_relations,
attributes=include_attributes,
),
run_python_on_parameters(
Locator(
CURRICULUM_NAME_FORMAT.format(
noise=add_noise,
shuffled=shuffle,
relations=include_relations,
attributes=include_attributes,
).split("-")),
generate_curriculum_script,
baseline_parameters.unify({
"train_curriculum":
Parameters.from_mapping(CURRICULUM_PARAMS).unify(
{
"add_noise": add_noise,
"shuffled": shuffle,
"include_attributes": include_attributes,
"include_relations": include_relations,
}).as_mapping()
}).unify(FIXED_PARAMETERS).unify(
{"curriculum_repository_path": curriculum_repository_path}),
depends_on=[],
),
Parameters.from_mapping(CURRICULUM_PARAMS).unify(
{
"add_noise": add_noise,
"shuffled": shuffle,
"include_attributes": include_attributes,
"include_relations": include_relations,
}),
) for add_noise in (True, False) for shuffle in (True, False))
# jobs to build experiment
for (curriculum_str, curriculum_dep,
curr_params) in curriculum_dependencies:
object_learner_type = params.string(
"object_learner.learner_type",
valid_options=["pursuit", "subset", "pbv"],
default="pursuit",
)
attribute_learner_type = params.string(
"attribute_learner.learner__type",
valid_options=["none", "pursuit", "subset"],
default="pursuit",
)
relation_learner_type = params.string(
"relation_learner.learner_type",
valid_options=["none", "pursuit", "subset"],
default="pursuit",
)
experiment_name_string = EXPERIMENT_NAME_FORMAT.format(
curriculum_name=curriculum_str.replace("-", "+"),
object_learner=object_learner_type,
attribute_learner=attribute_learner_type,
relation_learner=relation_learner_type,
)
experiment_name = Locator(experiment_name_string.split("-"))
# Note that the input parameters should include the root params and
# anything else we want.
experiment_params = baseline_parameters.unify(FIXED_PARAMETERS).unify({
"experiment":
experiment_name_string,
"experiment_group_dir":
directory_for(experiment_name),
"hypothesis_log_dir":
directory_for(experiment_name) / "hypotheses",
"learner_logging_path":
directory_for(experiment_name),
"log_learner_state":
True,
"resume_from_latest_logged_state":
True,
"load_from_curriculum_repository":
curriculum_repository_path,
"train_curriculum":
curr_params,
})
run_python_on_parameters(
experiment_name,
log_experiment_script,
experiment_params,
depends_on=[curriculum_dep],
resource_request=SlurmResourceRequest.from_parameters(
pursuit_resource_request_params) if "pursuit" in [
object_learner_type, attribute_learner_type,
relation_learner_type
] else None,
category="pursuit" if "pursuit" in [
object_learner_type, attribute_learner_type,
relation_learner_type
] else "subset",
use_pypy=True,
)
write_workflow_description()
def integrated_pursuit_learner_experiment_curriculum(
num_samples: Optional[int],
num_noise_objects: Optional[int],
language_generator: LanguageGenerator[
HighLevelSemanticsSituation, LinearizedDependencyTree
],
*,
params: Parameters = Parameters.empty(),
) -> Sequence[Phase1InstanceGroup]:
# Load Parameters
add_noise = params.boolean("add_noise", default=False)
block_multiple_of_same_type = params.boolean(
"block_multiple_of_same_type", default=True
)
include_targets_in_noise = params.boolean("include_targets_in_noise", default=False)
min_noise_objects = params.integer("min_noise_objects", default=0)
max_noise_objects = params.integer(
"max_noise_objects", default=num_noise_objects if num_noise_objects else 10
)
min_noise_relations = params.integer("min_noise_relations", default=0)
max_noise_relations = params.integer("max_noise_relations", default=5)
# This value ensure that pursuit gets at least 6 instances of any example
# As otherwise the lexicalization system might not lexicalize it
# But if there's lots of variants for noise we don't want to have thousands of examples
# As could happen combinatorially
min_samples_per_noise_object_relation_pair = (
max(
6
// (
max_noise_relations
- min_noise_relations
+ min_noise_objects
- max_noise_objects
),
1,
)
if add_noise
else 6
)
if num_samples is None:
num_samples = 50
# Random Number Generator for Curriculum Use
rng = random.Random()
rng.seed(params.integer("random_seed", default=0))
# Random Chooser for Curriculum Generation
chooser = RandomChooser.for_seed(params.integer("chooser_seed", default=0))
# Noise Elements
noise_objects_sets: ImmutableSet[ImmutableSet[TemplateObjectVariable]] = immutableset(
[
immutableset(
[
standard_object(
f"{i}_noise_object_{num}",
THING,
required_properties=[INTEGRATED_EXPERIMENT_PROP],
)
for num in range(i)
]
)
for i in range(min_noise_objects, max_noise_objects)
]
)
if noise_objects_sets.empty() or not add_noise:
noise_objects_sets = immutableset(immutableset())
target_objects = [
standard_object(node.handle, node)
for node in INTEGRATED_EXPERIMENT_CURRICULUM_OBJECTS
]
target_color_objects = [
standard_object(f"{node.handle}_{color.handle}", node, added_properties=[color])
for node in INTEGRATED_EXPERIMENT_CURRICULUM_OBJECTS
for color in INTEGRATED_EXPERIMENT_COLORS
if node not in [ZUP, SPAD, DAYGIN, MAWG, TOMBUR, GLIM]
]
# We use a max of 1 here to account for when noise values are not used as otherwise
# We'd be multiplying by 0 and cause div by 0 errors
samples_to_template_den = (
len(target_objects)
* max(len(noise_objects_sets), 1)
* max((max_noise_relations - min_noise_relations), 1)
)
ordered_curriculum = [
_single_object_described_curriculum(
num_samples,
target_objects,
noise_objects_sets,
min_noise_relations=min_noise_relations,
max_noise_relations=max_noise_relations,
add_noise=add_noise,
chooser=chooser,
samples_to_template_den=samples_to_template_den,
block_multiple_of_same_type=block_multiple_of_same_type,
language_generator=language_generator,
include_targets_in_noise=include_targets_in_noise,
min_samples=min_samples_per_noise_object_relation_pair,
)
]
if params.boolean("include_attributes", default=True):
ordered_curriculum.append(
_single_attribute_described_curriculum(
num_samples,
target_color_objects,
noise_objects_sets,
min_noise_relations=min_noise_relations,
max_noise_relations=max_noise_relations,
add_noise=add_noise,
chooser=chooser,
samples_to_template_den=samples_to_template_den,
block_multiple_of_same_type=block_multiple_of_same_type,
language_generator=language_generator,
include_targets_in_noise=include_targets_in_noise,
min_samples=min_samples_per_noise_object_relation_pair,
)
)
if params.boolean("include_relations", default=True):
ordered_curriculum.append(
_prepositional_relation_described_curriculum(
num_samples,
noise_objects_sets,
min_noise_relations=min_noise_relations,
max_noise_relations=max_noise_relations,
add_noise=add_noise,
chooser=chooser,
samples_to_template_den=samples_to_template_den,
block_multiple_of_same_type=block_multiple_of_same_type,
language_generator=language_generator,
include_targets_in_noise=include_targets_in_noise,
min_samples=min_samples_per_noise_object_relation_pair,
)
)
# Convert the 'from situation instances' into explicit instances this allows for
# 1) Less computation time on the learner experiment to generate the perception graphs
# 2) Allows us to shuffle the output order which we otherwise can't do
explicit_instances = [
instance for sit in ordered_curriculum for instance in sit.instances()
]
return [
ExplicitWithSituationInstanceGroup(
name="m18-integrated-learners-experiment",
instances=tuple(shuffle_curriculum(explicit_instances, rng=rng))
if params.boolean("shuffled", default=False)
else tuple(explicit_instances),
)
]
def learner_factory_from_params(
params: Parameters,
graph_logger: Optional[HypothesisLogger],
language_mode: LanguageMode = LanguageMode.ENGLISH,
) -> Callable[[], TopLevelLanguageLearner]: # type: ignore
learner_type = params.string(
"learner",
[
"pursuit",
"object-subset",
"preposition-subset",
"attribute-subset",
"verb-subset",
"integrated-learner",
"integrated-learner-recognizer-without-generics",
"integrated-learner-recognizer",
"pursuit-gaze",
"integrated-object-only",
"integrated-learner-params",
"integrated-pursuit-attribute-only",
],
)
beam_size = params.positive_integer("beam_size", default=10)
rng = random.Random()
rng.seed(0)
perception_generator = GAILA_PHASE_1_PERCEPTION_GENERATOR
objects = [YOU_HACK, ME_HACK]
objects.extend(PHASE_1_CURRICULUM_OBJECTS)
# Eval hack! This is specific to the Phase 1 ontology
object_recognizer = ObjectRecognizer.for_ontology_types(
objects,
determiners=ENGLISH_DETERMINERS,
ontology=GAILA_PHASE_1_ONTOLOGY,
language_mode=language_mode,
perception_generator=perception_generator,
)
if learner_type == "pursuit":
return lambda: ObjectPursuitLearner.from_parameters(
params.namespace("pursuit"), graph_logger=graph_logger)
elif learner_type == "pursuit-gaze":
return lambda: IntegratedTemplateLearner(
object_learner=PursuitObjectLearnerNew(
learning_factor=0.05,
graph_match_confirmation_threshold=0.7,
lexicon_entry_threshold=0.7,
rng=rng,
smoothing_parameter=0.002,
ontology=GAILA_PHASE_2_ONTOLOGY,
language_mode=language_mode,
rank_gaze_higher=True,
),
attribute_learner=SubsetAttributeLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
relation_learner=SubsetRelationLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
action_learner=SubsetVerbLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
)
elif learner_type == "object-subset":
return lambda: SubsetObjectLearner(ontology=GAILA_PHASE_1_ONTOLOGY,
language_mode=LanguageMode.ENGLISH)
elif learner_type == "attribute-subset":
return lambda: SubsetAttributeLearner(
ontology=GAILA_PHASE_1_ONTOLOGY,
object_recognizer=object_recognizer,
language_mode=LanguageMode.ENGLISH,
)
elif learner_type == "preposition-subset":
return lambda: SubsetPrepositionLearner(
# graph_logger=graph_logger,
object_recognizer=object_recognizer,
ontology=GAILA_PHASE_1_ONTOLOGY,
language_mode=LanguageMode.ENGLISH,
)
elif learner_type == "verb-subset":
return lambda: SubsetVerbLearner(
ontology=GAILA_PHASE_1_ONTOLOGY,
object_recognizer=object_recognizer,
language_mode=LanguageMode.ENGLISH,
)
elif learner_type == "integrated-learner":
return lambda: IntegratedTemplateLearner(
object_learner=SubsetObjectLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
attribute_learner=SubsetAttributeLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
relation_learner=SubsetRelationLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
action_learner=SubsetVerbLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
functional_learner=FunctionalLearner(language_mode=language_mode),
)
elif learner_type == "integrated-learner-recognizer":
return lambda: IntegratedTemplateLearner(
object_learner=ObjectRecognizerAsTemplateLearner(
object_recognizer=object_recognizer,
language_mode=language_mode),
attribute_learner=SubsetAttributeLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
relation_learner=SubsetRelationLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
action_learner=SubsetVerbLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
functional_learner=FunctionalLearner(language_mode=language_mode),
generics_learner=SimpleGenericsLearner(),
)
elif learner_type == "ic":
return lambda: IntegratedTemplateLearner(
object_learner=ObjectRecognizerAsTemplateLearner(
object_recognizer=object_recognizer,
language_mode=language_mode),
attribute_learner=SubsetAttributeLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
relation_learner=SubsetRelationLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
action_learner=SubsetVerbLearnerNew(
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
),
functional_learner=FunctionalLearner(language_mode=language_mode),
)
elif learner_type == "integrated-object-only":
object_learner_type = params.string(
"object_learner_type",
valid_options=["subset", "pbv", "pursuit"],
default="subset",
)
if params.has_namespace("learner_params"):
learner_params = params.namespace("learner_params")
else:
learner_params = params.empty(namespace_prefix="learner_params")
object_learner_factory: Callable[[], TemplateLearner]
if object_learner_type == "subset":
def subset_factory() -> SubsetObjectLearnerNew:
return SubsetObjectLearnerNew( # type: ignore
ontology=GAILA_PHASE_2_ONTOLOGY,
beam_size=beam_size,
language_mode=language_mode,
)
object_learner_factory = subset_factory
elif object_learner_type == "pbv":
def pbv_factory() -> ProposeButVerifyObjectLearner:
return ProposeButVerifyObjectLearner.from_params( # type: ignore
learner_params)
object_learner_factory = pbv_factory
elif object_learner_type == "pursuit":
def pursuit_factory() -> PursuitObjectLearnerNew:
return PursuitObjectLearnerNew( # type: ignore
learning_factor=learner_params.floating_point(
"learning_factor"),
graph_match_confirmation_threshold=learner_params.
floating_point("graph_match_confirmation_threshold"),
lexicon_entry_threshold=learner_params.floating_point(
"lexicon_entry_threshold"),
rng=rng,
smoothing_parameter=learner_params.floating_point(
"smoothing_parameter"),
ontology=GAILA_PHASE_2_ONTOLOGY,
language_mode=language_mode,
)
object_learner_factory = pursuit_factory
else:
raise RuntimeError(
f"Invalid Object Learner Type Selected: {learner_type}")
return lambda: IntegratedTemplateLearner(object_learner=
object_learner_factory())
elif learner_type == "integrated-learner-params":
object_learner = build_object_learner_factory( # type:ignore
params.namespace_or_empty("object_learner"), beam_size,
language_mode)
attribute_learner = build_attribute_learner_factory( # type:ignore
params.namespace_or_empty("attribute_learner"), beam_size,
language_mode)
relation_learner = build_relation_learner_factory( # type:ignore
params.namespace_or_empty("relation_learner"), beam_size,
language_mode)
action_learner = build_action_learner_factory( # type:ignore
params.namespace_or_empty("action_learner"), beam_size,
language_mode)
plural_learner = build_plural_learner_factory( # type:ignore
params.namespace_or_empty("plural_learner"), beam_size,
language_mode)
return lambda: IntegratedTemplateLearner(
object_learner=object_learner,
attribute_learner=attribute_learner,
relation_learner=relation_learner,
action_learner=action_learner,
functional_learner=FunctionalLearner(language_mode=language_mode)
if params.boolean("include_functional_learner", default=True) else
None,
generics_learner=SimpleGenericsLearner() if params.boolean(
"include_generics_learner", default=True) else None,
plural_learner=plural_learner,
suppress_error=params.boolean("suppress_error", default=True),
)
elif learner_type == "integrated-pursuit-attribute-only":
return lambda: IntegratedTemplateLearner(
object_learner=ObjectRecognizerAsTemplateLearner(
object_recognizer=object_recognizer,
language_mode=language_mode),
attribute_learner=PursuitAttributeLearnerNew(
learning_factor=0.05,
graph_match_confirmation_threshold=0.7,
lexicon_entry_threshold=0.7,
rng=rng,
smoothing_parameter=0.002,
rank_gaze_higher=False,
ontology=GAILA_PHASE_1_ONTOLOGY,
language_mode=language_mode,
),
)
else:
raise RuntimeError("can't happen")
def main(params: Parameters):
viz = SituationVisualizer()
# try to get the directory for rendering for an experiment
adam_root = params.existing_directory("adam_root")
root_output_directory = params.optional_creatable_directory(
"experiment_group_dir")
if root_output_directory is not None:
m9_experiments_dir = adam_root / "parameters" / "experiments" / "m9"
param_files: List[Path] = []
if params.boolean("include_objects"):
param_files.append(m9_experiments_dir / "objects.params")
if params.boolean("include_attributes"):
param_files.append(m9_experiments_dir / "attributes.params")
if params.boolean("include_relations"):
param_files.append(m9_experiments_dir / "relations.params")
if params.boolean("include_events"):
param_files.append(m9_experiments_dir / "events.params")
# This activates a special "debug" curriculum,
# which is meant to be edited in the code by a developer to do fine-grained debugging.
if params.boolean("include_debug", default=False):
param_files.append(m9_experiments_dir / "debug.params")
# loop over all experiment params files
for param_file in param_files:
experiment_params = YAMLParametersLoader().load(param_file)
if "curriculum" in experiment_params:
# get the experiment curriculum list (if there is one)
curriculum = curriculum_from_params(experiment_params)[0]
directory_name = experiment_params.string(
"experiment") + "/renders"
if not os.path.isdir(root_output_directory / directory_name):
os.mkdir(root_output_directory / directory_name)
for instance_group in curriculum:
try:
make_scenes(
params,
[instance_group],
root_output_directory / directory_name,
viz,
)
except RuntimeError as err:
print(f"uncaught exception: {err}")
else:
# render phase 1 scenes:
root_output_directory = params.optional_creatable_directory(
"screenshot_directory")
assert root_output_directory is not None
if not os.path.isdir(root_output_directory):
os.mkdir(root_output_directory)
for idx, instance_group in enumerate(
build_curriculum(None, None,
GAILA_PHASE_1_LANGUAGE_GENERATOR)):
# do any filtering here
if instance_group.name() in EXCLUDED_CURRICULA:
continue
directory_name = f"{idx:03}-{instance_group.name()}"
if not os.path.isdir(root_output_directory / directory_name):
os.mkdir(root_output_directory /
directory_name) # type: ignore
# then call some function from make_scenes.py to run the curriculum
make_scenes(params, [instance_group],
root_output_directory / directory_name, viz)
def curriculum_from_params(params: Parameters,
language_mode: LanguageMode = LanguageMode.ENGLISH):
str_to_train_test_curriculum: Mapping[str, Tuple[
CURRICULUM_BUILDER, Optional[CURRICULUM_BUILDER]]] = {
"m6-deniz": (make_m6_curriculum, None),
"each-object-by-itself": (
build_each_object_by_itself_curriculum_train,
build_each_object_by_itself_curriculum_test,
),
"pursuit": (
build_pursuit_curriculum,
build_each_object_by_itself_curriculum_test,
),
"m6-preposition": (build_m6_prepositions_curriculum, None),
"m9-objects": (build_gaila_phase1_object_curriculum, None),
"m9-attributes": (build_gaila_phase1_attribute_curriculum, None),
"chinese-classifiers": (build_classifier_curriculum, None),
"m9-relations": (build_gaila_phase1_relation_curriculum, None),
"m9-events": (build_gaila_phase1_verb_curriculum, None),
"m9-debug":
(build_debug_curriculum_train, build_debug_curriculum_test),
"m9-complete": (build_gaila_phase_1_curriculum, None),
"m13-imprecise-size": (make_imprecise_size_curriculum, None),
"m13-imprecise-temporal":
(make_imprecise_temporal_descriptions, None),
"m13-subtle-verb-distinction":
(make_subtle_verb_distinctions_curriculum, None),
"m13-object-restrictions":
(build_functionally_defined_objects_curriculum, None),
"m13-functionally-defined-objects": (
build_functionally_defined_objects_train_curriculum,
build_functionally_defined_objects_curriculum,
),
"m13-generics": (build_generics_curriculum, None),
"m13-complete": (build_gaila_m13_curriculum, None),
"m13-verbs-with-dynamic-prepositions": (
make_verb_with_dynamic_prepositions_curriculum,
None,
),
"m13-shuffled": (build_m13_shuffled_curriculum,
build_gaila_m13_curriculum),
"m13-relations": (make_prepositions_curriculum, None),
"actions-and-generics-curriculum":
(build_actions_and_generics_curriculum, None),
"m15-object-noise-experiments": (
build_object_learner_experiment_curriculum_train,
build_each_object_by_itself_curriculum_test,
),
"m18-integrated-learners-experiment": (
integrated_pursuit_learner_experiment_curriculum,
integrated_pursuit_learner_experiment_test,
),
}
curriculum_name = params.string("curriculum",
str_to_train_test_curriculum.keys())
language_generator = (
integrated_experiment_language_generator(language_mode)
if curriculum_name == "m18-integrated-learners-experiment" else
phase2_language_generator(language_mode))
if params.has_namespace("pursuit-curriculum-params"):
pursuit_curriculum_params = params.namespace(
"pursuit-curriculum-params")
else:
pursuit_curriculum_params = Parameters.empty()
use_path_instead_of_goal = params.boolean("use-path-instead-of-goal",
default=False)
(training_instance_groups,
test_instance_groups) = str_to_train_test_curriculum[curriculum_name]
num_samples = params.optional_positive_integer("num_samples")
# We need to be able to accept 0 as the number of noise objects but optional_integer doesn't currently
# support specifying a range of acceptable values: https://github.com/isi-vista/vistautils/issues/142
num_noise_objects = params.optional_integer("num_noise_objects")
if curriculum_name == "pursuit":
return (
training_instance_groups(
num_samples,
num_noise_objects,
language_generator,
pursuit_curriculum_params=pursuit_curriculum_params,
),
test_instance_groups(num_samples, num_noise_objects,
language_generator)
if test_instance_groups else [],
)
# optional argument to use path instead of goal
elif use_path_instead_of_goal and curriculum_name in [
"m13-complete",
"m13-shuffled",
"m13-verbs-with-dynamic-prepositions",
]:
return (
training_instance_groups(
num_samples,
num_noise_objects,
language_generator,
use_path_instead_of_goal,
),
test_instance_groups(num_samples, num_noise_objects,
language_generator)
if test_instance_groups else [],
)
elif curriculum_name in (
"m15-object-noise-experiments",
"m18-integrated-learners-experiment",
):
return (
training_instance_groups(
num_samples,
num_noise_objects,
language_generator,
params=params.namespace_or_empty("train_curriculum"),
),
test_instance_groups(
5,
0,
language_generator,
params=params.namespace_or_empty("test_curriculum"),
) if test_instance_groups else [],
)
return (
training_instance_groups(num_samples, num_noise_objects,
language_generator),
test_instance_groups(num_samples, num_noise_objects,
language_generator)
if test_instance_groups else [],
)
