python /home/admin/mtr/script_for_cron.py -j default -m 20 -a 'python3 ~/workarea/git/Velours/python/prod/datou.py -j batch_current -C 3911038' -s traitement_4877 -M 0 -S 0 -U 100,80,95 import MySQLdb succeeded Import error (python version) ['/Users/moilerat/Documents/Fotonower/install/caffe/distribute/python', '/home/admin/workarea/git/Velours/python/prod', '/home/admin/workarea/install/darknet', '/home/admin/workarea/git/Velours/python', '/home/admin/workarea/install/caffe_frcnn_python3/py-faster-rcnn/caffe-fast-rcnn/python', '/home/admin/mtr/.credentials', '/home/admin/workarea/install/caffe/python', '/home/admin/workarea/install/caffe_frcnn/py-faster-rcnn/tools', '/home/admin/workarea/git/fotonowerpip', '/home/admin/workarea/install/segment-anything', '/home/admin/workarea/git/pyfvs', '/home/admin/workarea/git/apy', '/usr/lib/python38.zip', '/usr/lib/python3.8', '/usr/lib/python3.8/lib-dynload', '/home/admin/.local/lib/python3.8/site-packages', '/usr/local/lib/python3.8/dist-packages', '/usr/lib/python3/dist-packages'] process id : 4124462 load datou : 0 # VR 17-11-17 : to create in DB ! Here we check the datou graph and we reorder steps ! Tree builded and cycle checked, now we need to re-order the steps ! We have currenlty an error because there is no dependence between the last step for the case tile - detect - glue We can either keep the depence of, it is better to keep an order compatible with the id of steps if we do not have sons, so a lexical order : (number_son, step_id) DONE and to test : checkNoCycle ! Here we check the consistency of inputs/outputs number between the given ones and the db ! eke 1-6-18 : checkConsistencyNbInputNbOutput should be processed after step reordering ! WARNING : step 0 init_dummy_multi_datou is not linked in the step_by_step architecture ! WARNING : step 1294 init_dummy_multi_datou is not linked in the step_by_step architecture ! Number of inputs / outputs for each step checked ! Here we check the consistency of outputs/inputs types during steps connections eke 1-6-18 : checkConsistencyTypeOutputInput should be processed after checkConsistencyNbInputNbOutput ! DataTypes for each output/input checked ! Unexpected type for variable list_input_json ERROR or WARNING : can't parse json string Expecting value: line 1 column 1 (char 0) Tried to parse : (photo_id, hashtag_id, score_max) was removed should we ? (x0, y0, x1, y1) was removed should we ? chemin de la photo was removed should we ? (photo_id, hashtag_id, score_max) was removed should we ? (x0, y0, x1, y1) was removed should we ? chemin de la photo was removed should we ? load thcls load pdts Running datou job : batch_current TODO datou_current to load to do maybe to take outside batchDatouExec updating current state to 1 list_input_json: [] Current got : datou_id : 4877, datou_cur_ids : ['3911038'] with mtr_portfolio_ids : ['27666970'] and first list_photo_ids : [] new path : /proc/4124462/ Inside batchDatouExec : verbose : 0 # VR 17-11-17 : to create in DB ! Here we check the datou graph and we reorder steps ! Tree builded and cycle checked, now we need to re-order the steps ! We have currenlty an error because there is no dependence between the last step for the case tile - detect - glue We can either keep the depence of, it is better to keep an order compatible with the id of steps if we do not have sons, so a lexical order : (number_son, step_id) All sons are already in current list ! All sons are already in current list ! All sons are already in current list ! All sons are already in current list ! All sons are already in current list ! All sons are already in current list ! All sons are already in current list ! All sons are already in current list ! DONE and to test : checkNoCycle ! Here we check the consistency of inputs/outputs number between the given ones and the db ! eke 1-6-18 : checkConsistencyNbInputNbOutput should be processed after step reordering ! WARNING : number of outputs for step 13611 mask_detect is not consistent : 3 used against 2 in the step definition ! Step 13613 crop_condition have less inputs used (1) than in the step definition (2) : maybe we manage optionnal inputs ! Step 13613 crop_condition have less outputs used (2) than in the step definition (3) : some outputs may be not used ! Step 13614 merge_mask_thcl_custom have less inputs used (2) than in the step definition (3) : maybe we manage optionnal inputs ! WARNING : number of outputs for step 13614 merge_mask_thcl_custom is not consistent : 4 used against 2 in the step definition ! WARNING : number of inputs for step 13615 rle_unique_nms_with_priority is not consistent : 2 used against 1 in the step definition ! WARNING : number of outputs for step 13615 rle_unique_nms_with_priority is not consistent : 2 used against 1 in the step definition ! Step 13621 crop_condition have less inputs used (1) than in the step definition (2) : maybe we manage optionnal inputs ! WARNING : number of outputs for step 13621 crop_condition is not consistent : 4 used against 3 in the step definition ! WARNING : number of inputs for step 13617 ventilate_hashtags_in_portfolio is not consistent : 2 used against 1 in the step definition ! Step 13617 ventilate_hashtags_in_portfolio have less outputs used (0) than in the step definition (1) : some outputs may be not used ! Number of inputs / outputs for each step checked ! Here we check the consistency of outputs/inputs types during steps connections eke 1-6-18 : checkConsistencyTypeOutputInput should be processed after checkConsistencyNbInputNbOutput ! WARNING : type of output 2 of step 13611 doesn't seem to be define in the database( WARNING : type of input 2 of step 13613 doesn't seem to be define in the database( WARNING : output 1 of step 13611 have datatype=2 whereas input 1 of step 13614 have datatype=7 WARNING : type of output 2 of step 13614 doesn't seem to be define in the database( WARNING : type of input 1 of step 13615 doesn't seem to be define in the database( WARNING : type of output 3 of step 13614 doesn't seem to be define in the database( WARNING : type of input 1 of step 13617 doesn't seem to be define in the database( WARNING : type of input 2 of step 13621 doesn't seem to be define in the database( WARNING : output 1 of step 13615 have datatype=7 whereas input 2 of step 13621 have datatype=None WARNING : type of output 3 of step 13621 doesn't seem to be define in the database( WARNING : type of input 2 of step 13617 doesn't seem to be define in the database( WARNING : output 0 of step 13614 have datatype=1 whereas input 0 of step 13615 have datatype=2 DataTypes for each output/input checked ! List Step Type Loaded in datou : mask_detect, crop_condition, thcl, merge_mask_thcl_custom, rle_unique_nms_with_priority, crop_condition, ventilate_hashtags_in_portfolio, tfhub_classification2, argmax over limit max, limiting to limit_max 20 list_input_json : [] origin We have 1 , BFBFBFwe have missing 0 photos in the step downloads : photo missing : [] try to delete the photos missing in DB length of list_filenames : 3 ; length of list_pids : 3 ; length of list_args : 3 time to download the photos : 0.47518253326416016 About to test input to load we should then remove the video here, and this would fix the bug of datou_current ! Calling datou_exec Inside datou_exec : verbose : 0 number of steps : 9 step1:mask_detect Fri Oct 10 09:28:02 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Beginning of datou step mask_detect ! save_polygon : True begin detect begin to check gpu status inside check gpu memory l 3637 free memory gpu now : 7216 max_wait_temp : 1 max_wait : 0 gpu_flag : 0 2025-10-10 09:28:05.366701: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA 2025-10-10 09:28:05.390635: I tensorflow/core/platform/profile_utils/cpu_utils.cc:102] CPU Frequency: 3493010000 Hz 2025-10-10 09:28:05.392378: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7f2290000b60 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2025-10-10 09:28:05.392427: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version 2025-10-10 09:28:05.396030: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1 2025-10-10 09:28:05.537427: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x3cf32c90 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2025-10-10 09:28:05.537472: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA GeForce RTX 2080 Ti, Compute Capability 7.5 2025-10-10 09:28:05.538725: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: pciBusID: 0000:41:00.0 name: NVIDIA GeForce RTX 2080 Ti computeCapability: 7.5 coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s 2025-10-10 09:28:05.539112: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:05.541993: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:28:05.544640: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10 2025-10-10 09:28:05.545136: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10 2025-10-10 09:28:05.547953: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10 2025-10-10 09:28:05.549115: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10 2025-10-10 09:28:05.553661: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2025-10-10 09:28:05.555146: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0 2025-10-10 09:28:05.555215: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:05.555876: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix: 2025-10-10 09:28:05.555891: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108] 0 2025-10-10 09:28:05.555899: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0: N 2025-10-10 09:28:05.557067: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6640 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 2080 Ti, pci bus id: 0000:41:00.0, compute capability: 7.5) WARNING:tensorflow:From /home/admin/workarea/git/Velours/python/mtr/mask_rcnn/mask_detection.py:69: The name tf.keras.backend.set_session is deprecated. Please use tf.compat.v1.keras.backend.set_session instead. 2025-10-10 09:28:05.994736: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: pciBusID: 0000:41:00.0 name: NVIDIA GeForce RTX 2080 Ti computeCapability: 7.5 coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s 2025-10-10 09:28:05.994873: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:05.994903: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:28:05.994929: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10 2025-10-10 09:28:05.994974: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10 2025-10-10 09:28:05.995001: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10 2025-10-10 09:28:05.995025: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10 2025-10-10 09:28:05.995051: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2025-10-10 09:28:05.996150: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0 2025-10-10 09:28:05.997295: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: pciBusID: 0000:41:00.0 name: NVIDIA GeForce RTX 2080 Ti computeCapability: 7.5 coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s 2025-10-10 09:28:05.997330: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:05.997346: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:28:05.997360: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10 2025-10-10 09:28:05.997373: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10 2025-10-10 09:28:05.997387: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10 2025-10-10 09:28:05.997401: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10 2025-10-10 09:28:05.997414: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2025-10-10 09:28:05.998440: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0 2025-10-10 09:28:05.998476: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix: 2025-10-10 09:28:05.998484: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108] 0 2025-10-10 09:28:05.998491: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0: N 2025-10-10 09:28:05.999512: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6640 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 2080 Ti, pci bus id: 0000:41:00.0, compute capability: 7.5) Using TensorFlow backend. WARNING:tensorflow:From /home/admin/workarea/install/Mask_RCNN/model.py:396: calling crop_and_resize_v1 (from tensorflow.python.ops.image_ops_impl) with box_ind is deprecated and will be removed in a future version. Instructions for updating: box_ind is deprecated, use box_indices instead WARNING:tensorflow:From /home/admin/workarea/install/Mask_RCNN/model.py:703: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version. Instructions for updating: Use `tf.cast` instead. WARNING:tensorflow:From /home/admin/workarea/install/Mask_RCNN/model.py:729: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version. Instructions for updating: Use `tf.cast` instead. Inside mask_sub_process Inside mask_detect About to load cache.load_thcl_param To do loadFromThcl(), then load ParamDescType : thcl2896 thcls : [{'id': 2896, 'mtr_user_id': 31, 'name': 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'background,carton_brun,carton_gris,cartonnette,kraft,autre_refus,metal,plastique,teint_dans_la_masse,environnement', 'svm_portfolios_learning': '0,0,0,0,0,0,0,0,0,0', 'photo_hashtag_type': 3663, 'photo_desc_type': 5309, 'type_classification': 'mask_rcnn', 'hashtag_id_list': '0,0,0,0,0,0,0,0,0,0'}] thcl {'id': 2896, 'mtr_user_id': 31, 'name': 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'background,carton_brun,carton_gris,cartonnette,kraft,autre_refus,metal,plastique,teint_dans_la_masse,environnement', 'svm_portfolios_learning': '0,0,0,0,0,0,0,0,0,0', 'photo_hashtag_type': 3663, 'photo_desc_type': 5309, 'type_classification': 'mask_rcnn', 'hashtag_id_list': '0,0,0,0,0,0,0,0,0,0'} Update svm_hashtag_type_desc : 5309 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (5309, 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 16384, 25088, 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 'pool5', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 1, datetime.datetime(2021, 5, 10, 19, 20, 46), datetime.datetime(2021, 5, 10, 19, 20, 46)) {'thcl': {'id': 2896, 'mtr_user_id': 31, 'name': 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'background,carton_brun,carton_gris,cartonnette,kraft,autre_refus,metal,plastique,teint_dans_la_masse,environnement', 'svm_portfolios_learning': '0,0,0,0,0,0,0,0,0,0', 'photo_hashtag_type': 3663, 'photo_desc_type': 5309, 'type_classification': 'mask_rcnn', 'hashtag_id_list': '0,0,0,0,0,0,0,0,0,0'}, 'list_hashtags': ['background', 'carton_brun', 'carton_gris', 'cartonnette', 'kraft', 'autre_refus', 'metal', 'plastique', 'teint_dans_la_masse', 'environnement'], 'list_hashtags_csv': 'background,carton_brun,carton_gris,cartonnette,kraft,autre_refus,metal,plastique,teint_dans_la_masse,environnement', 'svm_portfolios_learning': '0,0,0,0,0,0,0,0,0,0', 'photo_hashtag_type': 3663, 'svm_hashtag_type_desc': 5309, 'photo_desc_type': 5309, 'pb_hashtag_id_or_classifier': 0} list_class_names : ['background', 'carton_brun', 'carton_gris', 'cartonnette', 'kraft', 'autre_refus', 'metal', 'plastique', 'teint_dans_la_masse', 'environnement'] Configurations: BACKBONE resnet101 BACKBONE_SHAPES [[160 160] [ 80 80] [ 40 40] [ 20 20] [ 10 10]] BACKBONE_STRIDES [4, 8, 16, 32, 64] BATCH_SIZE 1 BBOX_STD_DEV [0.1 0.1 0.2 0.2] DETECTION_MAX_INSTANCES 100 DETECTION_MIN_CONFIDENCE 0.3 DETECTION_NMS_THRESHOLD 0.3 GPU_COUNT 1 IMAGES_PER_GPU 1 IMAGE_MAX_DIM 640 IMAGE_MIN_DIM 640 IMAGE_PADDING True IMAGE_SHAPE [640 640 3] LEARNING_MOMENTUM 0.9 LEARNING_RATE 0.001 LOSS_WEIGHTS {'rpn_class_loss': 1.0, 'rpn_bbox_loss': 1.0, 'mrcnn_class_loss': 1.0, 'mrcnn_bbox_loss': 1.0, 'mrcnn_mask_loss': 1.0} MASK_POOL_SIZE 14 MASK_SHAPE [28, 28] MAX_GT_INSTANCES 100 MEAN_PIXEL [123.7 116.8 103.9] MINI_MASK_SHAPE (56, 56) NAME learn_convoyeur_qualipapia_nantes_poly_100521_1 NUM_CLASSES 10 POOL_SIZE 7 POST_NMS_ROIS_INFERENCE 1000 POST_NMS_ROIS_TRAINING 2000 ROI_POSITIVE_RATIO 0.33 RPN_ANCHOR_RATIOS [0.5, 1, 2] RPN_ANCHOR_SCALES (16, 32, 64, 128, 256) RPN_ANCHOR_STRIDE 1 RPN_BBOX_STD_DEV [0.1 0.1 0.2 0.2] RPN_NMS_THRESHOLD 0.7 RPN_TRAIN_ANCHORS_PER_IMAGE 256 STEPS_PER_EPOCH 1000 TRAIN_ROIS_PER_IMAGE 200 USE_MINI_MASK True USE_RPN_ROIS True VALIDATION_STEPS 50 WEIGHT_DECAY 0.0001 model_param file didn't exist model_name : learn_convoyeur_qualipapia_nantes_poly_100521_1 model_type : mask_rcnn list file need : ['mask_model.h5'] file exist in s3 : ['mask_model.h5'] file manque in s3 : [] 2025-10-10 09:28:12.666900: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:28:12.857243: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 local folder : /data/models_weight/learn_convoyeur_qualipapia_nantes_poly_100521_1 /data/models_weight/learn_convoyeur_qualipapia_nantes_poly_100521_1/mask_model.h5 size_local : 256031040 size in s3 : 256031040 create time local : 2021-08-09 05:45:48 create time in s3 : 2021-08-06 18:59:51 mask_model.h5 already exist and didn't need to update list_images length : 3 NEW PHOTO Processing 1 images image shape: (1080, 1632, 3) min: 0.00000 max: 255.00000 molded_images shape: (1, 640, 640, 3) min: -123.70000 max: 151.10000 image_metas shape: (1, 18) min: 0.00000 max: 1632.00000 nb d'objets trouves : 4 NEW PHOTO Processing 1 images image shape: (1080, 1632, 3) min: 0.00000 max: 255.00000 molded_images shape: (1, 640, 640, 3) min: -123.70000 max: 151.10000 image_metas shape: (1, 18) min: 0.00000 max: 1632.00000 nb d'objets trouves : 5 NEW PHOTO Processing 1 images image shape: (1080, 1632, 3) min: 0.00000 max: 255.00000 molded_images shape: (1, 640, 640, 3) min: -123.70000 max: 151.10000 image_metas shape: (1, 18) min: 0.00000 max: 1632.00000 nb d'objets trouves : 25 Detection mask done ! Trying to reset tf kernel 4124730 begin to check gpu status inside check gpu memory l 3610 free memory gpu now : 5706 tf kernel not reseted sub process len(results) : 3 len(list_Values) 0 None max_time_sub_proc : 3600 parent process len(results) : 3 len(list_Values) 0 process is alive finish correctly or not : True after detect begin to check gpu status inside check gpu memory l 3610 free memory gpu now : 10998 list_Values should be empty [] To do loadFromThcl(), then load ParamDescType : thcl2896 Catched exception ! Connect or reconnect ! thcls : [{'id': 2896, 'mtr_user_id': 31, 'name': 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'background,carton_brun,carton_gris,cartonnette,kraft,autre_refus,metal,plastique,teint_dans_la_masse,environnement', 'svm_portfolios_learning': '0,0,0,0,0,0,0,0,0,0', 'photo_hashtag_type': 3663, 'photo_desc_type': 5309, 'type_classification': 'mask_rcnn', 'hashtag_id_list': '0,0,0,0,0,0,0,0,0,0'}] thcl {'id': 2896, 'mtr_user_id': 31, 'name': 'learn_convoyeur_qualipapia_nantes_poly_100521_1', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'background,carton_brun,carton_gris,cartonnette,kraft,autre_refus,metal,plastique,teint_dans_la_masse,environnement', 'svm_portfolios_learning': '0,0,0,0,0,0,0,0,0,0', 'photo_hashtag_type': 3663, 'photo_desc_type': 5309, 'type_classification': 'mask_rcnn', 'hashtag_id_list': '0,0,0,0,0,0,0,0,0,0'} Update svm_hashtag_type_desc : 5309 ['background', 'carton_brun', 'carton_gris', 'cartonnette', 'kraft', 'autre_refus', 'metal', 'plastique', 'teint_dans_la_masse', 'environnement'] time for calcul the mask position with numpy : 0.0021903514862060547 nb_pixel_total : 124166 time to create 1 rle with old method : 0.1375598907470703 length of segment : 360 time for calcul the mask position with numpy : 0.012640237808227539 nb_pixel_total : 903899 time to create 1 rle with new method : 0.028087377548217773 length of segment : 1633 time for calcul the mask position with numpy : 0.0010652542114257812 nb_pixel_total : 90948 time to create 1 rle with old method : 0.0937187671661377 length of segment : 350 time for calcul the mask position with numpy : 0.00042319297790527344 nb_pixel_total : 36115 time to create 1 rle with old method : 0.04000282287597656 length of segment : 229 time for calcul the mask position with numpy : 0.002088308334350586 nb_pixel_total : 153057 time to create 1 rle with new method : 0.0055332183837890625 length of segment : 935 time for calcul the mask position with numpy : 0.0065364837646484375 nb_pixel_total : 419912 time to create 1 rle with new method : 0.024052143096923828 length of segment : 1639 time for calcul the mask position with numpy : 0.0015499591827392578 nb_pixel_total : 90151 time to create 1 rle with old method : 0.11211609840393066 length of segment : 330 time for calcul the mask position with numpy : 0.00888681411743164 nb_pixel_total : 697246 time to create 1 rle with new method : 0.024122238159179688 length of segment : 1156 time for calcul the mask position with numpy : 0.00015974044799804688 nb_pixel_total : 5532 time to create 1 rle with old method : 0.006567239761352539 length of segment : 137 time for calcul the mask position with numpy : 0.0002987384796142578 nb_pixel_total : 9052 time to create 1 rle with old method : 0.010317087173461914 length of segment : 179 time for calcul the mask position with numpy : 3.814697265625e-05 nb_pixel_total : 376 time to create 1 rle with old method : 0.0004584789276123047 length of segment : 25 time for calcul the mask position with numpy : 7.200241088867188e-05 nb_pixel_total : 2291 time to create 1 rle with old method : 0.0027625560760498047 length of segment : 99 time for calcul the mask position with numpy : 4.38690185546875e-05 nb_pixel_total : 980 time to create 1 rle with old method : 0.0011758804321289062 length of segment : 40 time for calcul the mask position with numpy : 3.4809112548828125e-05 nb_pixel_total : 288 time to create 1 rle with old method : 0.00037217140197753906 length of segment : 29 time for calcul the mask position with numpy : 9.369850158691406e-05 nb_pixel_total : 4656 time to create 1 rle with old method : 0.005259275436401367 length of segment : 80 time for calcul the mask position with numpy : 0.005713224411010742 nb_pixel_total : 371385 time to create 1 rle with new method : 0.022112131118774414 length of segment : 1533 time for calcul the mask position with numpy : 0.0013134479522705078 nb_pixel_total : 98398 time to create 1 rle with old method : 0.10605287551879883 length of segment : 377 time for calcul the mask position with numpy : 0.00011205673217773438 nb_pixel_total : 1177 time to create 1 rle with old method : 0.0015895366668701172 length of segment : 37 time for calcul the mask position with numpy : 9.036064147949219e-05 nb_pixel_total : 953 time to create 1 rle with old method : 0.0011150836944580078 length of segment : 36 time for calcul the mask position with numpy : 0.0004057884216308594 nb_pixel_total : 8203 time to create 1 rle with old method : 0.009222030639648438 length of segment : 151 time for calcul the mask position with numpy : 0.00011873245239257812 nb_pixel_total : 6276 time to create 1 rle with old method : 0.006845712661743164 length of segment : 149 time for calcul the mask position with numpy : 0.001355886459350586 nb_pixel_total : 88722 time to create 1 rle with old method : 0.09172368049621582 length of segment : 383 time for calcul the mask position with numpy : 0.0007157325744628906 nb_pixel_total : 9743 time to create 1 rle with old method : 0.011022567749023438 length of segment : 252 time for calcul the mask position with numpy : 0.001940011978149414 nb_pixel_total : 53970 time to create 1 rle with old method : 0.05652117729187012 length of segment : 344 time for calcul the mask position with numpy : 0.0002186298370361328 nb_pixel_total : 2262 time to create 1 rle with old method : 0.002763032913208008 length of segment : 75 time for calcul the mask position with numpy : 9.965896606445312e-05 nb_pixel_total : 848 time to create 1 rle with old method : 0.0010306835174560547 length of segment : 40 time for calcul the mask position with numpy : 7.867813110351562e-05 nb_pixel_total : 799 time to create 1 rle with old method : 0.0011210441589355469 length of segment : 22 time for calcul the mask position with numpy : 4.792213439941406e-05 nb_pixel_total : 563 time to create 1 rle with old method : 0.0008156299591064453 length of segment : 25 time spent for convertir_results : 2.867941379547119 Inside saveOutput : final : False verbose : 0 eke 12-6-18 : saveMask need to be cleaned for new output ! Number saved : None batch 1 Loaded 28 chid ids of type : 3663 Number RLEs to save : 10645 save missing photos in datou_result : time spend for datou_step_exec : 19.48314332962036 time spend to save output : 1.3480846881866455 total time spend for step 1 : 20.831228017807007 step2:crop_condition Fri Oct 10 09:28:23 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed We should have FATAL ERROR but same_nb_input_output==True : this should be an optionnal input ! We should have FATAL ERROR but same_nb_input_output==True : this should be an optionnal input ! VR 22-3-18 : For now we do not clean correctly the datou structure Loading chi in step crop with photo_hashtag_type : 3663 Loading chi in step crop for list_pids : 3 ! batch 1 Loaded 28 chid ids of type : 3663 +++++++++++++++++++++++++++++++++++++++++++ begin to crop the class : teint_dans_la_masse param for this class : {'min_score': 0.7} filtre for class : teint_dans_la_masse hashtag_id of this class : 2107752385 we have both polygon and rles Next one ! map_result returned by crop_photo_return_map_crop : length : 1 About to insert : list_path_to_insert length 1 new photo from crops ! we have finished the crop for the class : teint_dans_la_masse begin to crop the class : autre_refus param for this class : {'min_score': 0.5} filtre for class : autre_refus hashtag_id of this class : 2107752406 we have both polygon and rles Next one ! we have both polygon and rles Next one ! map_result returned by crop_photo_return_map_crop : length : 2 About to insert : list_path_to_insert length 2 new photo from crops ! we have finished the crop for the class : autre_refus begin to crop the class : carton_gris param for this class : {'min_score': 0.5} filtre for class : carton_gris hashtag_id of this class : 2107753020 begin to crop the class : cartonnette param for this class : {'min_score': 0.5} filtre for class : cartonnette hashtag_id of this class : 702398920 we have both polygon and rles Next one ! we have both polygon and rles Next one ! we have both polygon and rles Next one ! we have both polygon and rles Next one ! we have both polygon and rles Next one ! we have both polygon and rles Next one ! map_result returned by crop_photo_return_map_crop : length : 6 About to insert : list_path_to_insert length 6 new photo from crops ! we have finished the crop for the class : cartonnette begin to crop the class : carton_brun param for this class : {'min_score': 0.7} filtre for class : carton_brun hashtag_id of this class : 2107753024 we have both polygon and rles Next one ! we have both polygon and rles Next one ! map_result returned by crop_photo_return_map_crop : length : 2 About to insert : list_path_to_insert length 2 new photo from crops ! we have finished the crop for the class : carton_brun begin to crop the class : plastique param for this class : {'min_score': 0.5} filtre for class : plastique hashtag_id of this class : 492725882 we have both polygon and rles Next one ! we have both polygon and rles Next one ! map_result returned by crop_photo_return_map_crop : length : 2 About to insert : list_path_to_insert length 2 new photo from crops ! we have finished the crop for the class : plastique begin to crop the class : kraft param for this class : {'min_score': 0.5} filtre for class : kraft hashtag_id of this class : 493202403 begin to crop the class : metal param for this class : {'min_score': 0.5} filtre for class : metal hashtag_id of this class : 492628673 delete rles for these photos Inside saveOutput : final : False verbose : 0 saveOutput not yet implemented for datou_step.type : crop_condition we use saveGeneral [1388575902, 1388575900, 1388575898] Looping around the photos to save general results len do output : 13 /-3991852383Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852398Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852391Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852376Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852381Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852380Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852397Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852390Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852393Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852388Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852392Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852377Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /-3991852394Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . before output type Here is an output not treated by saveGeneral : Here is an output not treated by saveGeneral : Here is an output not treated by saveGeneral : Managing all output in save final without adding information in the mtr_datou_result ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575902', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575900', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575898', None, None, None, None, None, '3911038') begin to insert list_values into mtr_datou_result : length of list_values in save_final : 42 time used for this insertion : 0.03864598274230957 save_final save missing photos in datou_result : time spend for datou_step_exec : 5.411890983581543 time spend to save output : 0.03915691375732422 total time spend for step 2 : 5.451047897338867 step3:thcl Fri Oct 10 09:28:28 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed complete output_args for input 0 VR 22-3-18 : For now we do not clean correctly the datou structure Beginning of datou step Thcl ! nombre de thcls : 2 we are using the classfication for multi_thcl [2456, 2868] time to import caffe and check if the image exist : 0.003021240234375 time to convert the images to numpy array : 2.384185791015625e-06 time to import caffe and check if the image exist : 0.0031943321228027344 time to convert the images to numpy array : 2.86102294921875e-06 time to import caffe and check if the image exist : 0.003116130828857422 time to convert the images to numpy array : 1.430511474609375e-06 time to import caffe and check if the image exist : 0.004280567169189453 time to convert the images to numpy array : 0.002353191375732422 time to import caffe and check if the image exist : 0.003704071044921875 time to convert the images to numpy array : 0.00522160530090332 time to import caffe and check if the image exist : 0.004173994064331055 time to convert the images to numpy array : 0.005132436752319336 time to import caffe and check if the image exist : 0.0038459300994873047 time to convert the images to numpy array : 0.006485700607299805 time to import caffe and check if the image exist : 0.0055501461029052734 time to convert the images to numpy array : 0.011377334594726562 time to import caffe and check if the image exist : 0.003640890121459961 time to convert the images to numpy array : 0.016382217407226562 time to import caffe and check if the image exist : 0.00586390495300293 time to convert the images to numpy array : 0.03702950477600098 total time to convert the images to numpy array : 0.22568511962890625 list photo_ids error: [] list photo_ids correct : [-3991852394, -3991852383, -3991852398, -3991852393, -3991852388, -3991852392, -3991852377, -3991852391, -3991852376, -3991852397, -3991852390, -3991852381, -3991852380] number of photos to traite : 13 try to delete the photos incorrect in DB tagging for thcl : 2456 To do loadFromThcl(), then load ParamDescType : thcl2456 thcls : [{'id': 2456, 'mtr_user_id': 31, 'name': 'learn_qualipapia_papier_refus_from_vlg_data_aug', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'papier,refus', 'svm_portfolios_learning': '3028087,3028251', 'photo_hashtag_type': 3049, 'photo_desc_type': 4999, 'type_classification': 'caffe', 'hashtag_id_list': '492668766,538914404'}] thcl {'id': 2456, 'mtr_user_id': 31, 'name': 'learn_qualipapia_papier_refus_from_vlg_data_aug', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'papier,refus', 'svm_portfolios_learning': '3028087,3028251', 'photo_hashtag_type': 3049, 'photo_desc_type': 4999, 'type_classification': 'caffe', 'hashtag_id_list': '492668766,538914404'} Update svm_hashtag_type_desc : 4999 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (4999, 'learn_qualipapia_papier_refus_from_vlg_data_aug', 16384, 25088, 'learn_qualipapia_papier_refus_from_vlg_data_aug', 'res5b', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 1, datetime.datetime(2020, 10, 23, 14, 27, 22), datetime.datetime(2020, 10, 23, 14, 27, 22)) To loadFromThcl() : net_4999 begin to check gpu status inside check gpu memory l 3637 free memory gpu now : 10998 max_wait_temp : 1 max_wait : 0 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (4999, 'learn_qualipapia_papier_refus_from_vlg_data_aug', 16384, 25088, 'learn_qualipapia_papier_refus_from_vlg_data_aug', 'res5b', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 1, datetime.datetime(2020, 10, 23, 14, 27, 22), datetime.datetime(2020, 10, 23, 14, 27, 22)) None mean_file_type : mean_file_path : prototxt_file_path : model : learn_qualipapia_papier_refus_from_vlg_data_aug Inside get_net Inside get_net before cache_data_model model_param file didn't exist Inside get_net before CDM.load_model_par_type model_name : learn_qualipapia_papier_refus_from_vlg_data_aug model_type : caffe list file need : ['caffemodel', 'deploy_conv_normal.prototxt', 'deploy_fc.prototxt', 'deploy.prototxt', 'mean.npy', 'synset_words.txt'] file exist in s3 : ['caffemodel', 'deploy.prototxt', 'mean.npy', 'synset_words.txt'] file manque in s3 : ['deploy_conv_normal.prototxt', 'deploy_fc.prototxt'] local folder : /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug/caffemodel size_local : 44972172 size in s3 : 44972172 create time local : 2021-08-09 05:55:48 create time in s3 : 2021-08-06 19:28:49 caffemodel already exist and didn't need to update /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug/deploy.prototxt size_local : 17311 size in s3 : 17311 create time local : 2021-08-09 05:55:48 create time in s3 : 2021-08-06 19:28:49 deploy.prototxt already exist and didn't need to update /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug/mean.npy size_local : 1572992 size in s3 : 1572992 create time local : 2021-08-09 05:55:48 create time in s3 : 2021-08-06 19:28:51 mean.npy already exist and didn't need to update /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug/synset_words.txt size_local : 57 size in s3 : 57 create time local : 2021-08-09 05:55:48 create time in s3 : 2021-08-06 19:28:49 synset_words.txt already exist and didn't need to update Inside get_net after CDM.load_model_par_type After if not only_with_local_cache: /home/admin/workarea/install/darknet/:/home/admin/workarea/git/Velours/python:/home/admin/workarea/install/caffe_frcnn_python3/py-faster-rcnn/caffe-fast-rcnn/python:/home/admin/mtr/.credentials:/home/admin/workarea/install/caffe/python:/home/admin/workarea/install/caffe_frcnn/py-faster-rcnn/tools/:/home/admin/workarea/git/fotonowerpip/:/home/admin/workarea/install/segment-anything:/home/admin//workarea/git/pyfvs/:/home/admin/workarea/git/apy/ Here before set mode gpu Doing nothing but we could set mode gpu after set mode gpu prototxt_filename : /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug/deploy.prototxt caffemodel_filename : /data/models_weight/learn_qualipapia_papier_refus_from_vlg_data_aug/caffemodel now we set caffe to gpu mode before predict begin to check gpu status inside check gpu memory l 3637 free memory gpu now : 10776 max_wait_temp : 1 max_wait : 0 dict_keys(['prob']) time used to do the prepocess of the images : 0.15992379188537598 time used to do the prediction : 0.06959843635559082 we don't save the descriptors for this thcl 2456 tagging for thcl : 2868 To do loadFromThcl(), then load ParamDescType : thcl2868 thcls : [{'id': 2868, 'mtr_user_id': 31, 'name': 'learn_papier_nantes_300421', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'Carton_brun,Carton_gris,Teint_Dans_La_Masse,autre_refus,cartonnette,environnement,kraft,metal,papier,plastique', 'svm_portfolios_learning': '3752117,3752118,3752123,3752106,3752116,3752124,3752119,3581575,3486029,3752122', 'photo_hashtag_type': 3632, 'photo_desc_type': 5288, 'type_classification': 'caffe', 'hashtag_id_list': '2107753024,2107753020,2107752385,2107752406,702398920,493012381,493202403,492628673,492668766,492725882'}] thcl {'id': 2868, 'mtr_user_id': 31, 'name': 'learn_papier_nantes_300421', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'Carton_brun,Carton_gris,Teint_Dans_La_Masse,autre_refus,cartonnette,environnement,kraft,metal,papier,plastique', 'svm_portfolios_learning': '3752117,3752118,3752123,3752106,3752116,3752124,3752119,3581575,3486029,3752122', 'photo_hashtag_type': 3632, 'photo_desc_type': 5288, 'type_classification': 'caffe', 'hashtag_id_list': '2107753024,2107753020,2107752385,2107752406,702398920,493012381,493202403,492628673,492668766,492725882'} Update svm_hashtag_type_desc : 5288 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (5288, 'learn_papier_nantes_300421', 512, 512, 'learn_papier_nantes_300421', 'pool5', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 1, datetime.datetime(2021, 4, 30, 17, 9, 41), datetime.datetime(2021, 4, 30, 17, 9, 41)) To loadFromThcl() : net_5288 begin to check gpu status inside check gpu memory l 3637 free memory gpu now : 10774 max_wait_temp : 1 max_wait : 0 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (5288, 'learn_papier_nantes_300421', 512, 512, 'learn_papier_nantes_300421', 'pool5', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 1, datetime.datetime(2021, 4, 30, 17, 9, 41), datetime.datetime(2021, 4, 30, 17, 9, 41)) None mean_file_type : mean_file_path : prototxt_file_path : model : learn_papier_nantes_300421 Inside get_net Inside get_net before cache_data_model model_param file didn't exist Inside get_net before CDM.load_model_par_type model_name : learn_papier_nantes_300421 model_type : caffe list file need : ['caffemodel', 'deploy_conv_normal.prototxt', 'deploy_fc.prototxt', 'deploy.prototxt', 'mean.npy', 'synset_words.txt'] file exist in s3 : ['caffemodel', 'deploy.prototxt', 'mean.npy', 'synset_words.txt'] file manque in s3 : ['deploy_conv_normal.prototxt', 'deploy_fc.prototxt'] local folder : /data/models_weight/learn_papier_nantes_300421 /data/models_weight/learn_papier_nantes_300421/caffemodel size_local : 44791983 size in s3 : 44791983 create time local : 2021-08-09 05:55:59 create time in s3 : 2021-08-06 19:22:13 caffemodel already exist and didn't need to update /data/models_weight/learn_papier_nantes_300421/deploy.prototxt size_local : 17255 size in s3 : 17255 create time local : 2021-08-09 05:55:59 create time in s3 : 2021-08-06 19:22:12 deploy.prototxt already exist and didn't need to update /data/models_weight/learn_papier_nantes_300421/mean.npy size_local : 1572992 size in s3 : 1572992 create time local : 2021-08-09 05:55:59 create time in s3 : 2021-08-06 19:22:14 mean.npy already exist and didn't need to update /data/models_weight/learn_papier_nantes_300421/synset_words.txt size_local : 331 size in s3 : 331 create time local : 2021-08-09 05:56:00 create time in s3 : 2021-08-06 19:22:12 synset_words.txt already exist and didn't need to update Inside get_net after CDM.load_model_par_type After if not only_with_local_cache: /home/admin/workarea/install/darknet/:/home/admin/workarea/git/Velours/python:/home/admin/workarea/install/caffe_frcnn_python3/py-faster-rcnn/caffe-fast-rcnn/python:/home/admin/mtr/.credentials:/home/admin/workarea/install/caffe/python:/home/admin/workarea/install/caffe_frcnn/py-faster-rcnn/tools/:/home/admin/workarea/git/fotonowerpip/:/home/admin/workarea/install/segment-anything:/home/admin//workarea/git/pyfvs/:/home/admin/workarea/git/apy/ Here before set mode gpu Doing nothing but we could set mode gpu after set mode gpu prototxt_filename : /data/models_weight/learn_papier_nantes_300421/deploy.prototxt caffemodel_filename : /data/models_weight/learn_papier_nantes_300421/caffemodel now we set caffe to gpu mode before predict begin to check gpu status inside check gpu memory l 3637 free memory gpu now : 10774 max_wait_temp : 1 max_wait : 0 dict_keys(['prob']) time used to do the prepocess of the images : 0.16041064262390137 time used to do the prediction : 0.05373239517211914 we don't save the descriptors for this thcl 2868 Inside saveOutput : final : False verbose : 0 time used to find the portfolios of the photos test new format of the output of the step_thcl begin to insert list_values into class_photo_scores : length of list_valuse in save_photo_hashtag_id_thcl_score : 0 time used for this insertion : 1.1205673217773438e-05 save missing photos in datou_result : time spend for datou_step_exec : 7.73178243637085 time spend to save output : 0.0001938343048095703 total time spend for step 3 : 7.731976270675659 step4:merge_mask_thcl_custom Fri Oct 10 09:28:36 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed We should have FATAL ERROR but same_nb_input_output==True : this should be an optionnal input ! complete output_args for input 1 Inconsistent number of input and output, step which parrallelize and manage error in input by avoiding sending an output for this data can't be used in tree dependencies of input and output complete output_args for input 2 VR 22-3-18 : For now we do not clean correctly the datou structure Begin step merge_mask_thcl_custom batch 1 Loaded 28 chid ids of type : 3663 +++++++++++++++++++++++++++++++++++++++++++As expected we have just one thcl present End of step merge_mask_thcl_custom Inside saveOutput : final : False verbose : 0 saveOutput not yet implemented for datou_step.type : merge_mask_thcl_custom we use saveGeneral [1388575902, 1388575900, 1388575898] Looping around the photos to save general results len do output : 3 /1388575902Didn't retrieve data .Didn't retrieve data . /1388575900Didn't retrieve data .Didn't retrieve data . /1388575898Didn't retrieve data .Didn't retrieve data . before output type Here is an output not treated by saveGeneral : Here is an output not treated by saveGeneral : Managing all output in save final without adding information in the mtr_datou_result ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575902', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575900', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575898', None, None, None, None, None, '3911038') begin to insert list_values into mtr_datou_result : length of list_values in save_final : 9 time used for this insertion : 0.04321646690368652 save_final save missing photos in datou_result : time spend for datou_step_exec : 0.19273805618286133 time spend to save output : 0.04353952407836914 total time spend for step 4 : 0.23627758026123047 step5:rle_unique_nms_with_priority Fri Oct 10 09:28:36 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed complete output_args for input 0 VR 22-3-18 : For now we do not clean correctly the datou structure Begin step rle-unique-nms nb_obj : 3 nb_hashtags : 2 time to prepare the origin masks : 0.24552559852600098 time for calcul the mask position with numpy : 0.009536027908325195 nb_pixel_total : 729394 time to create 1 rle with new method : 0.03233146667480469 time for calcul the mask position with numpy : 0.011987924575805664 nb_pixel_total : 818269 time to create 1 rle with new method : 0.034592390060424805 time for calcul the mask position with numpy : 0.0058040618896484375 nb_pixel_total : 124081 time to create 1 rle with old method : 0.1341545581817627 time for calcul the mask position with numpy : 0.005588054656982422 nb_pixel_total : 90816 time to create 1 rle with old method : 0.09766936302185059 create new chi : 0.33728599548339844 time to delete rle : 0.06765055656433105 batch 1 Loaded 4 chid ids of type : 3726 Number RLEs to save : 5420 TO DO : save crop sub photo not yet done ! save time : 0.6678965091705322 nb_obj : 5 nb_hashtags : 4 time to prepare the origin masks : 0.6255433559417725 time for calcul the mask position with numpy : 0.010115623474121094 nb_pixel_total : 698171 time to create 1 rle with new method : 0.030381202697753906 time for calcul the mask position with numpy : 0.00945734977722168 nb_pixel_total : 470647 time to create 1 rle with new method : 0.03185582160949707 time for calcul the mask position with numpy : 0.004982948303222656 nb_pixel_total : 36026 time to create 1 rle with old method : 0.0373992919921875 time for calcul the mask position with numpy : 0.006937265396118164 nb_pixel_total : 346883 time to create 1 rle with new method : 0.028203964233398438 time for calcul the mask position with numpy : 0.005949974060058594 nb_pixel_total : 120805 time to create 1 rle with old method : 0.1275017261505127 time for calcul the mask position with numpy : 0.006058216094970703 nb_pixel_total : 90028 time to create 1 rle with old method : 0.0963892936706543 create new chi : 0.39787793159484863 time to delete rle : 0.000675201416015625 batch 1 Loaded 6 chid ids of type : 3726 Number RLEs to save : 7275 TO DO : save crop sub photo not yet done ! save time : 0.8406391143798828 nb_obj : 16 nb_hashtags : 5 time to prepare the origin masks : 0.9490389823913574 time for calcul the mask position with numpy : 0.016988515853881836 nb_pixel_total : 1220928 time to create 1 rle with new method : 0.03164792060852051 time for calcul the mask position with numpy : 0.0049953460693359375 nb_pixel_total : 650 time to create 1 rle with old method : 0.0007290840148925781 time for calcul the mask position with numpy : 0.004682302474975586 nb_pixel_total : 557 time to create 1 rle with old method : 0.0006382465362548828 time for calcul the mask position with numpy : 0.004729747772216797 nb_pixel_total : 794 time to create 1 rle with old method : 0.0008997917175292969 time for calcul the mask position with numpy : 0.004827976226806641 nb_pixel_total : 4637 time to create 1 rle with old method : 0.004850864410400391 time for calcul the mask position with numpy : 0.004688739776611328 nb_pixel_total : 944 time to create 1 rle with old method : 0.0009837150573730469 time for calcul the mask position with numpy : 0.00496220588684082 nb_pixel_total : 10859 time to create 1 rle with old method : 0.011282920837402344 time for calcul the mask position with numpy : 0.00481867790222168 nb_pixel_total : 6562 time to create 1 rle with old method : 0.007170915603637695 time for calcul the mask position with numpy : 0.007075786590576172 nb_pixel_total : 400518 time to create 1 rle with new method : 0.025801658630371094 time for calcul the mask position with numpy : 0.004919767379760742 nb_pixel_total : 290 time to create 1 rle with old method : 0.0003345012664794922 time for calcul the mask position with numpy : 0.004615306854248047 nb_pixel_total : 889 time to create 1 rle with old method : 0.000934600830078125 time for calcul the mask position with numpy : 0.004878044128417969 nb_pixel_total : 54 time to create 1 rle with old method : 9.822845458984375e-05 time for calcul the mask position with numpy : 0.00465703010559082 nb_pixel_total : 8374 time to create 1 rle with old method : 0.008705377578735352 time for calcul the mask position with numpy : 0.0047931671142578125 nb_pixel_total : 2281 time to create 1 rle with old method : 0.002605915069580078 time for calcul the mask position with numpy : 0.0055005550384521484 nb_pixel_total : 98352 time to create 1 rle with old method : 0.10634422302246094 time for calcul the mask position with numpy : 0.005516767501831055 nb_pixel_total : 371 time to create 1 rle with old method : 0.0004367828369140625 time for calcul the mask position with numpy : 0.0049474239349365234 nb_pixel_total : 5500 time to create 1 rle with old method : 0.00624537467956543 create new chi : 0.31009531021118164 time to delete rle : 0.0007317066192626953 batch 1 Loaded 17 chid ids of type : 3726 Number RLEs to save : 6028 TO DO : save crop sub photo not yet done ! save time : 0.7455987930297852 map_output_result : {1388575902: (0.0, 'Should be the crop_list due to order', 0.0), 1388575900: (0.0, 'Should be the crop_list due to order', 0.0), 1388575898: (0.0, 'Should be the crop_list due to order', 0.0)} End step rle-unique-nms Inside saveOutput : final : False verbose : 0 saveOutput not yet implemented for datou_step.type : rle_unique_nms_with_priority we use saveGeneral [1388575902, 1388575900, 1388575898] Looping around the photos to save general results len do output : 3 /1388575902.Didn't retrieve data . /1388575900.Didn't retrieve data . /1388575898.Didn't retrieve data . before output type Used above Here is an output not treated by saveGeneral : Managing all output in save final without adding information in the mtr_datou_result ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575902', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575900', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575898', None, None, None, None, None, '3911038') begin to insert list_values into mtr_datou_result : length of list_values in save_final : 9 time used for this insertion : 0.03483247756958008 save_final save missing photos in datou_result : time spend for datou_step_exec : 5.324903249740601 time spend to save output : 0.03520393371582031 total time spend for step 5 : 5.360107183456421 step6:crop_condition Fri Oct 10 09:28:42 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed We should have FATAL ERROR but same_nb_input_output==True : this should be an optionnal input ! We should have FATAL ERROR but same_nb_input_output==True : this should be an optionnal input ! VR 22-3-18 : For now we do not clean correctly the datou structure some photos are not treated, begin crop_condition Loading chi in step crop with photo_hashtag_type : 3726 Loading chi in step crop for list_pids : 3 ! batch 1 Loaded 27 chid ids of type : 3726 begin to crop the class : teint_dans_la_masse param for this class : {'min_score': 0.7} filtre for class : teint_dans_la_masse hashtag_id of this class : 2107752385 map_result returned by crop_photo_return_map_crop : length : 0 About to insert : list_path_to_insert length 0 new photo from crops ! About to upload 0 photos WARNING : list_path_to_insert is empty, cannot upload ! we have finished the crop for the class : teint_dans_la_masse begin to crop the class : autre_refus param for this class : {'min_score': 0.5} filtre for class : autre_refus hashtag_id of this class : 2107752406 Next one ! map_result returned by crop_photo_return_map_crop : length : 1 About to insert : list_path_to_insert length 1 new photo from crops ! About to upload 1 photos upload in portfolio : 4869462 init cache_photo without model_param we have 1 photo to upload uploaded to storage server : ovh folder_temporaire : temp/1760081323_4124462 INSERT INTO MTRBack.photos (`timeStamp`, `latitude`, `longitude`, `right_categories`, `tags`, `speed`, `size`, `text`, `altitude`, `width`, `height`, `score`, `created_at`,`source_id`,`place_id`) VALUES (FROM_UNIXTIME(1760081323), 0.0, 0.0, 14, '', 0, 0, '1760081282_4124462_1388575900_382f59e38c0f7cb2b91d0a566592768c_rle_crop_3991852810_0.png', 0, 233, 199, 0, 1760081323,'0',0) batch_size : 0, verbose : False, strat_bulk_insert : ignore_different_from_first This is a hack ! we have uploaded 1 photos in the portfolio 4869462 time of upload the photos Elapsed time : 0.794111967086792 we have finished the crop for the class : autre_refus begin to crop the class : carton_gris param for this class : {'min_score': 0.5} filtre for class : carton_gris hashtag_id of this class : 2107753020 begin to crop the class : cartonnette param for this class : {'min_score': 0.5} filtre for class : cartonnette hashtag_id of this class : 702398920 begin to crop the class : carton_brun param for this class : {'min_score': 0.7} filtre for class : carton_brun hashtag_id of this class : 2107753024 Next one ! map_result returned by crop_photo_return_map_crop : length : 1 About to insert : list_path_to_insert length 1 new photo from crops ! About to upload 1 photos upload in portfolio : 4869462 init cache_photo without model_param we have 1 photo to upload uploaded to storage server : ovh folder_temporaire : temp/1760081324_4124462 INSERT INTO MTRBack.photos (`timeStamp`, `latitude`, `longitude`, `right_categories`, `tags`, `speed`, `size`, `text`, `altitude`, `width`, `height`, `score`, `created_at`,`source_id`,`place_id`) VALUES (FROM_UNIXTIME(1760081324), 0.0, 0.0, 14, '', 0, 0, '1760081282_4124462_1388575898_9bbec473c050c6f6830cae8871fbfe5b_rle_crop_3991852839_0.png', 0, 111, 132, 0, 1760081324,'0',0) batch_size : 0, verbose : False, strat_bulk_insert : ignore_different_from_first This is a hack ! we have uploaded 1 photos in the portfolio 4869462 time of upload the photos Elapsed time : 0.7118642330169678 we have finished the crop for the class : carton_brun begin to crop the class : plastique param for this class : {'min_score': 0.5} filtre for class : plastique hashtag_id of this class : 492725882 begin to crop the class : kraft param for this class : {'min_score': 0.5} filtre for class : kraft hashtag_id of this class : 493202403 begin to crop the class : metal param for this class : {'min_score': 0.5} filtre for class : metal hashtag_id of this class : 492628673 delete rles for these photos Inside saveOutput : final : False verbose : 0 saveOutput not yet implemented for datou_step.type : crop_condition we use saveGeneral [1388575902, 1388575900, 1388575898] Looping around the photos to save general results len do output : 2 /1388632331Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . /1388632332Didn't retrieve data .Didn't retrieve data .Didn't retrieve data . before output type Here is an output not treated by saveGeneral : Here is an output not treated by saveGeneral : Here is an output not treated by saveGeneral : Managing all output in save final without adding information in the mtr_datou_result ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575902', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575900', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575898', None, None, None, None, None, '3911038') begin to insert list_values into mtr_datou_result : length of list_values in save_final : 9 time used for this insertion : 0.03768634796142578 save_final save missing photos in datou_result : time spend for datou_step_exec : 2.639345645904541 time spend to save output : 0.03790760040283203 total time spend for step 6 : 2.677253246307373 step7:ventilate_hashtags_in_portfolio Fri Oct 10 09:28:44 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed We should have FATAL ERROR but same_nb_input_output==True : this should be an optionnal input ! VR 22-3-18 : For now we do not clean correctly the datou structure beginning of datou step ventilate_hashtags_in_portfolio : To implement ! To do loadFromThcl(), then load ParamDescType : thcl2725 thcls : [{'id': 2725, 'mtr_user_id': 31, 'name': 'learn_qualipapia_rle_210302_2', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'Carton_brun,Carton_gris,Teint_Dans_La_Masse,autre_refus,cartonnette,environnement,kraft,metal,papier,plastique', 'svm_portfolios_learning': '3460440,3460441,3460446,3460434,3460439,3467416,3460442,3460443,3486028,3460445', 'photo_hashtag_type': 3410, 'photo_desc_type': 5186, 'type_classification': 'caffe', 'hashtag_id_list': '2107753024,2107753020,2107752385,2107752406,702398920,493012381,493202403,492628673,492668766,492725882'}] thcl {'id': 2725, 'mtr_user_id': 31, 'name': 'learn_qualipapia_rle_210302_2', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'Carton_brun,Carton_gris,Teint_Dans_La_Masse,autre_refus,cartonnette,environnement,kraft,metal,papier,plastique', 'svm_portfolios_learning': '3460440,3460441,3460446,3460434,3460439,3467416,3460442,3460443,3486028,3460445', 'photo_hashtag_type': 3410, 'photo_desc_type': 5186, 'type_classification': 'caffe', 'hashtag_id_list': '2107753024,2107753020,2107752385,2107752406,702398920,493012381,493202403,492628673,492668766,492725882'} Update svm_hashtag_type_desc : 5186 Iterating over portfolio : 27666970 get user id for portfolio 27666970 SELECT mptpi.id, mptpi.mtr_portfolio_id_1, mptpi.mtr_portfolio_id_2, mptpi.type, mptpi.hashtag_id, mptpi.min_score, mptpi.mtr_user_id, mptpi.created_at, mptpi.updated_at, mptpi.last_updated_at_desc, mptpi.last_updated_at_asc, h.hashtag FROM MTRPhoto.mtr_port_to_port_ids mptpi, MTRBack.hashtags h WHERE h.hashtag_id=mptpi.hashtag_id AND mptpi.`mtr_portfolio_id_1`=27666970 AND mptpi.`type`=3726 AND mptpi.`hashtag_id` in (select hashtag_id FROM MTRBack.hashtags where hashtag in ('metal','kraft','cartonnette','autre_refus','flou','Carton_gris','Carton_brun','papier','mal_croppe','Teint_Dans_La_Masse','environnement','plastique')) AND mptpi.`min_score`=0.5 To do To do SELECT mptpi.id, mptpi.mtr_portfolio_id_1, mptpi.mtr_portfolio_id_2, mptpi.type, mptpi.hashtag_id, mptpi.min_score, mptpi.mtr_user_id, mptpi.created_at, mptpi.updated_at, mptpi.last_updated_at_desc, mptpi.last_updated_at_asc, h.hashtag FROM MTRPhoto.mtr_port_to_port_ids mptpi, MTRBack.hashtags h WHERE h.hashtag_id=mptpi.hashtag_id AND mptpi.`mtr_portfolio_id_1`=27666970 AND mptpi.`type`=3726 AND mptpi.`hashtag_id` in (select hashtag_id FROM MTRBack.hashtags where hashtag in ('metal','kraft','cartonnette','autre_refus','flou','Carton_gris','Carton_brun','papier','mal_croppe','Teint_Dans_La_Masse','environnement','plastique')) AND mptpi.`min_score`=0.5 To do To do ! Use context local managing function ! SELECT mptpi.id, mptpi.mtr_portfolio_id_1, mptpi.mtr_portfolio_id_2, mptpi.type, mptpi.hashtag_id, mptpi.min_score, mptpi.mtr_user_id, mptpi.created_at, mptpi.updated_at, mptpi.last_updated_at_desc, mptpi.last_updated_at_asc, h.hashtag FROM MTRPhoto.mtr_port_to_port_ids mptpi, MTRBack.hashtags h WHERE h.hashtag_id=mptpi.hashtag_id AND mptpi.`mtr_portfolio_id_1`=27666970 AND mptpi.`type`=4417 AND mptpi.`hashtag_id` in (select hashtag_id FROM MTRBack.hashtags where hashtag in ('metal','kraft','cartonnette','autre_refus','flou','Carton_gris','Carton_brun','papier','mal_croppe','Teint_Dans_La_Masse','environnement','plastique')) AND mptpi.`min_score`=0.5 To do To do SELECT mptpi.id, mptpi.mtr_portfolio_id_1, mptpi.mtr_portfolio_id_2, mptpi.type, mptpi.hashtag_id, mptpi.min_score, mptpi.mtr_user_id, mptpi.created_at, mptpi.updated_at, mptpi.last_updated_at_desc, mptpi.last_updated_at_asc, h.hashtag FROM MTRPhoto.mtr_port_to_port_ids mptpi, MTRBack.hashtags h WHERE h.hashtag_id=mptpi.hashtag_id AND mptpi.`mtr_portfolio_id_1`=27666970 AND mptpi.`type`=4417 AND mptpi.`hashtag_id` in (select hashtag_id FROM MTRBack.hashtags where hashtag in ('metal','kraft','cartonnette','autre_refus','flou','Carton_gris','Carton_brun','papier','mal_croppe','Teint_Dans_La_Masse','environnement','plastique')) AND mptpi.`min_score`=0.5 To do lien utilise dans velours : https://marlene.fotonower.com/velours/27685934,27685935,27685936,27685937,27685938,27685939,27685940,27685941,27685942,27685943,27685944,27685945?tags=metal,kraft,cartonnette,autre_refus,flou,Carton_gris,Carton_brun,papier,mal_croppe,Teint_Dans_La_Masse,environnement,plastique Inside saveOutput : final : False verbose : 0 saveOutput not yet implemented for datou_step.type : ventilate_hashtags_in_portfolio we use saveGeneral [1388575902, 1388575900, 1388575898] Looping around the photos to save general results len do output : 1 /27666970. before output type Here is an output not treated by saveGeneral : Managing all output in save final without adding information in the mtr_datou_result ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575902', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575900', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575898', None, None, None, None, None, '3911038') begin to insert list_values into mtr_datou_result : length of list_values in save_final : 4 time used for this insertion : 0.04086637496948242 save_final save missing photos in datou_result : time spend for datou_step_exec : 3.614222288131714 time spend to save output : 0.04107046127319336 total time spend for step 7 : 3.6552927494049072 step8:tfhub_classification2 Fri Oct 10 09:28:48 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed complete output_args for input 0 VR 22-3-18 : For now we do not clean correctly the datou structure Beginning of datou_step TFHub with tf2 ! nombre de thcls : 3 we are using the classfication for multi_thcl [3513, 3379, 3847] begin to check gpu status inside check gpu memory 2025-10-10 09:28:51.182133: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1 2025-10-10 09:28:51.183002: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: pciBusID: 0000:41:00.0 name: NVIDIA GeForce RTX 2080 Ti computeCapability: 7.5 coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s 2025-10-10 09:28:51.183064: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:51.183109: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:28:51.184782: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10 2025-10-10 09:28:51.184835: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10 2025-10-10 09:28:51.186855: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10 2025-10-10 09:28:51.187807: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10 2025-10-10 09:28:51.191575: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2025-10-10 09:28:51.193034: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0 2025-10-10 09:28:51.193400: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA 2025-10-10 09:28:51.218489: I tensorflow/core/platform/profile_utils/cpu_utils.cc:102] CPU Frequency: 3493010000 Hz 2025-10-10 09:28:51.220302: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7f217c000b60 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2025-10-10 09:28:51.220334: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version 2025-10-10 09:28:51.224290: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x4c61fd80 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2025-10-10 09:28:51.224313: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA GeForce RTX 2080 Ti, Compute Capability 7.5 2025-10-10 09:28:51.225271: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: pciBusID: 0000:41:00.0 name: NVIDIA GeForce RTX 2080 Ti computeCapability: 7.5 coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s 2025-10-10 09:28:51.225327: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:51.225348: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:28:51.225413: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10 2025-10-10 09:28:51.225435: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10 2025-10-10 09:28:51.225462: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10 2025-10-10 09:28:51.225491: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10 2025-10-10 09:28:51.225521: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2025-10-10 09:28:51.229727: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0 2025-10-10 09:28:51.229812: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 2025-10-10 09:28:51.229863: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix: 2025-10-10 09:28:51.229876: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108] 0 2025-10-10 09:28:51.229885: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0: N 2025-10-10 09:28:51.231537: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 3096 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 2080 Ti, pci bus id: 0000:41:00.0, compute capability: 7.5) l 3637 free memory gpu now : 10774 max_wait_temp : 1 max_wait : 5 1 Physical GPUs, 1 Logical GPUs tagging for thcl : 3513 To do loadFromThcl(), then load ParamDescType : thcl3513 thcls : [{'id': 3513, 'mtr_user_id': 31, 'name': 'Rungis_amount_dechets_fall_2018_v2_tf', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': '05102018_Papier_non_papier_dense,05102018_Papier_non_papier_peu_dense,05102018_Papier_non_papier_presque_vide,05102018_Papier_non_papier_tres_dense,05102018_Papier_non_papier_tres_peu_dense', 'svm_portfolios_learning': '1108385,1108386,1108388,1108384,1108387', 'photo_hashtag_type': 4557, 'photo_desc_type': 5767, 'type_classification': 'tf_classification2', 'hashtag_id_list': '2107751013,2107751014,2107751015,2107751016,2107751017'}] thcl {'id': 3513, 'mtr_user_id': 31, 'name': 'Rungis_amount_dechets_fall_2018_v2_tf', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': '05102018_Papier_non_papier_dense,05102018_Papier_non_papier_peu_dense,05102018_Papier_non_papier_presque_vide,05102018_Papier_non_papier_tres_dense,05102018_Papier_non_papier_tres_peu_dense', 'svm_portfolios_learning': '1108385,1108386,1108388,1108384,1108387', 'photo_hashtag_type': 4557, 'photo_desc_type': 5767, 'type_classification': 'tf_classification2', 'hashtag_id_list': '2107751013,2107751014,2107751015,2107751016,2107751017'} Update svm_hashtag_type_desc : 5767 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (5767, 'Rungis_amount_dechets_fall_2018_v2_tf', 2048, 2048, 'Rungis_amount_dechets_fall_2018_v2_tf', 'pool5', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 3, datetime.datetime(2023, 3, 16, 15, 52, 10), datetime.datetime(2023, 3, 16, 15, 52, 10)) model_name : Rungis_amount_dechets_fall_2018_v2_tf model_param file didn't exist model_name : Rungis_amount_dechets_fall_2018_v2_tf model_type : tf_classification2 list file need : ['Confusion_Matrix.png', 'Precision_Recall_05102018_Papier_non_papier_dense.jpg', 'Precision_Recall_05102018_Papier_non_papier_peu_dense.jpg', 'Precision_Recall_05102018_Papier_non_papier_presque_vide.jpg', 'Precision_Recall_05102018_Papier_non_papier_tres_dense.jpg', 'Precision_Recall_05102018_Papier_non_papier_tres_peu_dense.jpg', 'Result_Summary.txt', 'checkpoint', 'model_checkpoint.ckpt.data-00000-of-00001', 'model_checkpoint.ckpt.data-00000-of-00002', 'model_checkpoint.ckpt.data-00001-of-00002', 'model_checkpoint.ckpt.index', 'model_weights.h5'] file exist in s3 : ['Confusion_Matrix.png', 'Precision_Recall_05102018_Papier_non_papier_dense.jpg', 'Precision_Recall_05102018_Papier_non_papier_peu_dense.jpg', 'Precision_Recall_05102018_Papier_non_papier_presque_vide.jpg', 'Precision_Recall_05102018_Papier_non_papier_tres_dense.jpg', 'Precision_Recall_05102018_Papier_non_papier_tres_peu_dense.jpg', 'Result_Summary.txt', 'checkpoint', 'model_checkpoint.ckpt.data-00000-of-00001', 'model_checkpoint.ckpt.data-00000-of-00002', 'model_checkpoint.ckpt.data-00001-of-00002', 'model_checkpoint.ckpt.index', 'model_weights.h5'] file manque in s3 : [] local folder : /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Confusion_Matrix.png size_local : 67810 size in s3 : 67810 create time local : 2023-10-30 16:21:37 create time in s3 : 2023-10-30 14:09:29 Confusion_Matrix.png already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Precision_Recall_05102018_Papier_non_papier_dense.jpg size_local : 73949 size in s3 : 73949 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:30 Precision_Recall_05102018_Papier_non_papier_dense.jpg already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Precision_Recall_05102018_Papier_non_papier_peu_dense.jpg size_local : 85572 size in s3 : 85572 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:39 Precision_Recall_05102018_Papier_non_papier_peu_dense.jpg already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Precision_Recall_05102018_Papier_non_papier_presque_vide.jpg size_local : 72361 size in s3 : 72361 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:37 Precision_Recall_05102018_Papier_non_papier_presque_vide.jpg already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Precision_Recall_05102018_Papier_non_papier_tres_dense.jpg size_local : 83567 size in s3 : 83567 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:48 Precision_Recall_05102018_Papier_non_papier_tres_dense.jpg already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Precision_Recall_05102018_Papier_non_papier_tres_peu_dense.jpg size_local : 71611 size in s3 : 71611 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:29 Precision_Recall_05102018_Papier_non_papier_tres_peu_dense.jpg already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/Result_Summary.txt size_local : 1058 size in s3 : 1058 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:30 Result_Summary.txt already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/checkpoint size_local : 99 size in s3 : 99 create time local : 2023-10-30 16:21:38 create time in s3 : 2023-10-30 14:09:36 checkpoint already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/model_checkpoint.ckpt.data-00000-of-00001 size_local : 188538519 size in s3 : 188538519 create time local : 2023-10-30 16:21:41 create time in s3 : 2023-10-30 14:09:40 model_checkpoint.ckpt.data-00000-of-00001 already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/model_checkpoint.ckpt.data-00000-of-00002 size_local : 216572 size in s3 : 216572 create time local : 2023-10-30 16:21:41 create time in s3 : 2023-03-16 14:52:09 model_checkpoint.ckpt.data-00000-of-00002 already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/model_checkpoint.ckpt.data-00001-of-00002 size_local : 32279708 size in s3 : 32279708 create time local : 2023-10-30 16:21:42 create time in s3 : 2023-03-16 14:52:07 model_checkpoint.ckpt.data-00001-of-00002 already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/model_checkpoint.ckpt.index size_local : 28001 size in s3 : 28001 create time local : 2023-10-30 16:21:42 create time in s3 : 2023-10-30 14:09:30 model_checkpoint.ckpt.index already exist and didn't need to update /data/models_weight/Rungis_amount_dechets_fall_2018_v2_tf/model_weights.h5 size_local : 94501976 size in s3 : 94501976 create time local : 2023-10-30 16:21:44 create time in s3 : 2023-10-30 14:09:37 model_weights.h5 already exist and didn't need to update desc size : 2048 Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= module (KerasLayer) (None, 2048) 23561152 _________________________________________________________________ Rungis_amount_dechets_fall_2 (None, 5) 10245 ================================================================= Total params: 23,571,397 Trainable params: 10,245 Non-trainable params: 23,561,152 _________________________________________________________________ Loading Weights... time used to create the model : 7.57407808303833 time used to load_weights : 0.13528919219970703 0it [00:00, ?it/s] 3it [00:00, 3389.79it/s]2025-10-10 09:29:00.859583: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10 2025-10-10 09:29:00.887082: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 temp/1760081282_4124462_1388575902_47351dac3c30c89ff02bbbfc8556ea14.jpg temp/1760081282_4124462_1388575900_382f59e38c0f7cb2b91d0a566592768c.jpg temp/1760081282_4124462_1388575898_9bbec473c050c6f6830cae8871fbfe5b.jpg Found 3 images belonging to 1 classes. begin to do the prediction : time used to do the prediction : 2.1280148029327393 (3,) (3, 5) (3, 2048) shape of features : (3, 2048) shape of new features : (1, 3, 2048) save descriptor for thcl : 3513 time to traite the descriptors : 0.029751300811767578 storage_type for insertDescriptorsMulti : 3 To insert : 1388575902 To insert : 1388575900 To insert : 1388575898 Catched exception ! Connect or reconnect ! time to insert the descriptors : 2.395127296447754 tagging for thcl : 3379 To do loadFromThcl(), then load ParamDescType : thcl3379 thcls : [{'id': 3379, 'mtr_user_id': 31, 'name': 'learn_classif_flux_maj_generique_effnet_v2_s_02062022', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'aluminium,ela,film_pedb,flux_dev,jrm,pcm,pcnc,pehd_pp,pet_clair,refus,tapis_vide', 'svm_portfolios_learning': '5515864,5515840,5515844,5515850,6244400,6237996,6237998,5515847,5515841,5515868,5515866', 'photo_hashtag_type': 4374, 'photo_desc_type': 5680, 'type_classification': 'tf_classification2', 'hashtag_id_list': '493546845,492741797,2107760237,2107760238,495916461,560181804,1284539308,2107760239,2107755846,538914404,2107748999'}] thcl {'id': 3379, 'mtr_user_id': 31, 'name': 'learn_classif_flux_maj_generique_effnet_v2_s_02062022', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'aluminium,ela,film_pedb,flux_dev,jrm,pcm,pcnc,pehd_pp,pet_clair,refus,tapis_vide', 'svm_portfolios_learning': '5515864,5515840,5515844,5515850,6244400,6237996,6237998,5515847,5515841,5515868,5515866', 'photo_hashtag_type': 4374, 'photo_desc_type': 5680, 'type_classification': 'tf_classification2', 'hashtag_id_list': '493546845,492741797,2107760237,2107760238,495916461,560181804,1284539308,2107760239,2107755846,538914404,2107748999'} Update svm_hashtag_type_desc : 5680 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (5680, 'learn_classif_flux_maj_generique_effnet_v2_s_02062022', 1280, 1280, 'learn_classif_flux_maj_generique_effnet_v2_s_02062022', 'pool5', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 3, datetime.datetime(2022, 6, 4, 1, 31, 10), datetime.datetime(2022, 6, 4, 1, 31, 10)) model_name : learn_classif_flux_maj_generique_effnet_v2_s_02062022 model_param file didn't exist model_name : learn_classif_flux_maj_generique_effnet_v2_s_02062022 model_type : tf_classification2 list file need : ['Confusion_Matrix.png', 'Precision_Recall_aluminium.jpg', 'Precision_Recall_ela.jpg', 'Precision_Recall_film_pedb.jpg', 'Precision_Recall_flux_dev.jpg', 'Precision_Recall_jrm.jpg', 'Precision_Recall_pcm.jpg', 'Precision_Recall_pcnc.jpg', 'Precision_Recall_pehd_pp.jpg', 'Precision_Recall_pet_clair.jpg', 'Precision_Recall_refus.jpg', 'Precision_Recall_tapis_vide.jpg', 'Result_Summary.txt', 'checkpoint', 'model_checkpoint.ckpt.data-00000-of-00002', 'model_checkpoint.ckpt.data-00001-of-00002', 'model_checkpoint.ckpt.index', 'model_weights.h5'] file exist in s3 : ['Confusion_Matrix.png', 'Precision_Recall_aluminium.jpg', 'Precision_Recall_ela.jpg', 'Precision_Recall_film_pedb.jpg', 'Precision_Recall_flux_dev.jpg', 'Precision_Recall_jrm.jpg', 'Precision_Recall_pcm.jpg', 'Precision_Recall_pcnc.jpg', 'Precision_Recall_pehd_pp.jpg', 'Precision_Recall_pet_clair.jpg', 'Precision_Recall_refus.jpg', 'Precision_Recall_tapis_vide.jpg', 'Result_Summary.txt', 'checkpoint', 'model_checkpoint.ckpt.data-00000-of-00002', 'model_checkpoint.ckpt.data-00001-of-00002', 'model_checkpoint.ckpt.index', 'model_weights.h5'] file manque in s3 : [] local folder : /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022 /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Confusion_Matrix.png size_local : 82527 size in s3 : 82527 create time local : 2022-06-06 13:01:24 create time in s3 : 2022-06-03 23:31:09 Confusion_Matrix.png already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_aluminium.jpg size_local : 67602 size in s3 : 67602 create time local : 2022-06-06 13:01:24 create time in s3 : 2022-06-03 23:23:41 Precision_Recall_aluminium.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_ela.jpg size_local : 67189 size in s3 : 67189 create time local : 2022-06-06 13:01:25 create time in s3 : 2022-06-03 23:26:13 Precision_Recall_ela.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_film_pedb.jpg size_local : 71280 size in s3 : 71280 create time local : 2022-06-06 13:01:25 create time in s3 : 2022-06-03 23:26:14 Precision_Recall_film_pedb.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_flux_dev.jpg size_local : 59936 size in s3 : 59936 create time local : 2022-06-06 13:01:25 create time in s3 : 2022-06-03 23:23:41 Precision_Recall_flux_dev.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_jrm.jpg size_local : 67657 size in s3 : 67657 create time local : 2022-06-06 13:01:25 create time in s3 : 2022-06-03 23:31:06 Precision_Recall_jrm.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_pcm.jpg size_local : 83042 size in s3 : 83042 create time local : 2022-06-06 13:01:25 create time in s3 : 2022-06-03 23:23:40 Precision_Recall_pcm.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_pcnc.jpg size_local : 69972 size in s3 : 69972 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:31:05 Precision_Recall_pcnc.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_pehd_pp.jpg size_local : 67313 size in s3 : 67313 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:23:45 Precision_Recall_pehd_pp.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_pet_clair.jpg size_local : 67320 size in s3 : 67320 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:31:04 Precision_Recall_pet_clair.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_refus.jpg size_local : 59936 size in s3 : 59936 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:31:08 Precision_Recall_refus.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Precision_Recall_tapis_vide.jpg size_local : 80244 size in s3 : 80244 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:23:45 Precision_Recall_tapis_vide.jpg already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/Result_Summary.txt size_local : 1018 size in s3 : 1018 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:31:08 Result_Summary.txt already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/checkpoint size_local : 99 size in s3 : 99 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:31:06 checkpoint already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/model_checkpoint.ckpt.data-00000-of-00002 size_local : 216640 size in s3 : 216640 create time local : 2022-06-06 13:01:27 create time in s3 : 2022-06-03 23:23:43 model_checkpoint.ckpt.data-00000-of-00002 already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/model_checkpoint.ckpt.data-00001-of-00002 size_local : 32341196 size in s3 : 32341196 create time local : 2022-06-06 13:01:28 create time in s3 : 2022-06-03 23:26:15 model_checkpoint.ckpt.data-00001-of-00002 already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/model_checkpoint.ckpt.index size_local : 43546 size in s3 : 43546 create time local : 2022-06-06 13:01:28 create time in s3 : 2022-06-03 23:26:13 model_checkpoint.ckpt.index already exist and didn't need to update /data/models_weight/learn_classif_flux_maj_generique_effnet_v2_s_02062022/model_weights.h5 size_local : 16531224 size in s3 : 16531224 create time local : 2022-06-06 13:01:29 create time in s3 : 2022-06-03 23:23:46 model_weights.h5 already exist and didn't need to update desc size : 1280 Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= module (KerasLayer) (None, 1280) 4049564 _________________________________________________________________ learn_classif_flux_maj_gener (None, 11) 14091 ================================================================= Total params: 4,063,655 Trainable params: 14,091 Non-trainable params: 4,049,564 _________________________________________________________________ Loading Weights... time used to create the model : 6.909042119979858 time used to load_weights : 0.10359477996826172 0it [00:00, ?it/s] 3it [00:00, 3231.36it/s]temp/1760081282_4124462_1388575902_47351dac3c30c89ff02bbbfc8556ea14.jpg temp/1760081282_4124462_1388575900_382f59e38c0f7cb2b91d0a566592768c.jpg temp/1760081282_4124462_1388575898_9bbec473c050c6f6830cae8871fbfe5b.jpg Found 3 images belonging to 1 classes. begin to do the prediction : time used to do the prediction : 0.9784402847290039 (3,) (3, 11) (3, 1280) shape of features : (3, 1280) shape of new features : (1, 3, 1280) save descriptor for thcl : 3379 time to traite the descriptors : 0.03413891792297363 storage_type for insertDescriptorsMulti : 3 To insert : 1388575902 To insert : 1388575900 To insert : 1388575898 time to insert the descriptors : 1.0356597900390625 tagging for thcl : 3847 To do loadFromThcl(), then load ParamDescType : thcl3847 thcls : [{'id': 3847, 'mtr_user_id': 31, 'name': 'learn_MM_generique_050224', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'aluminium,ela,emr,film_pedb,flux_dev,jrm,pcm,pcnc,pehd_pp,pet_clair,refus,tapis_vide', 'svm_portfolios_learning': '13096157,13096155,13096163,13096159,13301956,13095886,13096162,13096160,13358264,13096158,5515868,13276803', 'photo_hashtag_type': 4932, 'photo_desc_type': 6032, 'type_classification': 'tf_classification2', 'hashtag_id_list': '493546845,492741797,616987804,2107760237,2107760238,495916461,560181804,1284539308,2107760239,2107755846,538914404,2107748999'}] thcl {'id': 3847, 'mtr_user_id': 31, 'name': 'learn_MM_generique_050224', 'pb_hashtag_id': 0, 'live': b'\x00', 'list_hashtags': 'aluminium,ela,emr,film_pedb,flux_dev,jrm,pcm,pcnc,pehd_pp,pet_clair,refus,tapis_vide', 'svm_portfolios_learning': '13096157,13096155,13096163,13096159,13301956,13095886,13096162,13096160,13358264,13096158,5515868,13276803', 'photo_hashtag_type': 4932, 'photo_desc_type': 6032, 'type_classification': 'tf_classification2', 'hashtag_id_list': '493546845,492741797,616987804,2107760237,2107760238,495916461,560181804,1284539308,2107760239,2107755846,538914404,2107748999'} Update svm_hashtag_type_desc : 6032 FOUND : 1 Here is data_from_sql_as_vec to set the ParamDescriptorType : (6032, 'learn_MM_generique_050224', 1280, 1280, 'learn_MM_generique_050224', 'pool5', 10.0, None, None, 256, None, 0, None, 8, None, None, -1000.0, 3, datetime.datetime(2024, 2, 5, 21, 42, 4), datetime.datetime(2024, 2, 5, 21, 42, 4)) model_name : learn_MM_generique_050224 model_param file didn't exist model_name : learn_MM_generique_050224 model_type : tf_classification2 list file need : ['Confusion_Matrix.png', 'Precision_Recall_aluminium.jpg', 'Precision_Recall_ela.jpg', 'Precision_Recall_emr.jpg', 'Precision_Recall_film_pedb.jpg', 'Precision_Recall_flux_dev.jpg', 'Precision_Recall_jrm.jpg', 'Precision_Recall_pcm.jpg', 'Precision_Recall_pcnc.jpg', 'Precision_Recall_pehd_pp.jpg', 'Precision_Recall_pet_clair.jpg', 'Precision_Recall_refus.jpg', 'Precision_Recall_tapis_vide.jpg', 'Result_Summary.txt', 'checkpoint', 'model_checkpoint.ckpt.data-00000-of-00001', 'model_checkpoint.ckpt.index', 'model_weights.h5'] file exist in s3 : ['Confusion_Matrix.png', 'Precision_Recall_aluminium.jpg', 'Precision_Recall_ela.jpg', 'Precision_Recall_emr.jpg', 'Precision_Recall_film_pedb.jpg', 'Precision_Recall_flux_dev.jpg', 'Precision_Recall_jrm.jpg', 'Precision_Recall_pcm.jpg', 'Precision_Recall_pcnc.jpg', 'Precision_Recall_pehd_pp.jpg', 'Precision_Recall_pet_clair.jpg', 'Precision_Recall_refus.jpg', 'Precision_Recall_tapis_vide.jpg', 'Result_Summary.txt', 'checkpoint', 'model_checkpoint.ckpt.data-00000-of-00001', 'model_checkpoint.ckpt.index', 'model_weights.h5'] file manque in s3 : [] local folder : /data/models_weight/learn_MM_generique_050224 /data/models_weight/learn_MM_generique_050224/Confusion_Matrix.png size_local : 88145 size in s3 : 88145 create time local : 2024-04-08 09:17:48 create time in s3 : 2024-02-05 20:41:54 Confusion_Matrix.png already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_aluminium.jpg size_local : 67983 size in s3 : 67983 create time local : 2024-04-08 09:17:48 create time in s3 : 2024-02-05 20:42:02 Precision_Recall_aluminium.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_ela.jpg size_local : 68484 size in s3 : 68484 create time local : 2024-04-08 09:17:48 create time in s3 : 2024-02-05 20:42:04 Precision_Recall_ela.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_emr.jpg size_local : 77930 size in s3 : 77930 create time local : 2024-04-08 09:17:48 create time in s3 : 2024-02-05 20:41:55 Precision_Recall_emr.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_film_pedb.jpg size_local : 66701 size in s3 : 66701 create time local : 2024-04-08 09:17:48 create time in s3 : 2024-02-05 20:41:55 Precision_Recall_film_pedb.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_flux_dev.jpg size_local : 59936 size in s3 : 59936 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:04 Precision_Recall_flux_dev.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_jrm.jpg size_local : 72890 size in s3 : 72890 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:04 Precision_Recall_jrm.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_pcm.jpg size_local : 93916 size in s3 : 93916 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:02 Precision_Recall_pcm.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_pcnc.jpg size_local : 74701 size in s3 : 74701 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:04 Precision_Recall_pcnc.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_pehd_pp.jpg size_local : 68765 size in s3 : 68765 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:03 Precision_Recall_pehd_pp.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_pet_clair.jpg size_local : 69649 size in s3 : 69649 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:03 Precision_Recall_pet_clair.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_refus.jpg size_local : 59936 size in s3 : 59936 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:41:59 Precision_Recall_refus.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Precision_Recall_tapis_vide.jpg size_local : 71529 size in s3 : 71529 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:41:54 Precision_Recall_tapis_vide.jpg already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/Result_Summary.txt size_local : 1076 size in s3 : 1076 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:41:55 Result_Summary.txt already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/checkpoint size_local : 99 size in s3 : 99 create time local : 2024-04-08 09:17:49 create time in s3 : 2024-02-05 20:42:00 checkpoint already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/model_checkpoint.ckpt.data-00000-of-00001 size_local : 32591049 size in s3 : 32591049 create time local : 2024-04-08 09:17:50 create time in s3 : 2024-02-05 20:42:02 model_checkpoint.ckpt.data-00000-of-00001 already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/model_checkpoint.ckpt.index size_local : 42495 size in s3 : 42495 create time local : 2024-04-08 09:17:50 create time in s3 : 2024-02-05 20:41:55 model_checkpoint.ckpt.index already exist and didn't need to update /data/models_weight/learn_MM_generique_050224/model_weights.h5 size_local : 16536200 size in s3 : 16536200 create time local : 2024-04-08 09:17:50 create time in s3 : 2024-02-05 20:42:00 model_weights.h5 already exist and didn't need to update desc size : 1280 Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= module (KerasLayer) (None, 1280) 4049564 _________________________________________________________________ learn_MM_generique_050224den (None, 12) 15372 ================================================================= Total params: 4,064,936 Trainable params: 15,372 Non-trainable params: 4,049,564 _________________________________________________________________ Loading Weights... time used to create the model : 6.9400434494018555 time used to load_weights : 0.10561704635620117 0it [00:00, ?it/s] 3it [00:00, 2482.33it/s]temp/1760081282_4124462_1388575902_47351dac3c30c89ff02bbbfc8556ea14.jpg temp/1760081282_4124462_1388575900_382f59e38c0f7cb2b91d0a566592768c.jpg temp/1760081282_4124462_1388575898_9bbec473c050c6f6830cae8871fbfe5b.jpg Found 3 images belonging to 1 classes. begin to do the prediction : time used to do the prediction : 0.8143820762634277 (3,) (3, 12) (3, 1280) shape of features : (3, 1280) shape of new features : (1, 3, 1280) save descriptor for thcl : 3847 time to traite the descriptors : 0.049283742904663086 storage_type for insertDescriptorsMulti : 3 To insert : 1388575902 To insert : 1388575900 To insert : 1388575898 time to insert the descriptors : 1.296703577041626 Inside saveOutput : final : False verbose : 0 saveOutput not yet implemented for datou_step.type : tfhub_classification2 we use saveGeneral [1388575902, 1388575900, 1388575898] Looping around the photos to save general results len do output : 3 /1388575902Didn't retrieve data . /1388575900Didn't retrieve data . /1388575898Didn't retrieve data . before output type Here is an output not treated by saveGeneral : Managing all output in save final without adding information in the mtr_datou_result ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575902', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575900', None, None, None, None, None, '3911038') ('4877', None, None, None, None, None, None, None, '3911038') ('4877', '27666970', '1388575898', None, None, None, None, None, '3911038') begin to insert list_values into mtr_datou_result : length of list_values in save_final : 6 time used for this insertion : 0.036992549896240234 save_final save missing photos in datou_result : time spend for datou_step_exec : 38.90656614303589 time spend to save output : 0.03724789619445801 total time spend for step 8 : 38.94381403923035 step9:argmax Fri Oct 10 09:29:27 2025 VR 17-11-17 : now, only for linear exec dependencies tree, some output goes to fill the input of the next VR 22-3-18 : now we test the dependencies tree, but keep two separate code for datou_prepare_output_input until the code is correctly tested, clean and works in both case VR 22-3-18 : but we use the first code for the first step id = -1, build in the code of datou_exec VR 22-3-18 : we should manage here the case when we are at the first step instead of building this step before datou_exec Currently we do not manage missing dependencies information, that could maybe be correctly interpreted with default behavior Some of the step done at execution of the step could be done before when the tree of execution is build and the dependencies of different step analysed complete output_args for input 0 VR 22-3-18 : For now we do not clean correctly the datou structure Beginning of datou_step Argmax ! Inside saveOutput : final : True verbose : 0 photo_id : 1388575902 output[photo_id] : [[(1388575902, '05102018_Papier_non_papier_presque_vide', 0.5940957, 4557, 3513), (1388575902, 'jrm', 0.5077127, 4374, 3379), (1388575902, 'jrm', 0.4112168, 4932, 3847)], 'temp/1760081282_4124462_1388575902_47351dac3c30c89ff02bbbfc8556ea14.jpg'] photo_id : 1388575900 output[photo_id] : [[(1388575900, '05102018_Papier_non_papier_presque_vide', 0.6041125, 4557, 3513), (1388575900, 'jrm', 0.708094, 4374, 3379), (1388575900, 'jrm', 0.8498574, 4932, 3847)], 'temp/1760081282_4124462_1388575900_382f59e38c0f7cb2b91d0a566592768c.jpg'] photo_id : 1388575898 output[photo_id] : [[(1388575898, '05102018_Papier_non_papier_tres_dense', 0.43152675, 4557, 3513), (1388575898, 'jrm', 0.84210294, 4374, 3379), (1388575898, 'jrm', 0.9070878, 4932, 3847)], 'temp/1760081282_4124462_1388575898_9bbec473c050c6f6830cae8871fbfe5b.jpg'] begin to insert list_values into photo_hahstag_ids : length of list_valuse in save_photo_hashtag_id_type : 9 time used for this insertion : 0.039691925048828125 begin to insert list_values into class_photo_scores : length of list_valuse in save_photo_hashtag_id_thcl_score : 9 time used for this insertion : 0.03470754623413086 len list_finale : 9, len picture : 3 begin to insert list_values into mtr_datou_result : length of list_values in save_final : 9 time used for this insertion : 0.03606367111206055 saving photo_ids in datou_result begin to insert list_values into mtr_datou_result : length of list_values in save_final : 3 time used for this insertion : 0.0362544059753418 save missing photos in datou_result : time spend for datou_step_exec : 0.01797032356262207 time spend to save output : 0.20109033584594727 total time spend for step 9 : 0.21906065940856934 caffe_path_current : About to save ! 2 After save, about to update current ! ret : 2 len(input) + len(total_photo_id_missing) : 3 set_done_treatment 54.07user 17.65system 1:29.78elapsed 79%CPU (0avgtext+0avgdata 3018324maxresident)k 8inputs+9160outputs (8major+2608730minor)pagefaults 0swaps