Batch size adjustments

Author: Stanislav Pidhorskyi

metrics/fid_rec.py

@@ -179,7 +179,7 @@ def sample(cfg, logger):
     decoder = nn.DataParallel(decoder)
 
     with torch.no_grad():
-        ppl = FID(cfg, num_images=50000, minibatch_size=8 * torch.cuda.device_count())
+        ppl = FID(cfg, num_images=50000, minibatch_size=16 * torch.cuda.device_count())
         ppl.evaluate(logger, mapping_fl, decoder, encoder, cfg.DATASET.MAX_RESOLUTION_LEVEL - 2)
 
 

metrics/fid_sep.py

@@ -27,7 +27,6 @@
 import utils
 
 dnnlib.tflib.init_tf()
-tf_config = {'rnd.np_random_seed': 1000}
 
 download.from_google_drive('1CIDc9i070KQhHlkr4yIwoJC8xqrwjE0_', directory="metrics")
 

metrics/lpips.py

@@ -1,13 +1,18 @@
 # Copyright 2019-2020 Stanislav Pidhorskyi
 #
-# Copyright (c) 2019, NVIDIA CORPORATION. All rights reserved.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-# This work is licensed under the Creative Commons Attribution-NonCommercial
-# 4.0 International License. To view a copy of this license, visit
-# http://creativecommons.org/licenses/by-nc/4.0/ or send a letter to
-# Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.
+#  http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
 
-"""Perceptual Path Length (PPL)."""
 
 import dnnlib.tflib
 import pickle
@@ -28,7 +33,6 @@
 import utils
 
 dnnlib.tflib.init_tf()
-tf_config = {'rnd.np_random_seed': 1000}
 
 download.from_google_drive('1CIDc9i070KQhHlkr4yIwoJC8xqrwjE0_', directory="metrics")
 
@@ -145,7 +149,7 @@ def sample(cfg, logger):
     encoder = nn.DataParallel(encoder)
 
     with torch.no_grad():
-        ppl = LPIPS(cfg, num_images=10000, minibatch_size=8 * torch.cuda.device_count())
+        ppl = LPIPS(cfg, num_images=10000, minibatch_size=16 * torch.cuda.device_count())
         ppl.evaluate(logger, mapping_fl, decoder, encoder, cfg.DATASET.MAX_RESOLUTION_LEVEL - 2)