Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-18 16:55:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'upstream' into concedo_experimental

Browse Source 
# Conflicts:
#	.devops/full-cuda.Dockerfile
#	.devops/full-rocm.Dockerfile
#	.devops/llama-cli-cuda.Dockerfile
#	.devops/llama-cli-rocm.Dockerfile
#	.devops/llama-cli-vulkan.Dockerfile
#	.devops/llama-cpp-cuda.srpm.spec
#	.devops/llama-server-cuda.Dockerfile
#	.devops/llama-server-rocm.Dockerfile
#	.devops/llama-server-vulkan.Dockerfile
#	.github/workflows/build.yml
#	.github/workflows/docker.yml
#	CMakeLists.txt
#	Makefile
#	README.md
#	examples/llama.android/llama/src/main/cpp/CMakeLists.txt
#	flake.lock
#	ggml/CMakeLists.txt
#	ggml/src/CMakeLists.txt
#	grammars/README.md
#	scripts/sync-ggml-am.sh
#	scripts/sync-ggml.last
#	tests/test-chat-template.cpp
#	tests/test-grammar-integration.cpp
#	tests/test-json-schema-to-grammar.cpp
This commit is contained in:
Concedo
2024-06-30 10:59:42 +08:00
parent 8421243c6d 72272b83a3
commit 02f92f6ecc
22 changed files with 632 additions and 182 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/infill/README.md
+1
View File
@@ -15,6 +15,7 @@ In this section, we cover the most commonly used options for running the `infill
- `-i, --interactive`: Run the program in interactive mode, allowing you to provide input directly and receive real-time responses.
- `-n N, --n-predict N`: Set the number of tokens to predict when generating text. Adjusting this value can influence the length of the generated text.
- `-c N, --ctx-size N`: Set the size of the prompt context. The default is 512, but LLaMA models were built with a context of 2048, which will provide better results for longer input/inference.
- `--spm-infill`: Use Suffix/Prefix/Middle pattern for infill (instead of Prefix/Suffix/Middle) as some models prefer this.
## Input Prompts
examples/infill/infill.cpp
+13 -11
View File
@@ -211,6 +211,7 @@ int main(int argc, char ** argv) {
suff_rm_leading_spc = false;
}
std::vector<llama_token> embd_inp;
std::vector<llama_token> embd_end;
std::vector<llama_token> inp_pfx = ::llama_tokenize(ctx, params.input_prefix, false);
std::vector<llama_token> inp_sfx = ::llama_tokenize(ctx, params.input_suffix, false);
const int space_token = 29871;
@@ -218,12 +219,13 @@ int main(int argc, char ** argv) {
inp_sfx.erase(inp_sfx.begin());
}
inp_pfx.insert(inp_pfx.begin(), llama_token_prefix(model));
if (add_bos) {
inp_pfx.insert(inp_pfx.begin(), llama_token_bos(model));
}
inp_sfx.insert(inp_sfx.begin(), llama_token_suffix(model));
embd_inp = inp_pfx;
embd_inp.insert(embd_inp.end(), inp_sfx.begin(), inp_sfx.end());
embd_inp = params.spm_infill ? inp_sfx : inp_pfx;
embd_end = params.spm_infill ? inp_pfx : inp_sfx;
if (add_bos) {
embd_inp.insert(embd_inp.begin(), llama_token_bos(model));
}
embd_inp.insert(embd_inp.end(), embd_end.begin(), embd_end.end());
const llama_token middle_token = llama_token_middle(model);
if (middle_token >= 0) {
@@ -527,14 +529,14 @@ int main(int argc, char ** argv) {
inp_sfx.erase(inp_sfx.begin());
}
inp_pfx.insert(inp_pfx.begin(), llama_token_prefix(model));
if (add_bos) {
inp_pfx.insert(inp_pfx.begin(), llama_token_bos(model));
}
inp_sfx.insert(inp_sfx.begin(), llama_token_suffix(model));
embd_inp = inp_pfx;
embd_inp.insert(embd_inp.end(), inp_sfx.begin(), inp_sfx.end());
embd_inp = params.spm_infill ? inp_sfx : inp_pfx;
embd_end = params.spm_infill ? inp_pfx : inp_sfx;
if (add_bos) {
embd_inp.insert(embd_inp.begin(), llama_token_bos(model));
}
embd_inp.insert(embd_inp.end(), embd_end.begin(), embd_end.end());
const llama_token middle_token = llama_token_middle(model);
if (middle_token >= 0) {
embd_inp.push_back(middle_token);
}
examples/json_schema_to_grammar.py
+3 -3
View File
@@ -231,7 +231,7 @@ GRAMMAR_RANGE_LITERAL_ESCAPE_RE = re.compile(r'[\r\n"\]\-\\]')
GRAMMAR_LITERAL_ESCAPES = {'\r': '\\r', '\n': '\\n', '"': '\\"', '-': '\\-', ']': '\\]'}
NON_LITERAL_SET = set('|.()[]{}*+?')
ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = set('[]()|{}*+?')
ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = set('^$.[]()|{}*+?')
class SchemaConverter:
@@ -602,7 +602,7 @@ class SchemaConverter:
else:
add_component(t, is_required=True)
return self._add_rule(rule_name, self._build_object_rule(properties, required, hybrid_name, additional_properties=[]))
return self._add_rule(rule_name, self._build_object_rule(properties, required, hybrid_name, additional_properties=None))
elif schema_type in (None, 'array') and ('items' in schema or 'prefixItems' in schema):
items = schema.get('items') or schema['prefixItems']
@@ -691,7 +691,7 @@ class SchemaConverter:
required_props = [k for k in sorted_props if k in required]
optional_props = [k for k in sorted_props if k not in required]
if additional_properties != False:
if additional_properties is not None and additional_properties != False:
sub_name = f'{name}{"-" if name else ""}additional'
value_rule = self.visit(additional_properties, f'{sub_name}-value') if isinstance(additional_properties, dict) else \
self._add_primitive('value', PRIMITIVE_RULES['value'])
examples/llama.android/llama/src/main/cpp/llama-android.cpp
+1 -1
View File
@@ -5,7 +5,7 @@
#include <string>
#include <unistd.h>
#include "llama.h"
#include "common/common.h"
#include "common.h"
// Write C++ code here.
//
examples/llava/clip.cpp
+13 -13
View File
@@ -1121,20 +1121,20 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
}
if (n < 32)
hparams.image_grid_pinpoints[n] = 0;
} catch (std::runtime_error & e) {
} catch (std::runtime_error & /*e*/) {
hparams.image_grid_pinpoints[0]=0;
}
try {
int idx = get_key_idx(ctx, KEY_MM_PATCH_MERGE_TYPE);
strcpy(hparams.mm_patch_merge_type, gguf_get_val_str(ctx, idx));
} catch (std::runtime_error & e) {
} catch (std::runtime_error & /*e*/) {
strcpy(hparams.mm_patch_merge_type, "flat");
}
try {
hparams.image_crop_resolution = get_u32(ctx, KEY_IMAGE_CROP_RESOLUTION); // llava-1.6
} catch(const std::exception& e) {
} catch(const std::exception& /*e*/) {
hparams.image_crop_resolution = hparams.image_size;
}
@@ -1173,7 +1173,7 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
try {
vision_model.class_embedding = get_tensor(new_clip->ctx_data, TN_CLASS_EMBD);
new_clip->has_class_embedding = true;
} catch (const std::exception& e) {
} catch (const std::exception& /*e*/) {
new_clip->has_class_embedding = false;
}
@@ -1181,7 +1181,7 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
vision_model.pre_ln_w = get_tensor(new_clip->ctx_data, format(TN_LN_PRE, "v", "weight"));
vision_model.pre_ln_b = get_tensor(new_clip->ctx_data, format(TN_LN_PRE, "v", "bias"));
new_clip->has_pre_norm = true;
} catch (std::exception & e) {
} catch (std::exception & /*e*/) {
new_clip->has_pre_norm = false;
}
@@ -1189,21 +1189,21 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
vision_model.post_ln_w = get_tensor(new_clip->ctx_data, format(TN_LN_POST, "v", "weight"));
vision_model.post_ln_b = get_tensor(new_clip->ctx_data, format(TN_LN_POST, "v", "bias"));
new_clip->has_post_norm = true;
} catch (std::exception & e) {
} catch (std::exception & /*e*/) {
new_clip->has_post_norm = false;
}
try {
vision_model.patch_bias = get_tensor(new_clip->ctx_data, TN_PATCH_BIAS);
new_clip->has_patch_bias = true;
} catch (std::exception & e) {
} catch (std::exception & /*e*/) {
new_clip->has_patch_bias = false;
}
try {
vision_model.patch_embeddings = get_tensor(new_clip->ctx_data, TN_PATCH_EMBD);
vision_model.position_embeddings = get_tensor(new_clip->ctx_data, format(TN_POS_EMBD, "v"));
} catch(const std::exception& e) {
} catch(const std::exception& /*e*/) {
LOG_TEE("%s: failed to load vision model tensors\n", __func__);
}
@@ -1215,26 +1215,26 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
// Yi-type llava
vision_model.mm_1_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 1, "weight"));
vision_model.mm_1_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 1, "bias"));
} catch (std::runtime_error & e) { }
} catch (std::runtime_error & /*e*/) { }
try {
// missing in Yi-type llava
vision_model.mm_2_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 2, "weight"));
vision_model.mm_2_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 2, "bias"));
} catch (std::runtime_error & e) { }
} catch (std::runtime_error & /*e*/) { }
try {
// Yi-type llava
vision_model.mm_3_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 3, "weight"));
vision_model.mm_3_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 3, "bias"));
} catch (std::runtime_error & e) { }
} catch (std::runtime_error & /*e*/) { }
try {
// Yi-type llava
vision_model.mm_4_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 4, "weight"));
vision_model.mm_4_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 4, "bias"));
} catch (std::runtime_error & e) { }
} catch (std::runtime_error & /*e*/) { }
try {
vision_model.image_newline = get_tensor(new_clip->ctx_data, TN_IMAGE_NEWLINE);
// LOG_TEE("%s: image_newline tensor (llava-1.6) found\n", __func__);
} catch (std::runtime_error & e) { }
} catch (std::runtime_error & /*e*/) { }
} else if (new_clip->proj_type == PROJECTOR_TYPE_LDP) {
// MobileVLM projection
vision_model.mm_model_mlp_1_w = get_tensor(new_clip->ctx_data, format(TN_MVLM_PROJ_MLP, 1, "weight"));
examples/main/main.cpp
+1 -1
View File
@@ -811,7 +811,7 @@ int main(int argc, char ** argv) {
is_antiprompt = true;
}
chat_add_and_format(model, chat_msgs, "system", assistant_ss.str());
chat_add_and_format(model, chat_msgs, "assistant", assistant_ss.str());
is_interacting = true;
printf("\n");
}
examples/server/README.md
+1
View File
@@ -73,6 +73,7 @@ The project is under active development, and we are [looking for feedback and co
- `-fa`, `--flash-attn` : enable flash attention (default: disabled).
- `-ctk TYPE`, `--cache-type-k TYPE` : KV cache data type for K (default: `f16`, options `f32`, `f16`, `q8_0`, `q4_0`, `q4_1`, `iq4_nl`, `q5_0`, or `q5_1`)
- `-ctv TYPE`, `--cache-type-v TYPE` : KV cache type for V (default `f16`, see `-ctk` for options)
- `--spm-infill` : Use Suffix/Prefix/Middle pattern for infill (instead of Prefix/Suffix/Middle) as some models prefer this.
**If compiled with `LLAMA_SERVER_SSL=ON`**
- `--ssl-key-file FNAME`: path to file a PEM-encoded SSL private key
examples/server/public/json-schema-to-grammar.mjs
+2 -2
View File
@@ -259,7 +259,7 @@ const GRAMMAR_RANGE_LITERAL_ESCAPE_RE = /[\n\r"\]\-\\]/g;
const GRAMMAR_LITERAL_ESCAPES = { '\r': '\\r', '\n': '\\n', '"': '\\"', '-': '\\-', ']': '\\]' };
const NON_LITERAL_SET = new Set('|.()[]{}*+?');
const ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = new Set('[]()|{}*+?');
const ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = new Set('^$.[]()|{}*+?');
export class SchemaConverter {
constructor(options) {
@@ -751,7 +751,7 @@ export class SchemaConverter {
const requiredProps = sortedProps.filter(k => required.has(k));
const optionalProps = sortedProps.filter(k => !required.has(k));
if (additionalProperties !== false) {
if (additionalProperties) {
const subName = `${name ?? ''}${name ? '-' : ''}additional`;
const valueRule =
additionalProperties != null && typeof additionalProperties === 'object' ? this.visit(additionalProperties, `${subName}-value`)
examples/server/server.cpp
+11 -5
View File
@@ -2021,6 +2021,7 @@ struct server_context {
slot.t_start_generation = 0;
if (slot.infill) {
const bool add_bos = llama_should_add_bos_token(model);
bool suff_rm_leading_spc = true;
if (params.input_suffix.find_first_of(' ') == 0 && params.input_suffix.size() > 1) {
params.input_suffix.erase(0, 1);
@@ -2036,16 +2037,21 @@ struct server_context {
}
prefix_tokens.insert(prefix_tokens.begin(), llama_token_prefix(model));
prefix_tokens.insert(prefix_tokens.begin(), llama_token_bos(model)); // always add BOS
prefix_tokens.insert(prefix_tokens.end(), llama_token_suffix(model));
prefix_tokens.insert(prefix_tokens.end(), suffix_tokens.begin(), suffix_tokens.end());
suffix_tokens.insert(suffix_tokens.begin(), llama_token_suffix(model));
auto embd_inp = params.spm_infill ? suffix_tokens : prefix_tokens;
auto embd_end = params.spm_infill ? prefix_tokens : suffix_tokens;
if (add_bos) {
embd_inp.insert(embd_inp.begin(), llama_token_bos(model));
}
embd_inp.insert(embd_inp.end(), embd_end.begin(), embd_end.end());
const llama_token middle_token = llama_token_middle(model);
if (middle_token >= 0) {
prefix_tokens.push_back(middle_token);
embd_inp.push_back(middle_token);
}
prompt_tokens = prefix_tokens;
prompt_tokens = embd_inp;
} else {
prompt_tokens = tokenize(slot.prompt, system_prompt.empty()); // add BOS if there isn't system prompt
}