Merge pull request #12 from NxNiki/dev

add analysis

.gitattributes

@@ -0,0 +1 @@
+results/**/*.html filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -35,5 +35,10 @@ src/brain_decoding/__pycache__/
 data/
 ._data
 config/*.yaml
-results/
+
+results/**/*.npy
+results/**/*.png
+results/**/*.tar
+!results/**/*.html
+
 wandb/

.run/run_twilight_merge.run.xml

@@ -0,0 +1,25 @@
+<component name="ProjectRunConfigurationManager">
+  <configuration default="false" name="run_twilight_merge" type="PythonConfigurationType" factoryName="Python">
+    <module name="brain_decoding" />
+    <option name="ENV_FILES" value="" />
+    <option name="INTERPRETER_OPTIONS" value="" />
+    <option name="PARENT_ENVS" value="true" />
+    <envs>
+      <env name="PYTHONUNBUFFERED" value="1" />
+    </envs>
+    <option name="SDK_HOME" value="" />
+    <option name="SDK_NAME" value="movie_decoding" />
+    <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+    <option name="IS_MODULE_SDK" value="false" />
+    <option name="ADD_CONTENT_ROOTS" value="true" />
+    <option name="ADD_SOURCE_ROOTS" value="true" />
+    <option name="SCRIPT_NAME" value="$PROJECT_DIR$/src/scripts/run_model_twilight_merge.py" />
+    <option name="PARAMETERS" value="" />
+    <option name="SHOW_COMMAND_LINE" value="false" />
+    <option name="EMULATE_TERMINAL" value="false" />
+    <option name="MODULE_MODE" value="false" />
+    <option name="REDIRECT_INPUT" value="false" />
+    <option name="INPUT_FILE" value="" />
+    <method v="2" />
+  </configuration>
+</component>

.run/run_twilight_vs_24.run.xml

@@ -0,0 +1,25 @@
+<component name="ProjectRunConfigurationManager">
+  <configuration default="false" name="run_twilight_vs_24" type="PythonConfigurationType" factoryName="Python">
+    <module name="brain_decoding" />
+    <option name="ENV_FILES" value="" />
+    <option name="INTERPRETER_OPTIONS" value="" />
+    <option name="PARENT_ENVS" value="true" />
+    <envs>
+      <env name="PYTHONUNBUFFERED" value="1" />
+    </envs>
+    <option name="SDK_HOME" value="" />
+    <option name="SDK_NAME" value="movie_decoding" />
+    <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+    <option name="IS_MODULE_SDK" value="false" />
+    <option name="ADD_CONTENT_ROOTS" value="true" />
+    <option name="ADD_SOURCE_ROOTS" value="true" />
+    <option name="SCRIPT_NAME" value="$PROJECT_DIR$/src/scripts/run_model_twilight_vs_24.py" />
+    <option name="PARAMETERS" value="" />
+    <option name="SHOW_COMMAND_LINE" value="false" />
+    <option name="EMULATE_TERMINAL" value="false" />
+    <option name="MODULE_MODE" value="false" />
+    <option name="REDIRECT_INPUT" value="false" />
+    <option name="INPUT_FILE" value="" />
+    <method v="2" />
+  </configuration>
+</component>

src/brain_decoding/config/ensure_list.py

@@ -0,0 +1,46 @@
+from pydantic import BaseModel
+
+
+class Config(BaseModel):
+    class Config:
+        # arbitrary_types_allowed = True
+        extra = "allow"  # Allow arbitrary attributes
+
+    _list_fields = set()  # A set to track which fields should be treated as lists
+
+    def ensure_list(self, name: str):
+        value = getattr(self, name, None)
+        if value is not None and not isinstance(value, list):
+            setattr(self, name, [value])
+        # Mark the field to always be treated as a list
+        self._list_fields.add(name)
+
+    def __setattr__(self, name, value):
+        if name in self._list_fields and not isinstance(value, list):
+            # Automatically convert to a list if it's in the list fields
+            value = [value]
+        super().__setattr__(name, value)
+
+
+class SupConfig(Config):
+    pass
+
+
+# Example usage
+config = SupConfig()
+
+# Dynamically adding attributes
+config.param1 = "a"
+
+# Ensuring param1 is a list
+config.ensure_list("param1")
+print(config.param1)  # Output: ['a']
+
+# Assigning new value to param1
+config.param1 = "ab"
+print(config.param1)  # Output: ['ab'] gets automatically converted to ['ab']
+
+# Adding another parameter and ensuring it's a list
+config.ensure_list("param2")
+config.param2 = 123
+print(config.param2)  # Output: [123]

src/brain_decoding/config/export_extra.py

@@ -0,0 +1,64 @@
+from typing import Any, Dict, Set
+
+from pydantic import BaseModel, Field
+
+
+class BaseConfig(BaseModel):
+    class Config:
+        extra = "allow"  # Allow arbitrary attributes
+
+    def __init__(self, **data: Any) -> None:
+        super().__init__(**data)
+        self.__dict__["_list_fields"]: Set[str] = set()
+        self.__dict__["_alias"]: Dict[str, str] = {}
+
+    def __getitem__(self, key: str) -> Any:
+        return getattr(self, key)
+
+    def __setitem__(self, key: str, value: Any):
+        setattr(self, key, value)
+
+    def __getattr__(self, name):
+        """Handles alias access and custom parameters."""
+        if name in self._alias:
+            return getattr(self, self._alias[name])
+
+    def __setattr__(self, name, value):
+        """Handles alias assignment, field setting, or adding to _param."""
+        if name in self._alias:
+            name = self._alias[name]
+        if name in self._list_fields and not isinstance(value, list):
+            value = [value]
+        super().__setattr__(name, value)
+
+    def __contains__(self, key: str) -> bool:
+        return hasattr(self, key)
+
+    def __repr__(self):
+        attrs = {k: v for k, v in self.__dict__.items() if not k.startswith("_")}
+        attr_str = "\n".join(f"    {key}: {value!r}" for key, value in attrs.items())
+        return f"{self.__class__.__name__}(\n{attr_str}\n)"
+
+    def set_alias(self, name: str, alias: str) -> None:
+        self.__dict__["_alias"][alias] = name
+
+    def ensure_list(self, name: str):
+        """Mark the field to always be treated as a list"""
+        value = getattr(self, name, None)
+        if value is not None and not isinstance(value, list):
+            setattr(self, name, [value])
+        self._list_fields.add(name)
+
+
+class Foo(BaseConfig):
+    a: int = 1
+
+    class Config:
+        extra = "allow"
+
+
+print(Foo(**{"a": 1, "b": 2}).model_dump())  # == {'a': 1, 'b': 2}
+
+foo = Foo()
+foo.b = 2
+print(foo.model_dump())

src/brain_decoding/config/file_path.py

@@ -6,3 +6,7 @@
 SURROGATE_FILE_PATH = ROOT_PATH / "data/surrogate_windows"
 CONFIG_FILE_PATH = ROOT_PATH / "config"
 RESULT_PATH = ROOT_PATH / "results"
+MOVIE24_LABEL_PATH = f"{DATA_PATH}/8concepts_merged.npy"
+TWILIGHT_LABEL_PATH = f"{DATA_PATH}/twilight_concepts.npy"
+TWILIGHT_MERGE_LABEL_PATH = f"{DATA_PATH}/twilight_concepts_merged.npy"
+MOVIE_LABEL_TWILIGHT_VS_24 = f"{DATA_PATH}/twilight_vs_24.npy"

scripts/save_config.py → src/brain_decoding/config/save_config.py

@@ -3,6 +3,8 @@
 Custom parameters can be added to any of the three fields of config (experiment, model, data).
 """
 
+from torch import nn
+
 from brain_decoding.config.config import ExperimentConfig, PipelineConfig
 from brain_decoding.config.file_path import CONFIG_FILE_PATH, DATA_PATH, RESULT_PATH
 
@@ -18,7 +20,7 @@
 config.model.lr_drop = 50
 config.model.validation_step = 10
 config.model.early_stop = 75
-config.model.num_labels = 8
+config.model.num_labels = 18  # 8 for 24, 18 for twilight
 config.model.merge_label = True
 config.model.img_embedding_size = 192
 config.model.hidden_size = 256
@@ -27,6 +29,7 @@
 config.model.patch_size = (1, 5)
 config.model.intermediate_size = 192 * 2
 config.model.classifier_proj_size = 192
+config.model.train_loss = nn.BCEWithLogitsLoss(reduction="none")
 
 config.experiment.seed = 42
 config.experiment.use_spike = True
@@ -44,8 +47,8 @@
 config.experiment.use_shuffle_diagnostic = True
 config.experiment.testing_mode = False  # in testing mode, a maximum of 1e4 clusterless data will be loaded.
 config.experiment.model_aggregate_type = "sum"
-config.experiment.train_phases = ["movie_1"]
-config.experiment.test_phases = ["sleep_2"]
+config.experiment.train_phases = ["twilight_1"]
+config.experiment.test_phases = ["sleep_1"]
 config.experiment.compute_accuracy = False
 
 config.experiment.ensure_list("train_phases")
@@ -61,6 +64,8 @@
 config.data.spike_data_sd_inference = 3.5
 config.data.model_aggregate_type = "sum"
 config.data.movie_label_path = str(DATA_PATH / "8concepts_merged.npy")
+config.data.movie_label_sr = 1
 config.data.movie_sampling_rate = 30
+config.data.filter_low_occurrence_samples = True
 
 # config.export_config(CONFIG_FILE_PATH)

src/brain_decoding/dataloader/clusterless_clean.py

@@ -1,6 +1,7 @@
 import glob
 import os
 import re
+from typing import List, Union
 
 import numpy as np
 import pandas as pd
@@ -16,6 +17,11 @@ def __init__(self, operator: str, threshold: int):
         self.threshold = threshold
 
 
+def sort_file_name(filenames: str) -> List[Union[int, str]]:
+    """Extract the numeric part of the filename and use it as the sort key"""
+    return [int(x) if x.isdigit() else x for x in re.findall(r"\d+|\D+", filenames)]
+
+
 def find_true_indices(mask, op_thresh: OpThresh = None):
     """
     Returns an nx3 matrix containing start, end, and length of all true samples in a 1D boolean mask.
@@ -161,14 +167,10 @@ def load_data_from_bundle(clu_bundle_filepaths):
 
 
 def get_oneshot_clean(patient_number, desired_samplerate, mode, category="recall", phase=None, version="notch"):
-    def sort_filename(filename):
-        """Extract the numeric part of the filename and use it as the sort key"""
-        return [int(x) if x.isdigit() else x for x in re.findall(r"\d+|\D+", filename)]
-
     # folder contains the clustless data, I saved the folder downloaded from the drive as '562/clustless_raw'
     spike_path = f"/mnt/SSD2/yyding/Datasets/neuron/spike_data/{patient_number}/raw_{mode}/"
     spike_files = glob.glob(os.path.join(spike_path, "*.csv"))
-    spike_files = sorted(spike_files, key=sort_filename)
+    spike_files = sorted(spike_files, key=sort_file_name)
 
     for bundle in range(0, len(spike_files), 8):
         df = load_data_from_bundle(spike_files[bundle : bundle + 8])