FiddlmZmZmZmZmZddlmZ erJeddusJdaedZ ddl m Z ddl Z ddl mZdd ed ed dfd Zedd ZddZddZddZddZereeeedZyiZy#eeeef$rdZYCwxYw))LOGGERSETTINGS TESTS_RUNNINGcolorstr torch_utils)smart_inference_mode tensorboardTNz TensorBoard: )deepcopy) SummaryWriterscalarsstepreturncptr0|jD]\}}tj|||yy)aLog scalar values to TensorBoard. Args: scalars (dict): Dictionary of scalar values to log to TensorBoard. Keys are scalar names and values are the corresponding scalar values. step (int): Global step value to record with the scalar values. Used for x-axis in TensorBoard graphs. Examples: Log training metrics >>> metrics = {"loss": 0.5, "accuracy": 0.95} >>> _log_scalars(metrics, step=100) N)WRITERitems add_scalar)r r kvs i/var/www/pod-logistic/pod-ai/venv/lib/python3.12/site-packages/ultralytics/utils/callbacks/tensorboard.py _log_scalarsrs7MMO *DAq   aD ) *c j|jj}t|tr||fn|}t |j j }tjddg||j|j} |j jtjtjjt!j"|j |dgt%j&t(dy#t*$r} t-t!j"|j }|j|j/d}|j1D]}t3|d sd |_d |_||tjtjj||dgt%j&t(dn<#t*$r+}t%j8t(d |d |Yd}~nd}~wwxYwYd}~yYd}~yd}~wwxYw)aLog model graph to TensorBoard. This function attempts to visualize the model architecture in TensorBoard by tracing the model with a dummy input tensor. It first tries a simple method suitable for YOLO models, and if that fails, falls back to a more complex approach for models like RTDETR that may require special handling. Args: trainer (ultralytics.engine.trainer.BaseTrainer): The trainer object containing the model to visualize. Must have attributes model and args with imgsz. Notes: This function requires TensorBoard integration to be enabled and the global WRITER to be initialized. It handles potential warnings from the PyTorch JIT tracer and attempts to gracefully handle different model architectures. )devicedtypeF)strictu#model graph visualization added ✅N)verboseexportT torchscriptz)TensorBoard graph visualization failure: z -> )argsimgsz isinstanceintnextmodel parameterstorchzerosrrevalr add_graphjittracer unwrap_modelrinfoPREFIX Exceptionr fusemoduleshasattrrformatwarning)trainerr"pime1r&me2s r_log_tensorboard_graphr=*s$ LL  E(4UEN%E W]] % % '(A a^U^AHHAGG DB] )A)A'--)PRT]bcegh vhABC ] ][55gmmDEE JJLJJuJ-E]]_ -1h'#AH,AH - "I   UYY__UBu_Er J KK6("EF G ] NNfX%NrdRVWYVZ[ \ \ ] \ H]sEB D H2A)G,A*G,+H-, H 5!HH-H  H--H2c trI tt|jat j t d|jdyy#t$r(}t jt d|Yd}~yd}~wwxYw)z2Initialize TensorBoard logging with SummaryWriter.z!Start with 'tensorboard --logdir z!', view at http://localhost:6006/z=TensorBoard not initialized correctly, not logging this run. N) r strsave_dirrrr/r0r1r6)r7es ron_pretrain_routine_startrBXsx h"3w'7'7#89F KK6("CGDTDTCUUvw x   h NNfX%bcdbef g g hsAA BA==Bc(tr t|yy)zLog TensorBoard graph.N)rr=r7s ron_train_startrEcs w'rct|j|jd|jdzt|j|jdzy)z5Log scalar statistics at the end of a training epoch.train)prefixrN)rlabel_loss_itemstlossepochlrrDs ron_train_epoch_endrMisA))'--)H'--Z[J[\W]]Q./rcJt|j|jdzy)z+Log epoch metrics at end of training epoch.rN)rmetricsrKrDs ron_fit_epoch_endrPos'--!"34r)rBrErPrM)r)rN)ultralytics.utilsrrrrrultralytics.utils.torch_utilsrrr0copyr r(torch.utils.tensorboardr ImportErrorAssertionError TypeErrorAttributeErrordictr$rr=rBrErMrP callbacksrrr\sUT>  M "d ** * F o &F5*$*c*$*$*]*]Zh( 0 5 &?(,0    E ^Y?Ms'A33 BB