|-jddlZddlmZejeZ ddejdejdejd ejd ejd ed eed e de dejfdZ ddZ dS)N)loggingmean token_logitsduration_logitstargets logit_lengthstarget_lengthsblank_token_id durationssigma reductionreturnc  |dvrtd|d|j} |j\} } } } |}|}t j|d|z }t j|d}t j| | | ftd| }d|d d d d f<|d d d d d d |f}| d kr||d d| d}t j |d d d d d | d z d d fd |d  d}t j td| }td | | zd z D]}td || z d z}t|d z| }t j||| }||z }g}t!|D]k\}}||z }|d k}|s&|d }|d krh|d d ||f|d d ||fz|d d |||fz} t j|d | |} || |d k}!||!z}"|"r|d z d }#| d kr|#| dz n|#}$|d d ||#f|d d ||$fz|d d ||#|fz} t j|"d | |} || m|r4t j|d }%t j|%d |d d ||f<t j| | }&t j| ftd| }'t!|D]\}}|d kr ||z }(|(d k})|)s,|(d }*||&|*|f||&|*||fz||&|*||fz}+t j|'|+gd },t j|)t j|,d |'}'|' }-|dkr)|-|z S|dkr|-S|-S)a Compute TDT (Token-and-Duration Transducer) loss (https://arxiv.org/abs/2304.06795). Ported from NeMo's `TDTLossPytorch` with anti-diagonal processing. Unlike standard RNNT loss, this loss trains both the token prediction head and the duration prediction head. It uses vectorized anti-diagonal processing for efficiency: all (t, u) pairs on each anti-diagonal t+u=n are computed in parallel as batched tensor operations. Args: token_logits: Token logits of shape `(batch, T, U+1, vocab_size+1)`. duration_logits: Duration logits of shape `(batch, T, U+1, num_durations)`. targets: Target labels of shape `(batch, U)`. logit_lengths: Encoder output lengths of shape `(batch,)`. target_lengths: Target lengths of shape `(batch,)`. blank_token_id: Blank token id. durations: List of duration values (e.g., `[0, 1, 2, 3, 4]`). sigma: Logit undernormalization constant (see TDT paper). Defaults to `0.0`. reduction: Loss reduction method. One of `"mean"`, `"sum"`, or `"none"`. Defaults to `"mean"`. Returns: Scalar loss tensor (or per-example losses if `reduction="none"`). )rsumnonezInvalid reduction mode "z,". Expected one of "mean", "sum", or "none".)dimz-inf)devicerNr)rindex)minr)maxrr) ValueErrorrshapefloattorch log_softmaxfull unsqueezeexpandgathersqueezetensorrangerrarange enumerateanyclampwhereappendstack logsumexprr).rrr r r r r rrr batch_sizemax_tmax_u_token_log_probsduration_log_probs log_alphablank_log_probstargets_expandedlabel_log_probsneg_infnu_startu_end u_indices t_indicesall_candidatesidurt_prevvalid_tt_srccontribvalid_u valid_bothu_src u_src_labelstacked batch_idx log_probst_finalvalid t_clampedterminalcombinedlossess. Z/var/www/html/banglarbhumi/venv/lib/python3.11/site-packages/transformers/loss/loss_tdt.pytdt_lossrUsD///kIkkklll  F"."4Jua%%''L%++--O' "===EO*?CCC Ju5uV}}VTTTIIaaaAg&aaaAAA~&=>O qyy",,Q//66r5"EE, AAAqqq+EAI+qqq0 1",,R00    '"++  l5==888G1eema' ( ()Q)QaUQ''AE5!!L%???  M  ** / /FAs_FkG;;== LLQL''EQwwaaa 12%aaa &9:;(E9a)?@A  +g&7&7&:&:GWMM%%g... !mG 7*J~~ /"Q--!-44rfs   H % %FF,F\F\F< F L F  FCyF FF \FFFFb rV