diff --git a/mhcflurry/batch_generator.py b/mhcflurry/batch_generator.py
index ca51207201b98d83040a1ecefe37e35c23310666..96c1a15d7712f19d53fc92d73c487c4f9e48fa3e 100644
--- a/mhcflurry/batch_generator.py
+++ b/mhcflurry/batch_generator.py
@@ -80,9 +80,9 @@ class BatchPlan(object):
("Batch %5d: " % i) + ", ".join(
"{key}[{value}]".format(key=key, value=value)
for (key, value) in label_counts.iteritems()))
- if i == 5:
+ if i == 5 and len(self.batch_compositions) > i + 3:
lines.append("...")
- i = len(self.batch_compositions) - 4
+ i = len(self.batch_compositions) - i + 1
i += 1
indent_spaces = " " * indent
diff --git a/mhcflurry/class1_presentation_neural_network.py b/mhcflurry/class1_presentation_neural_network.py
index 26742d45a6251566a150e8e6d7df1c0f16061cbf..6ac7b1a220981410193e810a85a5f756305f0d0a 100644
--- a/mhcflurry/class1_presentation_neural_network.py
+++ b/mhcflurry/class1_presentation_neural_network.py
@@ -129,7 +129,7 @@ class Class1PresentationNeuralNetwork(object):
output_dim=1029,
input_length=self.hyperparameters['max_alleles'],
trainable=False,
- mask_zero=True)(input_alleles)
+ mask_zero=False)(input_alleles)
allele_flat = allele_representation
@@ -427,7 +427,8 @@ class Class1PresentationNeuralNetwork(object):
allele_representations_hash = self.set_allele_representations(
allele_representations)
self.network.compile(
- loss=[affinities_loss.loss, mms_loss.loss, ZeroLoss.loss],
+ loss=[affinities_loss.keras_wrapped(), mms_loss.keras_wrapped(), ZeroLoss().keras_wrapped()],
+ #loss_weights=[1.0, 1.0, 1.0],
optimizer=self.hyperparameters['optimizer'])
if self.hyperparameters['learning_rate'] is not None:
K.set_value(
@@ -545,6 +546,10 @@ class Class1PresentationNeuralNetwork(object):
for (key, value) in fit_history.history.items():
fit_info[key].extend(value)
+ if numpy.isnan(fit_info['loss'][-1]):
+ import ipdb ; ipdb.set_trace()
+ raise ValueError("NaN loss")
+
# Print progress no more often than once every few seconds.
if progress_print_interval is not None and (
not last_progress_print or (
diff --git a/mhcflurry/custom_loss.py b/mhcflurry/custom_loss.py
index 65a6586b54b394f972596bc5d2e05df8f52a15a5..e6f72264ae5bc1a5f4ba4aa6d33fcf367cc948a4 100644
--- a/mhcflurry/custom_loss.py
+++ b/mhcflurry/custom_loss.py
@@ -57,6 +57,12 @@ class Loss(object):
def __str__(self):
return "<Loss: %s>" % self.name
+ def keras_wrapped(self, reduction="sum_over_batch_size"):
+ from keras.losses import LossFunctionWrapper
+ return LossFunctionWrapper(
+ self.loss, reduction=reduction, name=self.name)
+
+
class StandardKerasLoss(Loss):
"""
@@ -132,6 +138,7 @@ class MSEWithInequalities(Loss):
# We always delay import of Keras so that mhcflurry can be imported
# initially without tensorflow debug output, etc.
from keras import backend as K
+ import tensorflow as tf
# Handle (=) inequalities
diff1 = y_pred - y_true
@@ -153,8 +160,7 @@ class MSEWithInequalities(Loss):
K.sum(K.square(diff1)) +
K.sum(K.square(diff2)) +
K.sum(K.square(diff3))) / K.cast(K.shape(y_pred)[0], "float32")
-
- return result
+ return tf.where(tf.is_nan(result), tf.zeros_like(result), result)
class MSEWithInequalitiesAndMultipleOutputs(Loss):
@@ -281,7 +287,8 @@ class ZeroLoss(Loss):
@staticmethod
def loss(y_true, y_pred):
- return 0.0
+ import keras.backend as K
+ return K.constant(0.0)
def check_shape(name, arr, expected_shape):
diff --git a/test/test_class1_presentation_predictor.py b/test/test_class1_presentation_predictor.py
index 9c04c618116342846cba659ab379b97e1a52c281..7cf4480c3499649d494a7456048587412195764e 100644
--- a/test/test_class1_presentation_predictor.py
+++ b/test/test_class1_presentation_predictor.py
@@ -24,6 +24,7 @@ import sys
import copy
import os
import tempfile
+import pickle
from numpy.testing import assert_, assert_equal, assert_allclose, assert_array_equal
from nose.tools import assert_greater, assert_less
@@ -90,11 +91,14 @@ def teardown():
cleanup()
-def test_basic():
+def Xtest_basic():
affinity_predictor = PAN_ALLELE_PREDICTOR_NO_MASS_SPEC
models = []
for affinity_network in affinity_predictor.class1_pan_allele_models:
- presentation_network = Class1PresentationNeuralNetwork()
+ presentation_network = Class1PresentationNeuralNetwork(
+ max_epochs=1,
+ learning_rate=0.0001,
+ batch_generator_batch_size=256)
presentation_network.load_from_class1_neural_network(affinity_network)
print(presentation_network.network.get_config())
models.append(presentation_network)
@@ -112,13 +116,13 @@ def test_basic():
df["tightest_affinity"] = df.min(1)
df["tightest_allele"] = df.idxmin(1)
- df2 = predictor.predict_to_dataframe(
+ # Test untrained predictor gives predictions that match the affinity
+ # predictor
+ df_predictor = predictor.predict_to_dataframe(
peptides=df.index.values,
alleles=alleles)
merged_df = pandas.merge(
- df, df2.set_index("peptide"), left_index=True, right_index=True)
-
- #import ipdb ; ipdb.set_trace()
+ df, df_predictor.set_index("peptide"), left_index=True, right_index=True)
assert_allclose(
merged_df["tightest_affinity"], merged_df["affinity"], rtol=1e-5)
@@ -126,18 +130,59 @@ def test_basic():
merged_df["tightest_affinity"], to_ic50(merged_df["score"]), rtol=1e-5)
assert_array_equal(merged_df["tightest_allele"], merged_df["allele"])
+ # Test saving and loading
models_dir = tempfile.mkdtemp("_models")
print(models_dir)
predictor.save(models_dir)
predictor2 = Class1PresentationPredictor.load(models_dir)
- df3 = predictor2.predict_to_dataframe(
+ df_predictor2 = predictor2.predict_to_dataframe(
peptides=df.index.values,
alleles=alleles)
-
- assert_array_equal(df2.values, df3.values)
-
- # TODO: test fitting, saving, and loading
+ assert_array_equal(df_predictor.values, df_predictor2.values)
+
+ # Test pickling
+ predictor3 = pickle.loads(
+ pickle.dumps(predictor, protocol=pickle.HIGHEST_PROTOCOL))
+ predictor4 = pickle.loads(
+ pickle.dumps(predictor2, protocol=pickle.HIGHEST_PROTOCOL))
+ df_predictor3 = predictor3.predict_to_dataframe(
+ peptides=df.index.values,
+ alleles=alleles)
+ df_predictor4 = predictor4.predict_to_dataframe(
+ peptides=df.index.values,
+ alleles=alleles)
+ assert_array_equal(df_predictor.values, df_predictor3.values)
+ assert_array_equal(df_predictor.values, df_predictor4.values)
+
+ # Test that fitting a model changes the predictor but not the original model
+ train_df = pandas.DataFrame({
+ "peptide": random_peptides(256),
+ })
+ train_df["allele"] = "HLA-A*02:20"
+ train_df["experiment"] = "experiment1"
+ train_df["label"] = train_df.peptide.str.match("^[KSEQ]").astype("float32")
+ allele_encoding = MultipleAlleleEncoding(
+ experiment_names=train_df.experiment.values,
+ experiment_to_allele_list={
+ "experiment1": ["HLA-A*02:20", "HLA-A*02:01"],
+ },
+ allele_to_sequence=predictor4.allele_to_sequence)
+ model = predictor4.models[0]
+ new_predictor = Class1PresentationPredictor(
+ models=[model],
+ allele_to_sequence=predictor4.allele_to_sequence)
+ train_df["original_score"] = new_predictor.predict(
+ train_df.peptide.values, alleles=["HLA-A*02:20"])
+ model.fit(
+ peptides=train_df.peptide.values,
+ labels=train_df.label.values,
+ allele_encoding=allele_encoding)
+ train_df["updated_score"] = new_predictor.predict(
+ train_df.peptide.values,
+ alleles=["HLA-A*02:20"])
+ print(train_df)
+ import ipdb ; ipdb.set_trace()
def scramble_peptide(peptide):
@@ -402,7 +447,8 @@ def Xtest_real_data_multiallelic_refinement(max_epochs=10):
import ipdb ; ipdb.set_trace()
-def Xtest_synthetic_allele_refinement_with_affinity_data(max_epochs=10):
+def test_synthetic_allele_refinement_with_affinity_data(
+ max_epochs=10, include_affinities=False):
refine_allele = "HLA-C*01:02"
alleles = [
"HLA-A*02:01", "HLA-B*27:01", "HLA-C*07:01",
@@ -456,28 +502,34 @@ def Xtest_synthetic_allele_refinement_with_affinity_data(max_epochs=10):
mms_train_df = pandas.concat([hits_df, decoys_df], ignore_index=True)
mms_train_df["label"] = mms_train_df.hit
mms_train_df["is_affinity"] = False
+ mms_train_df["measurement_inequality"] = None
+
+ if include_affinities:
+ affinity_train_df = pandas.read_csv(
+ get_path(
+ "models_class1_pan", "models.with_mass_spec/train_data.csv.bz2"))
+ affinity_train_df = affinity_train_df.loc[
+ affinity_train_df.allele.isin(alleles),
+ ["peptide", "allele", "measurement_inequality", "measurement_value"]]
+
+ affinity_train_df["label"] = affinity_train_df["measurement_value"]
+ del affinity_train_df["measurement_value"]
+ affinity_train_df["is_affinity"] = True
+ else:
+ affinity_train_df = None
- affinity_train_df = pandas.read_csv(
- get_path(
- "models_class1_pan", "models.with_mass_spec/train_data.csv.bz2"))
- affinity_train_df = affinity_train_df.loc[
- affinity_train_df.allele.isin(alleles),
- ["peptide", "allele", "measurement_inequality", "measurement_value"]]
-
- affinity_train_df["label"] = affinity_train_df["measurement_value"]
- del affinity_train_df["measurement_value"]
- affinity_train_df["is_affinity"] = True
-
- predictor = Class1PresentationNeuralNetwork(
- PAN_ALLELE_PREDICTOR_NO_MASS_SPEC,
+ (affinity_model,) = PAN_ALLELE_PREDICTOR_NO_MASS_SPEC.class1_pan_allele_models
+ presentation_model = Class1PresentationNeuralNetwork(
auxiliary_input_features=["gene"],
- max_ensemble_size=1,
max_epochs=max_epochs,
learning_rate=0.0001,
patience=5,
min_delta=0.0,
- random_negative_rate=1.0,
- random_negative_constant=25)
+ random_negative_rate=0.0,
+ random_negative_constant=0) # WHY DOES THIS BREAK WITH RANDOM NEG?
+ presentation_model.load_from_class1_neural_network(affinity_model)
+
+ presentation_model = pickle.loads(pickle.dumps(presentation_model))
mms_allele_encoding = MultipleAlleleEncoding(
experiment_names=["experiment1"] * len(mms_train_df),
@@ -488,27 +540,24 @@ def Xtest_synthetic_allele_refinement_with_affinity_data(max_epochs=10):
allele_to_sequence=PAN_ALLELE_PREDICTOR_NO_MASS_SPEC.allele_to_sequence,
)
allele_encoding = copy.deepcopy(mms_allele_encoding)
- allele_encoding.append_alleles(affinity_train_df.allele.values)
- allele_encoding = allele_encoding.compact()
+ if affinity_train_df is not None:
+ allele_encoding.append_alleles(affinity_train_df.allele.values)
+ train_df = pandas.concat(
+ [mms_train_df, affinity_train_df], ignore_index=True, sort=False)
+ else:
+ train_df = mms_train_df
- train_df = pandas.concat(
- [mms_train_df, affinity_train_df], ignore_index=True, sort=False)
+ allele_encoding = allele_encoding.compact()
- pre_predictions = from_ic50(
- predictor.predict(
- output="affinities_matrix",
+ pre_predictions = presentation_model.predict(
peptides=mms_train_df.peptide.values,
- alleles=mms_allele_encoding))
+ allele_encoding=mms_allele_encoding).score
- (model,) = PAN_ALLELE_PREDICTOR_NO_MASS_SPEC.class1_pan_allele_models
expected_pre_predictions = from_ic50(
- model.predict(
+ affinity_model.predict(
peptides=numpy.repeat(mms_train_df.peptide.values, len(alleles)),
allele_encoding=mms_allele_encoding.allele_encoding,
)).reshape((-1, len(alleles)))
-
- #import ipdb ; ipdb.set_trace()
-
mms_train_df["pre_max_prediction"] = pre_predictions.max(1)
pre_auc = roc_auc_score(mms_train_df.hit.values, mms_train_df.pre_max_prediction.values)
print("PRE_AUC", pre_auc)
@@ -526,6 +575,7 @@ def Xtest_synthetic_allele_refinement_with_affinity_data(max_epochs=10):
metric_rows = []
+ """
def progress():
(_, presentation_prediction, affinities_predictions) = (
predictor.predict(
@@ -573,24 +623,26 @@ def Xtest_synthetic_allele_refinement_with_affinity_data(max_epochs=10):
update_motifs()
return (presentation_prediction, auc)
+
print("Pre fitting:")
progress()
update_motifs()
print("Fitting...")
-
- predictor.fit(
+ """
+ presentation_model.fit(
peptides=train_df.peptide.values,
labels=train_df.label.values,
inequalities=train_df.measurement_inequality.values,
affinities_mask=train_df.is_affinity.values,
allele_encoding=allele_encoding,
- progress_callback=progress,
+ #progress_callback=progress,
)
- (predictions, final_auc) = progress()
- print("Final AUC", final_auc)
+ #(predictions, final_auc) = progress()
+ #print("Final AUC", final_auc)
+ """
update_motifs()
motifs = pandas.DataFrame(
@@ -612,6 +664,7 @@ def Xtest_synthetic_allele_refinement_with_affinity_data(max_epochs=10):
])
return (predictor, predictions, metrics, motifs)
+ """