diff --git a/examples/llava/clip.cpp b/examples/llava/clip.cpp index ba28c07c6..463b7c865 100644 --- a/examples/llava/clip.cpp +++ b/examples/llava/clip.cpp @@ -896,7 +896,7 @@ static ggml_cgraph * clip_image_build_graph(clip_ctx * ctx, const clip_image_f32 mlp_3 = ggml_cont(ctx0, ggml_permute(ctx0, mlp_3, 1, 0, 2, 3)); mlp_3 = ggml_reshape_4d(ctx0, mlp_3, n_patch, n_patch, mlp_3->ne[1], mlp_3->ne[2]); // stride = 1, padding = 1, bias is nullptr - block_1 = ggml_conv_depthwise_2d(ctx0, model.mm_model_block_1_block_0_0_w, mlp_3, 1, 1, 1, 1, 1, 1); + block_1 = ggml_conv_2d_dw(ctx0, model.mm_model_block_1_block_0_0_w, mlp_3, 1, 1, 1, 1, 1, 1); // layer norm // // block_1 shape = [1, 2048, 24, 24], ne = [24, 24, 2048, 1] @@ -944,7 +944,7 @@ static ggml_cgraph * clip_image_build_graph(clip_ctx * ctx, const clip_image_f32 // block_2 { // stride = 2 - block_1 = ggml_conv_depthwise_2d(ctx0, model.mm_model_block_2_block_0_0_w, block_1, 2, 2, 1, 1, 1, 1); + block_1 = ggml_conv_2d_dw(ctx0, model.mm_model_block_2_block_0_0_w, block_1, 2, 2, 1, 1, 1, 1); // block_1 shape = [1, 2048, 12, 12], ne = [12, 12, 2048, 1] // layer norm @@ -1005,7 +1005,7 @@ static ggml_cgraph * clip_image_build_graph(clip_ctx * ctx, const clip_image_f32 // mlp_2 ne [24, 24, 2048, 1] mlp_2 = ggml_pool_2d(ctx0, mlp_2, GGML_OP_POOL_AVG, 2, 2, 2, 2, 0, 0); // weight ne = [3, 3, 2048, 1] - struct ggml_tensor * peg_0 = ggml_conv_depthwise_2d(ctx0, model.mm_model_peg_0_w, mlp_2, 1, 1, 1, 1, 1, 1); + struct ggml_tensor * peg_0 = ggml_conv_2d_dw(ctx0, model.mm_model_peg_0_w, mlp_2, 1, 1, 1, 1, 1, 1); peg_0 = ggml_cont(ctx0, ggml_permute(ctx0, peg_0, 1, 2, 0, 3)); peg_0 = ggml_add(ctx0, peg_0, model.mm_model_peg_0_b); mlp_2 = ggml_cont(ctx0, ggml_permute(ctx0, mlp_2, 1, 2, 0, 3)); diff --git a/examples/tts/tts.cpp b/examples/tts/tts.cpp index 8ba13944f..e2306a5a8 100644 --- a/examples/tts/tts.cpp +++ b/examples/tts/tts.cpp @@ -651,6 +651,8 @@ lovely<|t_0.56|><|code_start|><|634|><|596|><|1766|><|1556|><|1306|><|1285|><|14 LOG_INF("%s: time for decoder: %.3f ms\n", __func__, (ggml_time_us() - t_dec_start) / 1000.0f); } + common_perf_print(ctx_ttc, smpl[0]); + //std::vector codes = {198, 88225, 155856, 151669, 152205, // 153064, 152537, 153421, 153209, 152524, 151689, 152993, 152438, 152695, // 153091, 152945, 152829, 152534, 152934, 153020, 151997, 152263, 153010,