!34569 add 310 infer ST testcase of yolov4

Merge pull request !34569 from GongLiyao/st_testcase_yolov4

tests/models

@@ -1 +1 @@
-Subproject commit c819236923d1bd072f431f39b352e1e77f348b7b
+Subproject commit 5cd93e4d7af7e98e05fdcc98c901c33c971523c8

tests/st/model_zoo_tests/310_infer_yolov4/test_infer_310_yolov4.py

@@ -0,0 +1,110 @@
+# Copyright 2022 Huawei Technologies Co., Ltd
+#
+# 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
+#
+# 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.
+# ============================================================================
+
+import os
+import pytest
+
+from tests.st.model_zoo_tests import utils
+
+
+def check_log_file(infer_ret_path):
+    """
+    Feature：check the log file and get the 310infer results.
+    Description: infer_ret_path -- the path of 310 inference.
+    Expectation: accuracy_data, fps_data.
+    """
+
+    build_log = "{0}/../ascend310_infer/src/build.log".format(infer_ret_path)
+    infer_log = os.path.join(infer_ret_path, "infer.log")
+    acc_log = os.path.join(infer_ret_path, "acc.log")
+    log_file_list = [build_log, infer_log, acc_log]
+    check_type = True
+    for log_file in log_file_list:
+        if not os.path.exists(log_file):
+            print(f"{log_file} is not exists.")
+            check_type = False
+        os.system(r""" grep "ERROR" {} > log_error.txt""".format(log_file))
+        with open("./log_error.txt", "r") as f:
+            check_err_info = f.read()
+            if check_err_info != '':
+                print(check_err_info)
+                check_type = False
+        os.remove("./log_error.txt")
+    assert check_type, "Exist 'ERROR' in log files."
+
+    # verify accuracy and performance
+    get_accuracy_shell = "cd %s; grep 'Average Precision' acc.log | grep 'IoU=0.50:0.95' \
+                          | grep 'area=   all' | awk -F ' ] = ' '{print $2}' >acc_data.txt" % infer_ret_path
+    get_perf_shell = "cd %s; grep 'cost average time' infer.log | awk -F : '{ print $2 }'\
+                          | awk -F 'ms' '{ print $1 }' > pref_data.txt" % infer_ret_path
+    os.system(get_accuracy_shell)
+    os.system(get_perf_shell)
+    accuracy_data, fps_data = 0, 0
+    with open(f"{infer_ret_path}/acc_data.txt", "r") as acc:
+        accuracy_data = float(acc.read())
+    with open(f"{infer_ret_path}/pref_data.txt", "r") as pref:
+        perf_data = float(pref.read())
+        fps_data = 4 * 1000 / perf_data
+    os.remove(f"{infer_ret_path}/acc_data.txt")
+    os.remove(f"{infer_ret_path}/pref_data.txt")
+
+    return accuracy_data, fps_data
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_ascend310_inference
+@pytest.mark.platform_arm_ascend310_inference
+@pytest.mark.env_onecard
+def test_infer_310_yolov4():
+    """
+    Feature：Verify the yolov4 310 infer process.
+    Description: export DATASET_PATH(option), CKPT_FILE(option), DEVICE_ID(option).
+    Expectation: success or accuracy/fps less than standard.
+    """
+
+    dataset_path = os.path.abspath(os.getenv('DATASET_PATH', "/home/workspace/mindspore_dataset"))
+    ckpt_file = os.path.abspath(os.getenv('CKPT_FILE',\
+        "/home/workspace/mindspore_ckpt/ckpt/yolov4_ascend_v130_coco2017_official_cv_bs8_acc44.ckpt"))
+    device_id = int(os.getenv('DEVICE_ID', '0'))
+    print("Specifying path by setting environment variables 'DEVICE_ID', 'DATASET_PATH' and 'CKPT_FILE'.")
+    assert os.path.isfile(ckpt_file), f"Ckpt_File:{ckpt_file} is not exist."
+    assert os.path.isdir(dataset_path), f"Dataset_Path:{dataset_path} is not exist."
+
+    # export to MindIR
+    cur_path = os.path.dirname(os.path.abspath(__file__))
+    model_path = "{}/../../../../tests/models/official/cv".format(cur_path)
+    model_name = "yolov4"
+    assert os.path.isdir(os.path.join(model_path, model_name)), "models dir is not exist."
+    utils.copy_files(model_path, cur_path, model_name)
+    cur_model_path = os.path.join(cur_path, model_name)
+    exec_export_shell = "cd {0}; python export.py --ckpt_file {1} --file_name yolov4 --file_format MINDIR \
+                         --batch_size 1 --device_id {2}".format(model_name, ckpt_file, device_id)
+    os.system(exec_export_shell)
+    assert os.path.exists(os.path.join(cur_model_path, "yolov4.mindir")), "MINDIR_File is not exist."
+
+    # do 310 infer
+    old_list, new_list = ["&> acc.log &"], ["> acc.log"]
+    infer_ret_path = os.path.join(cur_model_path, "scripts")
+    utils.exec_sed_command(old_list, new_list, os.path.join(infer_ret_path, "run_infer_310.sh"))
+    exec_infer_shell = "cd {0}; bash run_infer_310.sh ../yolov4.mindir {1} {2} {3}".format(infer_ret_path, \
+        os.path.join(dataset_path, "coco", "coco2017_500", "val2017_500"), device_id, \
+        os.path.join(dataset_path, "coco", "coco2017_500", "annotations", "instances_val2017_500.json"))
+    os.system(exec_infer_shell)
+
+    accuracy_data, fps_data = check_log_file(infer_ret_path)
+    standard_value = {'acc': 0.470, 'fps': 55}
+    print(f"accuracy:{accuracy_data}, fps:{fps_data}")
+    assert accuracy_data > standard_value.get('acc')
+    assert fps_data > standard_value.get('fps')

tests/st/model_zoo_tests/bgcf/test_BGCF_amazon_beauty.py

@@ -47,13 +47,13 @@ def test_BGCF_amazon_beauty():
     print("train process finished.")
 
     log_file = os.path.join(cur_model_path, "log")
-    pattern1 = r"loss ([\d\.\+]+)\,"
+    pattern1 = r"loss: ([\d\.\+]+)\,"
     loss_list = utils.parse_log_file(pattern1, log_file)
     loss_list = loss_list[-5:]
     print("last 5 epoch average loss is", sum(loss_list) / len(loss_list))
     assert sum(loss_list) / len(loss_list) < 6400
 
-    pattern1 = r"cost:([\d\.\+]+)"
+    pattern1 = r"epoch time: ([\d\.\+]+)"
     epoch_time_list = utils.parse_log_file(pattern1, log_file)[1:]
     print("per epoch time:", sum(epoch_time_list) / len(epoch_time_list))
     assert sum(epoch_time_list) / len(epoch_time_list) < 2.2