2023-10-12 15:23:18 +00:00
# include "clip.h"
# include "common.h"
# include "llama.h"
2023-11-06 21:36:23 +00:00
# include "llava.h"
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
# include "base64.hpp"
2023-10-12 15:23:18 +00:00
# include <cstdio>
# include <cstdlib>
# include <vector>
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
# include <numeric>
// RGB uint8 image
struct clip_image_u8 {
int nx ;
int ny ;
std : : vector < uint8_t > buf ;
} ;
// RGB float32 image (NHWC)
// Memory layout: RGBRGBRGB...
struct clip_image_f32 {
int nx ;
int ny ;
std : : vector < float > buf ;
} ;
struct clip_image_grid_shape {
int first ;
int second ;
} ;
/**
* Selects the best resolution from a list of possible resolutions based on the original size .
*
* @ param original_size The original size of the image in the format ( width , height ) .
* @ param possible_resolutions A list of possible resolutions in the format [ ( width1 , height1 ) , ( width2 , height2 ) , . . . ] .
* @ return The best fit resolution in the format ( width , height ) .
*/
static std : : pair < int , int > select_best_resolution ( const std : : pair < int , int > & original_size , const std : : vector < std : : pair < int , int > > & possible_resolutions ) {
int original_width = original_size . first ;
int original_height = original_size . second ;
std : : pair < int , int > best_fit ;
int max_effective_resolution = 0 ;
int min_wasted_resolution = std : : numeric_limits < int > : : max ( ) ;
for ( const auto & resolution : possible_resolutions ) {
int width = resolution . first ;
int height = resolution . second ;
float scale = std : : min ( static_cast < float > ( width ) / original_width , static_cast < float > ( height ) / original_height ) ;
int downscaled_width = static_cast < int > ( original_width * scale ) ;
int downscaled_height = static_cast < int > ( original_height * scale ) ;
int effective_resolution = std : : min ( downscaled_width * downscaled_height , original_width * original_height ) ;
int wasted_resolution = ( width * height ) - effective_resolution ;
2024-04-21 12:19:04 +00:00
// LOG_TEE("resolution: %d %d, scale: %f, downscaled: %d %d, effective: %d, wasted: %d\n", width, height, scale, downscaled_width, downscaled_height, effective_resolution, wasted_resolution);
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
if ( effective_resolution > max_effective_resolution | | ( effective_resolution = = max_effective_resolution & & wasted_resolution < min_wasted_resolution ) ) {
max_effective_resolution = effective_resolution ;
min_wasted_resolution = wasted_resolution ;
best_fit = resolution ;
}
}
return best_fit ;
}
/**
* @ brief Get the anyres image grid shape object
*
* @ param image_size
* @ param grid_pinpoints
* @ param image_patch_size
* @ return < int , int >
*/
static struct clip_image_grid_shape get_anyres_image_grid_shape ( const std : : pair < int , int > & image_size , const std : : vector < std : : pair < int , int > > & grid_pinpoints , int image_patch_size ) {
/**
Conversion from gguf flat array to vector :
std : : vector < std : : pair < int , int > > possible_resolutions ;
for ( int i = 0 ; i < 32 & & params . image_grid_pinpoints [ i ] ! = 0 ; i + = 2 ) {
possible_resolutions . push_back ( { params . image_grid_pinpoints [ i ] , params . image_grid_pinpoints [ i + 1 ] } ) ;
}
*/
auto best_resolution = select_best_resolution ( image_size , grid_pinpoints ) ;
return { best_resolution . first / image_patch_size , best_resolution . second / image_patch_size } ;
}
// Take the image segments in a grid configuration and return the embeddings and the number of embeddings into preallocated memory (image_embd_out)
static bool clip_llava_handle_patches ( clip_ctx * ctx_clip , std : : vector < float * > & image_embd_v , struct clip_image_grid_shape grid_shape , float * image_embd_out , int * n_img_pos_out ) {
struct {
struct ggml_context * ctx ;
} model ;
const int32_t image_size = clip_image_size ( ctx_clip ) ;
const int32_t patch_size = clip_patch_size ( ctx_clip ) ;
int32_t num_patches_per_side = image_size / patch_size ; // 336 / 14 = 24 - used for embedding-patching boxes (24*24 = 576 patches)
int num_patches_width = grid_shape . first ; // grid 1-4
int num_patches_height = grid_shape . second ; // grid 1-4
2024-02-15 07:59:18 +00:00
const size_t num_images = num_patches_width * num_patches_height + 1 ;
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
// TODO: size calculation is not calculated - it's only tens of MB
size_t ctx_size = 0 ;
{
ctx_size + = clip_embd_nbytes ( ctx_clip ) * num_images * 8 ; // image_features
ctx_size + = 1024 * 1024 * ggml_type_size ( GGML_TYPE_F32 ) ;
}
struct ggml_init_params params {
/*.mem_size =*/ ctx_size ,
/*.mem_buffer =*/ NULL ,
/*.no_alloc =*/ false , // NOTE: this should be false when using the legacy API
} ;
// Python reference code for full unpad:
/*
base_image_feature = image_feature [ 0 ]
image_feature = image_feature [ 1 : ]
image_feature = image_feature . permute ( 4 , 0 , 2 , 1 , 3 ) . contiguous ( )
image_feature = image_feature . flatten ( 1 , 2 ) . flatten ( 2 , 3 )
image_feature = unpad_image ( image_feature , image_sizes [ image_idx ] )
image_feature = torch . cat ( (
image_feature ,
self . model . image_newline [ : , None , None ] . expand ( * image_feature . shape [ : - 1 ] , 1 )
) , dim = - 1 )
image_feature = image_feature . flatten ( 1 , 2 ) . transpose ( 0 , 1 )
image_feature = torch . cat ( ( base_image_feature , image_feature ) , dim = 0 )
*/
// We now have two options: unpad or no unpad. Unpad removes tokens for faster llm eval.
// In terms of result quality it appears to make no difference, so we'll start with the easier approach given 5D tensors are not supported in ggml yet.
// Without unpad we have to split the sub-image embeddings into patches of 24 features each and permute them.
// Once all images are processed to prepended the base_image_features without any changes.
// Pytorch reference simplified, modified for ggml compatibility - confirmed identical output in python (for a 2x2 grid image (676x676 scaling))
/*
image_feature = image_feature . view ( 2 , 2 , 24 , 24 , 4096 )
image_feature = image_feature . permute ( 0 , 2 , 1 , 3 , 4 ) . contiguous ( )
image_feature = image_feature . view ( 2 , 24 , 2 , 24 , 4096 )
image_feature = image_feature . flatten ( 0 , 3 )
// Reshape to 4D tensor by merging the last two dimensions
image_feature = image_feature . view ( 2 , 2 , 24 , 24 * 4096 )
image_feature = image_feature . permute ( 0 , 2 , 1 , 3 ) . contiguous ( )
image_feature = image_feature . view ( - 1 , 4096 )
*/
model . ctx = ggml_init ( params ) ;
struct ggml_tensor * image_features = ggml_new_tensor_3d ( model . ctx , GGML_TYPE_F32 , clip_n_mmproj_embd ( ctx_clip ) , clip_n_patches ( ctx_clip ) , num_images - 1 ) ; // example: 4096 x 576 x 4
// ggml_tensor_printf(image_features,"image_features",__LINE__,false,false);
// fill it with the image embeddings, ignoring the base
for ( size_t i = 1 ; i < num_images ; i + + ) {
size_t offset = ( i - 1 ) * clip_embd_nbytes ( ctx_clip ) ;
memcpy ( ( uint8_t * ) ( image_features - > data ) + offset , image_embd_v [ i ] , clip_embd_nbytes ( ctx_clip ) ) ;
}
struct ggml_cgraph * gf = ggml_new_graph ( model . ctx ) ;
size_t size_ele = ggml_type_size ( GGML_TYPE_F32 ) ;
struct ggml_tensor * image_features_patchview = ggml_view_4d ( model . ctx , image_features ,
num_patches_per_side * clip_n_mmproj_embd ( ctx_clip ) ,
num_patches_per_side ,
num_patches_width ,
num_patches_height ,
size_ele * num_patches_per_side * clip_n_mmproj_embd ( ctx_clip ) ,
size_ele * num_patches_per_side * clip_n_mmproj_embd ( ctx_clip ) * num_patches_per_side ,
size_ele * num_patches_per_side * clip_n_mmproj_embd ( ctx_clip ) * num_patches_per_side * num_patches_width , 0 ) ;
// ggml_tensor_printf(image_features_patchview,"image_features_patchview",__LINE__,false,false);
struct ggml_tensor * permuted_cont = ggml_cont ( model . ctx , ggml_permute ( model . ctx , image_features_patchview , 0 , 2 , 1 , 3 ) ) ;
/**
At the end of each row we have to add the row_end embeddings , which are the same as the newline embeddings
image_feature = torch . cat ( (
image_feature ,
self . model . image_newline [ : , None , None ] . expand ( * image_feature . shape [ : - 1 ] , 1 ) . to ( image_feature . device )
) , dim = - 1 )
*
*/
// ggml_tensor_printf(permuted_cont,"permuted_cont",__LINE__,false,false);
struct ggml_tensor * flatten = ggml_view_2d ( model . ctx , permuted_cont , clip_n_mmproj_embd ( ctx_clip ) , num_patches_height * num_patches_width * num_patches_per_side * num_patches_per_side , size_ele * clip_n_mmproj_embd ( ctx_clip ) , 0 ) ;
// ggml_tensor_printf(flatten,"flatten",__LINE__,false,false);
ggml_build_forward_expand ( gf , flatten ) ;
ggml_graph_compute_with_ctx ( model . ctx , gf , 1 ) ;
struct ggml_tensor * result = gf - > nodes [ gf - > n_nodes - 1 ] ;
memcpy ( image_embd_out , image_embd_v [ 0 ] , clip_embd_nbytes ( ctx_clip ) ) ; // main image as global context
// append without newline tokens (default behavior in llava_arch when not using unpad ):
memcpy ( image_embd_out + clip_n_patches ( ctx_clip ) * clip_n_mmproj_embd ( ctx_clip ) , ( float * ) result - > data , clip_embd_nbytes ( ctx_clip ) * ( num_images - 1 ) ) ; // grid patches
* n_img_pos_out = static_cast < int > ( result - > ne [ 1 ] + clip_n_patches ( ctx_clip ) ) ;
// Debug: Test single segments
// Current findings: sending base image, sending a segment embedding all works similar to python
// However, permuted embeddings do not work yet (stride issue?)
// memcpy(image_embd_out, image_embd_v[0], clip_embd_nbytes(ctx_clip)); // main image as context
// memcpy(image_embd_out, (float*)prepared_cont->data, clip_embd_nbytes(ctx_clip)); // main image as context
// *n_img_pos_out=576;
ggml_free ( model . ctx ) ;
return true ;
}
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
static bool encode_image_with_clip ( clip_ctx * ctx_clip , int n_threads , const clip_image_u8 * img , float * image_embd , int * n_img_pos ) {
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
// std::vector<clip_image_f32*> img_res_v; // format VectN x H x W x RGB (N x 336 x 336 x 3), so interleaved RGB - different to the python implementation which is N x 3 x 336 x 336
clip_image_f32_batch img_res_v ;
img_res_v . size = 0 ;
img_res_v . data = nullptr ;
2024-03-15 14:31:05 +00:00
if ( ! clip_image_preprocess ( ctx_clip , img , & img_res_v ) ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: unable to preprocess image \n " , __func__ ) ;
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
delete [ ] img_res_v . data ;
2023-11-06 21:36:23 +00:00
return false ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
const int64_t t_img_enc_start_us = ggml_time_us ( ) ;
2023-10-12 15:23:18 +00:00
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
const char * mm_patch_merge_type = clip_patch_merge_type ( ctx_clip ) ;
if ( strcmp ( mm_patch_merge_type , " spatial_unpad " ) ! = 0 ) {
// flat / default llava-1.5 type embedding
* n_img_pos = clip_n_patches ( ctx_clip ) ;
bool encoded = clip_image_encode ( ctx_clip , n_threads , & img_res_v . data [ 0 ] , image_embd ) ; // image_embd shape is 576 x 4096
delete [ ] img_res_v . data ;
if ( ! encoded ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " Unable to encode image \n " ) ;
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
return false ;
}
} else {
// spatial_unpad llava-1.6 type embedding
// TODO: CLIP needs batching support - in HF the llm projection is separate after encoding, which might be a solution to quickly get batching working
std : : vector < float * > image_embd_v ;
image_embd_v . resize ( img_res_v . size ) ;
for ( size_t i = 0 ; i < img_res_v . size ; i + + ) {
image_embd_v [ i ] = ( float * ) malloc ( clip_embd_nbytes ( ctx_clip ) ) ; // 576 patches * 4096 embeddings * 4 bytes = 9437184
const bool encoded = clip_image_encode ( ctx_clip , n_threads , & img_res_v . data [ i ] , image_embd_v [ i ] ) ; // image data is in 3x336x336 format and will be converted to 336x336x3 inside
if ( ! encoded ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " Unable to encode image - spatial_unpad - subimage %d of %d \n " , ( int ) i + 1 , ( int ) img_res_v . size ) ;
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
return false ;
}
}
const int64_t t_img_enc_batch_us = ggml_time_us ( ) ;
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: %d segments encoded in %8.2f ms \n " , __func__ , ( int ) img_res_v . size , ( t_img_enc_batch_us - t_img_enc_start_us ) / 1000.0 ) ;
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
const int32_t * image_grid = clip_image_grid ( ctx_clip ) ;
std : : vector < std : : pair < int , int > > grid_pinpoints ;
for ( int i = 0 ; i < 32 & & image_grid [ i ] ! = 0 ; i + = 2 ) {
grid_pinpoints . push_back ( { image_grid [ i ] , image_grid [ i + 1 ] } ) ;
}
// free all img_res_v - not needed anymore
delete [ ] img_res_v . data ;
img_res_v . size = 0 ;
img_res_v . data = nullptr ;
const int32_t image_size = clip_image_size ( ctx_clip ) ;
struct clip_image_grid_shape grid_shape = get_anyres_image_grid_shape ( { img - > nx , img - > ny } , grid_pinpoints , image_size ) ;
int n_img_pos_out ;
clip_llava_handle_patches ( ctx_clip , image_embd_v , grid_shape , image_embd , & n_img_pos_out ) ;
* n_img_pos = n_img_pos_out ;
for ( size_t i = 0 ; i < image_embd_v . size ( ) ; i + + ) {
free ( image_embd_v [ i ] ) ;
}
image_embd_v . clear ( ) ;
// debug image/segment/normalization content:
// clip_image_u8 * tmp = clip_image_u8_init();
// clip_image_convert_f32_to_u8(*image_feature, *tmp);
// clip_image_save_to_bmp(*tmp, "image_feature.bmp");
2023-10-12 15:23:18 +00:00
}
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: image embedding created: %d tokens \n " , __func__ , * n_img_pos ) ;
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
2023-11-06 21:36:23 +00:00
const int64_t t_img_enc_end_us = ggml_time_us ( ) ;
float t_img_enc_ms = ( t_img_enc_end_us - t_img_enc_start_us ) / 1000.0 ;
2023-10-12 15:23:18 +00:00
2024-04-21 12:19:04 +00:00
LOG_TEE ( " \n %s: image encoded in %8.2f ms by CLIP (%8.2f ms per image patch) \n " , __func__ , t_img_enc_ms , t_img_enc_ms / * n_img_pos ) ;
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
return true ;
}
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
bool llava_validate_embed_size ( const llama_context * ctx_llama , const clip_ctx * ctx_clip ) {
// make sure that the correct mmproj was used, i.e., compare apples to apples
int n_llama_embd = llama_n_embd ( llama_get_model ( ctx_llama ) ) ;
auto n_image_embd = clip_n_mmproj_embd ( ctx_clip ) ;
if ( n_image_embd ! = n_llama_embd ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: embedding dim of the multimodal projector (%d) is not equal to that of LLaMA (%d). Make sure that you use the correct mmproj file. \n " , __func__ , n_image_embd , n_llama_embd ) ;
2023-11-06 21:36:23 +00:00
return false ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
return true ;
}
2023-10-12 15:23:18 +00:00
2024-02-20 19:07:22 +00:00
bool llava_image_embed_make_with_clip_img ( clip_ctx * ctx_clip , int n_threads , const clip_image_u8 * img , float * * image_embd_out , int * n_img_pos_out ) {
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
float * image_embd = ( float * ) malloc ( clip_embd_nbytes ( ctx_clip ) * 6 ) ; // TODO: base on gridsize/llava model
2023-10-12 15:23:18 +00:00
if ( ! image_embd ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " Unable to allocate memory for image embeddings \n " ) ;
2023-11-06 21:36:23 +00:00
return false ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
int n_img_pos ;
if ( ! encode_image_with_clip ( ctx_clip , n_threads , img , image_embd , & n_img_pos ) ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: cannot encode image, aborting \n " , __func__ ) ;
2023-11-06 21:36:23 +00:00
free ( image_embd ) ;
return false ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
* image_embd_out = image_embd ;
* n_img_pos_out = n_img_pos ;
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
return true ;
}
2023-10-14 10:52:44 +00:00
2023-11-06 21:36:23 +00:00
bool llava_eval_image_embed ( llama_context * ctx_llama , const struct llava_image_embed * image_embed , int n_batch , int * n_past ) {
int n_embd = llama_n_embd ( llama_get_model ( ctx_llama ) ) ;
for ( int i = 0 ; i < image_embed - > n_image_pos ; i + = n_batch ) {
int n_eval = image_embed - > n_image_pos - i ;
if ( n_eval > n_batch ) {
n_eval = n_batch ;
}
llama_batch batch = { int32_t ( n_eval ) , nullptr , ( image_embed - > embed + i * n_embd ) , nullptr , nullptr , nullptr , nullptr , * n_past , 1 , 0 , } ;
if ( llama_decode ( ctx_llama , batch ) ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s : failed to eval \n " , __func__ ) ;
2023-11-06 21:36:23 +00:00
return false ;
}
* n_past + = n_eval ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
return true ;
}
2023-10-12 15:23:18 +00:00
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
struct llava_image_embed * llava_image_embed_make_with_bytes ( struct clip_ctx * ctx_clip , int n_threads , const unsigned char * image_bytes , int image_bytes_length ) {
2023-12-30 21:24:42 +00:00
clip_image_u8 * img = clip_image_u8_init ( ) ;
2023-11-06 21:36:23 +00:00
if ( ! clip_image_load_from_bytes ( image_bytes , image_bytes_length , img ) ) {
clip_image_u8_free ( img ) ;
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: can't load image from bytes, is it a valid image? " , __func__ ) ;
2023-11-06 21:36:23 +00:00
return NULL ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
float * image_embed = NULL ;
int n_image_pos = 0 ;
bool image_embed_result = llava_image_embed_make_with_clip_img ( ctx_clip , n_threads , img , & image_embed , & n_image_pos ) ;
if ( ! image_embed_result ) {
clip_image_u8_free ( img ) ;
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: coulnd't embed the image \n " , __func__ ) ;
2023-11-06 21:36:23 +00:00
return NULL ;
2023-10-12 15:23:18 +00:00
}
2023-11-06 21:36:23 +00:00
clip_image_u8_free ( img ) ;
auto result = ( llava_image_embed * ) malloc ( sizeof ( llava_image_embed ) ) ;
result - > embed = image_embed ;
result - > n_image_pos = n_image_pos ;
return result ;
}
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
static bool load_file_to_bytes ( const char * path , unsigned char * * bytesOut , long * sizeOut ) {
auto file = fopen ( path , " rb " ) ;
if ( file = = NULL ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: can't read file %s \n " , __func__ , path ) ;
2023-11-06 21:36:23 +00:00
return false ;
}
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
fseek ( file , 0 , SEEK_END ) ;
auto fileSize = ftell ( file ) ;
fseek ( file , 0 , SEEK_SET ) ;
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
auto buffer = ( unsigned char * ) malloc ( fileSize ) ; // Allocate memory to hold the file data
if ( buffer = = NULL ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: failed to alloc %ld bytes for file %s \n " , __func__ , fileSize , path ) ;
2023-11-06 21:36:23 +00:00
perror ( " Memory allocation error " ) ;
fclose ( file ) ;
return false ;
2023-10-12 15:23:18 +00:00
}
2023-11-17 15:22:56 +00:00
errno = 0 ;
size_t ret = fread ( buffer , 1 , fileSize , file ) ; // Read the file into the buffer
if ( ferror ( file ) ) {
die_fmt ( " read error: %s " , strerror ( errno ) ) ;
}
if ( ret ! = ( size_t ) fileSize ) {
die ( " unexpectedly reached end of file " ) ;
}
2023-11-06 21:36:23 +00:00
fclose ( file ) ; // Close the file
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
* bytesOut = buffer ;
* sizeOut = fileSize ;
return true ;
}
2023-10-12 15:23:18 +00:00
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
struct llava_image_embed * llava_image_embed_make_with_filename ( struct clip_ctx * ctx_clip , int n_threads , const char * image_path ) {
2023-11-06 21:36:23 +00:00
unsigned char * image_bytes ;
long image_bytes_length ;
auto loaded = load_file_to_bytes ( image_path , & image_bytes , & image_bytes_length ) ;
if ( ! loaded ) {
2024-04-21 12:19:04 +00:00
LOG_TEE ( " %s: failed to load %s \n " , __func__ , image_path ) ;
2023-11-06 21:36:23 +00:00
return NULL ;
2023-10-12 15:23:18 +00:00
}
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
llava_image_embed * embed = llava_image_embed_make_with_bytes ( ctx_clip , n_threads , image_bytes , image_bytes_length ) ;
2023-11-06 21:36:23 +00:00
free ( image_bytes ) ;
2023-10-12 15:23:18 +00:00
2023-11-06 21:36:23 +00:00
return embed ;
}
2023-10-12 15:23:18 +00:00
llava : support v1.6 (#5267)
* Create llava-survery-v2.py
* Update convert-image-encoder-to-gguf.py
* Update convert-image-encoder-to-gguf.py
* Rename llava-survery-v2.py to llava-surgery-v2.py
* Update convert-image-encoder-to-gguf.py
will now search for projector
* Update convert-image-encoder-to-gguf.py
whoops
* Update llava-surgery-v2.py
* Clip: Bugfix for normalization (it did not loat the 3 std and mean values)
Clip: bicubic resize function
Clip: added save-to-bmp/pil for debugging and conversion from/to 32/8 images
Clip: added normalization with FP16 precision simulation (image tensors match HF implementation, can be switched off, only used for llava-1.6)
Clip: added newline tensor, mergetype kv, image-grid kv, new resize-pad function with resolution from gridpoints
Clip: clip_image_preprocess now returns a float * vector instead of float, this way llava 1.5 and 1.6 is supported
llava: added ggml cpu graph for embedding patching, added spatial_unpad preliminary support, added a lot of comments that need to be cleaned when all is final
convert-image-encoder: fixed image-grid flattening
* whitespace corrections
* ws
* Tensors are now properly permuted.
Before the embeddings were inserted 1:1, now they are split into the 24x24 patches as in reference.
* ws
* added verbose_prompt support into cli
added stopwords for llava-1.6 into cli
* moved llava functions to llava.cpp, made clip.h C compatible API, replaced vector style functions with pointers, added a debug define to remove functions from compilation while not needed
* ws
* convert : skip unknown tensors (need for LLaVA)
* llava : update readme
* llava : fix compile warnings
* llava : style
* convert : add --skip-unknown CLI arg
* server : remove clip structs
* bugfix for non llava-1.6
It should now work with llava-1.5 as well
* clip : minor code rearrange
* llava : update readme a bit
---------
Co-authored-by: John <cmt-nct@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-02-14 07:38:35 +00:00
void llava_image_embed_free ( struct llava_image_embed * embed ) {
2023-11-06 21:36:23 +00:00
free ( embed - > embed ) ;
free ( embed ) ;
2023-10-12 15:23:18 +00:00
}