fix tests

mhcflurry/__init__.py

-from mhcflurry.class1_affinity_predictor import Class1AffinityPredictor
-from mhcflurry.class1_neural_network import Class1NeuralNetwork
-
-__version__ = "1.1.0"
+from .class1_affinity_predictor import Class1AffinityPredictor
+from .class1_neural_network import Class1NeuralNetwork
+from .version import __version__
 
 __all__ = [
     "__version__",

mhcflurry/class1_affinity_predictor.py

@@ -7,6 +7,8 @@ import time
 import warnings
 from os.path import join, exists
 from os import mkdir
+from socket import gethostname
+from getpass import getuser
 
 import mhcnames
 import numpy
@@ -20,6 +22,7 @@ from .downloads import get_path
 from .encodable_sequences import EncodableSequences
 from .percent_rank_transform import PercentRankTransform
 from .regression_target import to_ic50
+from .version import __version__
 
 
 class Class1AffinityPredictor(object):
@@ -193,7 +196,8 @@ class Class1AffinityPredictor(object):
         the configurations of each Class1NeuralNetwork, along with per-network
         files giving the model weights. If there are pan-allele predictors in
         the ensemble, the allele pseudosequences are also stored in the
-        directory.
+        directory. There is also a small file "index.txt" with basic metadata:
+        when the models were trained, by whom, on what host.
         
         Parameters
         ----------
@@ -234,6 +238,18 @@ class Class1AffinityPredictor(object):
         write_manifest_df.to_csv(manifest_path, index=False)
         logging.info("Wrote: %s" % manifest_path)
 
+        # Write "info.txt"
+        info_path = join(models_dir, "info.txt")
+        rows = [
+            ("trained on", time.asctime()),
+            ("package   ", "mhcflurry %s" % __version__),
+            ("hostname  ", gethostname()),
+            ("user      ", getuser()),
+        ]
+        pandas.DataFrame(rows).to_csv(
+            info_path, sep="\t", header=False, index=False)
+        print("Wrote: %s" % info_path)
+
         if self.allele_to_percent_rank_transform:
             percent_ranks_df = None
             for (allele, transform) in self.allele_to_percent_rank_transform.items():

mhcflurry/version.py

+__version__ = "1.1.0"

setup.py

@@ -40,7 +40,7 @@ except:
     logging.warning("Conversion of long_description from MD to RST failed")
     pass
 
-with open('mhcflurry/__init__.py', 'r') as f:
+with open('mhcflurry/version.py', 'r') as f:
     version = re.search(
         r'^__version__\s*=\s*[\'"]([^\'"]*)[\'"]',
         f.read(),

test/test_class1_affinity_predictor.py

@@ -51,7 +51,7 @@ def test_a1_known_epitopes_in_newly_trained_model():
     allele = "HLA-A*01:01"
     df = pandas.read_csv(
         get_path(
-            "data_curated", "curated_training_data.csv.bz2"))
+            "data_curated", "curated_training_data.no_mass_spec.csv.bz2"))
     df = df.ix[
         (df.allele == allele) &
         (df.peptide.str.len() >= 8) &
@@ -137,7 +137,7 @@ def test_class1_affinity_predictor_a0205_memorize_training_data():
 
     df = pandas.read_csv(
         get_path(
-            "data_curated", "curated_training_data.csv.bz2"))
+            "data_curated", "curated_training_data.no_mass_spec.csv.bz2"))
     df = df.ix[
         df.allele == allele
     ]

test/test_class1_neural_network.py

@@ -37,7 +37,7 @@ def test_class1_neural_network_a0205_training_accuracy():
 
     df = pandas.read_csv(
         get_path(
-            "data_curated", "curated_training_data.csv.bz2"))
+            "data_curated", "curated_training_data.no_mass_spec.csv.bz2"))
     df = df.ix[
         df.allele == allele
     ]

test/test_train_allele_specific_models_command.py

@@ -57,12 +57,13 @@ def run_and_check(n_jobs=0):
         json.dump(HYPERPARAMETERS, fd)
 
     args = [
-        "--data", get_path("data_curated", "curated_training_data.csv.bz2"),
+        "--data", get_path("data_curated", "curated_training_data.no_mass_spec.csv.bz2"),
         "--hyperparameters", hyperparameters_filename,
         "--allele", "HLA-A*02:01", "HLA-A*01:01", "HLA-A*03:01",
         "--out-models-dir", models_dir,
         "--percent-rank-calibration-num-peptides-per-length", "10000",
         "--parallelization-num-jobs", str(n_jobs),
+        "--ignore-inequalities",
     ]
     print("Running with args: %s" % args)
     train_allele_specific_models_command.run(args)