diff mbox series

[FFmpeg-devel,1/5] lavfi/dnn_backend_native_layer_avgpool.h: Documentation

Message ID 20210508121016.12413-1-shubhanshu.e01@gmail.com
State Accepted
Headers show
Series [FFmpeg-devel,1/5] lavfi/dnn_backend_native_layer_avgpool.h: Documentation | expand

Checks

Context Check Description
andriy/x86_make success Make finished
andriy/x86_make_fate success Make fate finished
andriy/PPC64_make success Make finished
andriy/PPC64_make_fate success Make fate finished

Commit Message

Shubhanshu Saxena May 8, 2021, 12:10 p.m. UTC
Add documentation for Average Pool Layer

Signed-off-by: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
---
 .../dnn/dnn_backend_native_layer_avgpool.h    | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)

Comments

Guo, Yejun May 12, 2021, 2:21 a.m. UTC | #1
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Shubhanshu Saxena
> Sent: 2021年5月8日 20:10
> To: ffmpeg-devel@ffmpeg.org
> Cc: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
> Subject: [FFmpeg-devel] [PATCH 1/5]
> lavfi/dnn_backend_native_layer_avgpool.h: Documentation
> 
> Add documentation for Average Pool Layer
> 
> Signed-off-by: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
> ---
>  .../dnn/dnn_backend_native_layer_avgpool.h    | 27
> +++++++++++++++++++
>  1 file changed, 27 insertions(+)
> 
> diff --git a/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> b/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> index 75d9eb187b..0f629b9165 100644
> --- a/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> +++ b/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> @@ -33,7 +33,34 @@ typedef struct AvgPoolParams{
>      DNNPaddingParam padding_method;
>  } AvgPoolParams;
> 
> +/**
> + * @brief Load Average Pooling Layer.
> + *
> + * It assigns the layer parameters to the hyperparameters
> + * like strides, padding method, and kernel size after
> + * parsing from the model file context.
> + *<space>

please run 'git show' for every patch to make sure there's no
tailing spaces in the change.

> + * @param layer pointer to the DNN layer instance
> + * @param model_file_context pointer to model file context
> + * @param file_size model file size
> + * @param operands_num number of operands for the layer

operands_num is the operand count of the whole model,
it is used to check the data read from model file is correct,
just like the usage of file_size.

> + * @return Size of DNN Layer
Size -> size.
return the number of bytes read from model file.

> + * @retval 0 if model file context contains invalid hyperparameters.
return 0 for error.

there's another case to return 0 for out of memory.

> + */
>  int ff_dnn_load_layer_avg_pool(Layer *layer, AVIOContext
> *model_file_context, int file_size, int operands_num);
> +
> +/**
> + * @brief Execute the Average Pooling Layer.
> + * Padding in channel dimensions is currently not supported.
> + *
> + * @param operands input operands

operands contain all the operands of the model

> + * @param input_operand_indexes input operand indexes

input operand indexes for this layer.

> + * @param output_operand_index output operand index

output operand index for this layer.

> + * @param parameters average pooling parameters
> + * @param ctx pointer to Native model context

and its usage is for logging only.

> + * @retval 0 if the execution succeeds
> + * @retval DNN_ERROR if the execution fails
> + */
>  int ff_dnn_execute_layer_avg_pool(DnnOperand *operands, const int32_t
> *input_operand_indexes,
>                                    int32_t output_operand_index,
> const void *parameters, NativeContext *ctx);
> 
> --
> 2.27.0
> 
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
Shubhanshu Saxena May 12, 2021, 9:01 a.m. UTC | #2
On Wed, May 12, 2021 at 7:52 AM Guo, Yejun <yejun.guo@intel.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Shubhanshu Saxena
> > Sent: 2021年5月8日 20:10
> > To: ffmpeg-devel@ffmpeg.org
> > Cc: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
> > Subject: [FFmpeg-devel] [PATCH 1/5]
> > lavfi/dnn_backend_native_layer_avgpool.h: Documentation
> >
> > Add documentation for Average Pool Layer
> >
> > Signed-off-by: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
> > ---
> >  .../dnn/dnn_backend_native_layer_avgpool.h    | 27
> > +++++++++++++++++++
> >  1 file changed, 27 insertions(+)
> >
> > diff --git a/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> > b/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> > index 75d9eb187b..0f629b9165 100644
> > --- a/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> > +++ b/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
> > @@ -33,7 +33,34 @@ typedef struct AvgPoolParams{
> >      DNNPaddingParam padding_method;
> >  } AvgPoolParams;
> >
> > +/**
> > + * @brief Load Average Pooling Layer.
> > + *
> > + * It assigns the layer parameters to the hyperparameters
> > + * like strides, padding method, and kernel size after
> > + * parsing from the model file context.
> > + *<space>
>
> please run 'git show' for every patch to make sure there's no
> tailing spaces in the change.
>
> > + * @param layer pointer to the DNN layer instance
> > + * @param model_file_context pointer to model file context
> > + * @param file_size model file size
> > + * @param operands_num number of operands for the layer
>
> operands_num is the operand count of the whole model,
> it is used to check the data read from model file is correct,
> just like the usage of file_size.
>
> > + * @return Size of DNN Layer
> Size -> size.
> return the number of bytes read from model file.
>
> > + * @retval 0 if model file context contains invalid hyperparameters.
> return 0 for error.
>
> there's another case to return 0 for out of memory.
>
> > + */
> >  int ff_dnn_load_layer_avg_pool(Layer *layer, AVIOContext
> > *model_file_context, int file_size, int operands_num);
> > +
> > +/**
> > + * @brief Execute the Average Pooling Layer.
> > + * Padding in channel dimensions is currently not supported.
> > + *
> > + * @param operands input operands
>
> operands contain all the operands of the model
>
> > + * @param input_operand_indexes input operand indexes
>
> input operand indexes for this layer.
>
> > + * @param output_operand_index output operand index
>
> output operand index for this layer.
>
> > + * @param parameters average pooling parameters
> > + * @param ctx pointer to Native model context
>
> and its usage is for logging only.
>
> > + * @retval 0 if the execution succeeds
> > + * @retval DNN_ERROR if the execution fails
> > + */
> >  int ff_dnn_execute_layer_avg_pool(DnnOperand *operands, const int32_t
> > *input_operand_indexes,
> >                                    int32_t output_operand_index,
> > const void *parameters, NativeContext *ctx);
> >
> > --
> > 2.27.0
> >
> > _______________________________________________
> > ffmpeg-devel mailing list
> > ffmpeg-devel@ffmpeg.org
> > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> >
> > To unsubscribe, visit link above, or email
> > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>

Okay, I'll remove the spaces and correct these lines. Thank you.

Also, since the parameters for loading and execution functions are the same
in other layers, I need to correct them as well. Right?
Guo, Yejun May 12, 2021, 3:01 p.m. UTC | #3
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Shubhanshu Saxena
> Sent: Wednesday, May 12, 2021 5:02 PM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] [PATCH 1/5]
> lavfi/dnn_backend_native_layer_avgpool.h: Documentation
> 
> On Wed, May 12, 2021 at 7:52 AM Guo, Yejun <yejun.guo@intel.com> wrote:
> 
> >
> >
> > > -----Original Message-----
> > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > Shubhanshu Saxena
> > > Sent: 2021年5月8日 20:10
> > > To: ffmpeg-devel@ffmpeg.org
> > > Cc: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
> > > Subject: [FFmpeg-devel] [PATCH 1/5]
> >
> 
> Okay, I'll remove the spaces and correct these lines. Thank you.
> 
> Also, since the parameters for loading and execution functions are the same
> in other layers, I need to correct them as well. Right?

Yes, right.
Shubhanshu Saxena May 12, 2021, 6:48 p.m. UTC | #4
On Wed, May 12, 2021 at 8:31 PM Guo, Yejun <yejun.guo@intel.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Shubhanshu Saxena
> > Sent: Wednesday, May 12, 2021 5:02 PM
> > To: FFmpeg development discussions and patches <ffmpeg-
> > devel@ffmpeg.org>
> > Subject: Re: [FFmpeg-devel] [PATCH 1/5]
> > lavfi/dnn_backend_native_layer_avgpool.h: Documentation
> >
> > On Wed, May 12, 2021 at 7:52 AM Guo, Yejun <yejun.guo@intel.com> wrote:
> >
> > >
> > >
> > > > -----Original Message-----
> > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > > Shubhanshu Saxena
> > > > Sent: 2021年5月8日 20:10
> > > > To: ffmpeg-devel@ffmpeg.org
> > > > Cc: Shubhanshu Saxena <shubhanshu.e01@gmail.com>
> > > > Subject: [FFmpeg-devel] [PATCH 1/5]
> > >
> >
> > Okay, I'll remove the spaces and correct these lines. Thank you.
> >
> > Also, since the parameters for loading and execution functions are the
> same
> > in other layers, I need to correct them as well. Right?
>
> Yes, right.
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

I have sent the newer versions of the patches. Thanks
diff mbox series

Patch

diff --git a/libavfilter/dnn/dnn_backend_native_layer_avgpool.h b/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
index 75d9eb187b..0f629b9165 100644
--- a/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
+++ b/libavfilter/dnn/dnn_backend_native_layer_avgpool.h
@@ -33,7 +33,34 @@  typedef struct AvgPoolParams{
     DNNPaddingParam padding_method;
 } AvgPoolParams;
 
+/**
+ * @brief Load Average Pooling Layer.
+ *
+ * It assigns the layer parameters to the hyperparameters 
+ * like strides, padding method, and kernel size after 
+ * parsing from the model file context.
+ * 
+ * @param layer pointer to the DNN layer instance
+ * @param model_file_context pointer to model file context
+ * @param file_size model file size
+ * @param operands_num number of operands for the layer
+ * @return Size of DNN Layer
+ * @retval 0 if model file context contains invalid hyperparameters.
+ */
 int ff_dnn_load_layer_avg_pool(Layer *layer, AVIOContext *model_file_context, int file_size, int operands_num);
+
+/**
+ * @brief Execute the Average Pooling Layer.
+ * Padding in channel dimensions is currently not supported.
+ * 
+ * @param operands input operands
+ * @param input_operand_indexes input operand indexes
+ * @param output_operand_index output operand index
+ * @param parameters average pooling parameters
+ * @param ctx pointer to Native model context
+ * @retval 0 if the execution succeeds
+ * @retval DNN_ERROR if the execution fails
+ */
 int ff_dnn_execute_layer_avg_pool(DnnOperand *operands, const int32_t *input_operand_indexes,
                                   int32_t output_operand_index, const void *parameters, NativeContext *ctx);