Merge pull request #11 from fish-quant/release/v0.1.0

v0.1.0

.gitignore

@@ -0,0 +1,20 @@
+# Dot files
+.idea/
+.DS_Store
+
+# Packaging related files
+MANIFEST
+build/
+dist/
+big_fish.egg-info/
+
+# Notebooks
+notebooks/old
+notebooks/.ipynb_checkpoints
+
+# Data
+data/input/*
+data/output/*
+
+# Cache
+__pycache__/

Makefile

@@ -0,0 +1,2 @@
+init:
+    pip install -r requirements.txt

bigfish/classification/__init__.py

@@ -0,0 +1,17 @@
+# -*- coding: utf-8 -*-
+
+"""
+The bigfish.classification module includes models to classify the localization
+patterns of the RNA.
+"""
+
+# from .squeezenet import SqueezeNet0
+from .features import get_features, get_features_name
+
+# ### Load models ###
+
+_features = ["get_features", "get_features_name"]
+
+# _squeezenet = ["SqueezeNet0"]
+
+__all__ = _features

bigfish/classification/base.py

@@ -0,0 +1,101 @@
+# -*- coding: utf-8 -*-
+
+"""
+General classes and methods to use the models.
+"""
+
+from abc import ABCMeta, abstractmethod
+
+from tensorflow.python.keras.optimizers import (Adam, Adadelta, Adagrad,
+                                                Adamax, SGD)
+
+
+# ### General models ###
+
+class BaseModel(object, metaclass=ABCMeta):
+
+    def __init__(self):
+        pass
+
+    @abstractmethod
+    def fit(self, train_data, train_label, validation_data, validation_label,
+            batch_size, nb_epochs):
+        pass
+
+    @abstractmethod
+    def fit_generator(self, train_generator, validation_generator, nb_epochs,
+                      nb_workers=1, multiprocessing=False):
+        pass
+
+    @abstractmethod
+    def predict(self, data, return_probability=False):
+        pass
+
+    @abstractmethod
+    def predict_generator(self, generator, return_probability=False,
+                          nb_workers=1, multiprocessing=False):
+        pass
+
+    @abstractmethod
+    def predict_probability(self, data):
+        pass
+
+    @abstractmethod
+    def predict_probability_generator(self, generator,
+                                      nb_workers=1, multiprocessing=False):
+        pass
+
+    @abstractmethod
+    def evaluate(self, data, label):
+        pass
+
+    @abstractmethod
+    def evaluate_generator(self, generator, nb_workers=1,
+                           multiprocessing=False):
+        pass
+
+
+# ### optimizer ###
+
+def get_optimizer(optimizer_name="adam", **kwargs):
+    """Instantiate the optimizer.
+
+    Parameters
+    ----------
+    optimizer_name : str
+        Name of the optimizer to use.
+
+    Returns
+    -------
+    optimizer : tf.keras.optimizers
+        Optimizer instance used in the model.
+
+    """
+    # TODO use tensorflow optimizer
+    if optimizer_name == "adam":
+        optimizer = Adam(**kwargs)
+    elif optimizer_name == "adadelta":
+        optimizer = Adadelta(**kwargs)
+    elif optimizer_name == "adagrad":
+        optimizer = Adagrad(**kwargs)
+    elif optimizer_name == "adamax":
+        optimizer = Adamax(**kwargs)
+    elif optimizer_name == "sgd":
+        optimizer = SGD(**kwargs)
+    else:
+        raise ValueError("Instead of {0}, optimizer must be chosen among "
+                         "['adam', 'adadelta', 'adagrad', adamax', sgd']."
+                         .format(optimizer_name))
+
+    return optimizer
+
+
+
+
+#print(globals())
+#print()
+#print(globals()["BaseModel"])
+#print()
+#print(locals())
+#print()
+#print(BaseModel.__subclasses__())