0
0
Fork 0
mirror of https://github.com/matrix-construct/construct synced 2024-12-12 00:22:58 +01:00
construct/ircd/gpt_model.cc

758 lines
19 KiB
C++
Raw Normal View History

2021-03-05 02:03:33 +01:00
// Tensor Construct
//
// Copyright (C) Matrix Construct Developers, Authors & Contributors
// Copyright (C) 2016-2021 Jason Volk <jason@zemos.net>
//
// Permission to use, copy, modify, and/or distribute this software for any
// purpose with or without fee is hereby granted, provided that the above
// copyright notice and this permission notice is present in all copies. The
// full license for this software is available in the LICENSE file.
namespace ircd::gpt::model
{
using init_func = void (*)(decoder &, const string_view &, const size_t &, const json::array &);
using init_handler = std::pair<string_view, init_func>;
2021-03-05 02:03:33 +01:00
static void
init_f_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_f_bias(decoder &, const string_view &, const size_t &, const json::array &),
init_wpe_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_wte_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ffnn_fc_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ffnn_fc_bias(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ffnn_proj_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ffnn_proj_bias(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ln_1_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ln_1_bias(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ln_2_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_ln_2_bias(decoder &, const string_view &, const size_t &, const json::array &),
init_h_attn_attn_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_attn_attn_bias(decoder &, const string_view &, const size_t &, const json::array &),
init_h_attn_proj_weight(decoder &, const string_view &, const size_t &, const json::array &),
init_h_attn_proj_bias(decoder &, const string_view &, const size_t &, const json::array &);
2021-03-05 02:03:33 +01:00
static bool init_dataset(const string_view &);
static bool init_from_cache(const string_view &);
static void init_from_json_handle(decoder &, const init_handler &, const size_t &);
static void init_from_json(const string_view &, const string_view &);
static void init(), fini() noexcept;
extern const init_handler
2021-03-05 02:03:33 +01:00
manifest[],
manifest_h[];
extern conf::item<bool>
cache_mapped,
cache_locked,
cache_shared,
cache_hugepage;
extern conf::item<std::string>
path,
cache_path,
dataset_path;
static fs::map
default_model_shm,
default_dataset_shm;
2021-03-05 02:03:33 +01:00
}
2022-06-20 03:59:29 +02:00
constexpr const char
*const ircd::gpt::model::prop::ended,
*const ircd::gpt::model::prop::id,
*const ircd::gpt::model::prop::length,
*const ircd::gpt::model::prop::text;
2021-03-05 02:03:33 +01:00
decltype(ircd::gpt::model::manifest_h)
ircd::gpt::model::manifest_h
{
{ "h.%u.mlp.c_fc.weight.json", init_h_ffnn_fc_weight, },
{ "h.%u.mlp.c_fc.bias.json", init_h_ffnn_fc_bias, },
{ "h.%u.mlp.c_proj.weight.json", init_h_ffnn_proj_weight, },
{ "h.%u.mlp.c_proj.bias.json", init_h_ffnn_proj_bias, },
{ "h.%u.ln_1.weight.json", init_h_ln_1_weight, },
{ "h.%u.ln_1.bias.json", init_h_ln_1_bias, },
{ "h.%u.ln_2.weight.json", init_h_ln_2_weight, },
{ "h.%u.ln_2.bias.json", init_h_ln_2_bias, },
{ "h.%u.attn.c_attn.weight.json", init_h_attn_attn_weight, },
{ "h.%u.attn.c_attn.bias.json", init_h_attn_attn_bias, },
{ "h.%u.attn.c_proj.weight.json", init_h_attn_proj_weight, },
{ "h.%u.attn.c_proj.bias.json", init_h_attn_proj_bias },
};
decltype(ircd::gpt::model::manifest)
ircd::gpt::model::manifest
{
{ "ln_f.weight.json", init_f_weight, },
{ "ln_f.bias.json", init_f_bias, },
{ "wpe.weight.json", init_wpe_weight },
{ "wte.weight.json", init_wte_weight },
};
decltype(ircd::gpt::model::cache_mapped)
ircd::gpt::model::cache_mapped
{
{ "name", "ircd.gpt.model.cache.mapped" },
{ "default", true },
};
decltype(ircd::gpt::model::cache_locked)
ircd::gpt::model::cache_locked
{
{ "name", "ircd.gpt.model.cache.locked" },
{ "default", false },
};
decltype(ircd::gpt::model::cache_shared)
ircd::gpt::model::cache_shared
{
{ "name", "ircd.gpt.model.cache.shared" },
{ "default", false },
};
decltype(ircd::gpt::model::cache_hugepage)
ircd::gpt::model::cache_hugepage
{
{ "name", "ircd.gpt.model.cache.hugepage" },
2022-06-20 03:59:29 +02:00
{ "default", false },
};
decltype(ircd::gpt::model::cache_path)
ircd::gpt::model::cache_path
{
{ "name", "ircd.gpt.model.cache.path" },
{ "default", "model.cache.localhost" },
};
decltype(ircd::gpt::model::dataset_path)
ircd::gpt::model::dataset_path
{
{ "name", "ircd.gpt.model.dataset.path" },
{ "default", string_view{} },
};
2021-03-05 02:03:33 +01:00
decltype(ircd::gpt::model::path)
ircd::gpt::model::path
{
{
{ "name", "ircd.gpt.model.path" },
{ "default", string_view{} },
},
init
};
decltype(ircd::gpt::model::default_model)
ircd::gpt::model::default_model;
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
decltype(ircd::gpt::model::default_moment)
ircd::gpt::model::default_moment;
decltype(ircd::gpt::model::default_checkpoint)
ircd::gpt::model::default_checkpoint;
decltype(ircd::gpt::model::default_dataset)
ircd::gpt::model::default_dataset;
decltype(ircd::gpt::model::default_data)
ircd::gpt::model::default_data;
2021-03-05 02:03:33 +01:00
void
ircd::gpt::model::init()
{
if(!model::path)
return;
if(model::dataset_path)
init_dataset(model::dataset_path);
2022-06-20 03:59:29 +02:00
if(likely(init_from_cache(model::cache_path)))
return;
init_from_json(model::cache_path, model::path);
if(unlikely(!init_from_cache(model::cache_path)))
throw error
{
"Failed to find and/or initialize model."
};
}
void
ircd::gpt::model::fini()
noexcept
{
2022-06-20 03:59:29 +02:00
default_checkpoint[2] = nullptr;
default_checkpoint[1] = nullptr;
default_checkpoint[0] = nullptr;
default_moment[1] = nullptr;
default_moment[0] = nullptr;
if(!cache_mapped)
delete default_model;
default_model = nullptr;
default_model_shm = {};
default_dataset = nullptr;
default_data.clear();
default_dataset_shm = {};
}
bool
ircd::gpt::model::init_from_cache(const string_view &cache_path)
{
if(!fs::is_reg(cache_path))
return false;
2022-06-20 03:59:29 +02:00
const auto file_size
2021-03-05 02:03:33 +01:00
{
fs::size(cache_path)
2021-03-05 02:03:33 +01:00
};
2022-06-20 03:59:29 +02:00
const auto decoder_size
{
sizeof(model::decoder)
};
const bool has_params
{
file_size >= decoder_size
};
const bool has_moments
{
file_size >= decoder_size * 6
};
if(unlikely(!has_params))
throw error
2021-03-05 02:03:33 +01:00
{
2022-06-20 03:59:29 +02:00
"Cached model `%s' size %zu insufficient for decoder size %zu.",
cache_path,
2022-06-20 03:59:29 +02:00
file_size,
decoder_size,
2021-03-05 02:03:33 +01:00
};
const auto mode
{
cache_shared?
std::ios::in | std::ios::out:
std::ios::in
};
const fs::fd fd
{
2022-06-20 03:59:29 +02:00
cache_path, fs::fd::opts
{
.mode = mode,
},
2021-04-17 20:59:30 +02:00
};
2022-06-20 03:59:29 +02:00
const bool map_moments
2021-04-17 20:59:30 +02:00
{
2022-06-20 03:59:29 +02:00
has_moments || cache_shared
};
if(!has_moments && map_moments)
{
fs::truncate(fd, decoder_size * 6);
fs::allocate(fd, decoder_size * 5, decoder_size);
}
const auto map_size
{
map_moments?
decoder_size * 6:
decoder_size
};
fs::map::opts map_opts
2022-06-20 03:59:29 +02:00
{
.alignment = alignof(model::decoder),
.shared = bool(cache_shared),
.locked = bool(cache_locked),
.huge2mb = bool(cache_hugepage),
};
map_opts.mode = mode;
// amdgpu requires both anon and shms to be read-write even if we
// open the fd read-only and use read-only cl_mems.
if(cache_mapped)
map_opts.mode |= std::ios::out;
default_model_shm = fs::map
{
fd, map_size, map_opts,
};
2021-03-05 02:03:33 +01:00
default_model = reinterpret_cast<decoder *>
(
cache_mapped?
data(default_model_shm):
allocator::allocate(info::page_size, map_size)
);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
if(map_moments)
{
default_moment[0] = reinterpret_cast<float *>(default_model + 1);
default_moment[1] = reinterpret_cast<float *>(default_model + 2);
default_checkpoint[0] = reinterpret_cast<float *>(default_model + 3);
default_checkpoint[1] = reinterpret_cast<float *>(default_model + 4);
default_checkpoint[2] = reinterpret_cast<float *>(default_model + 5);
}
if(cache_mapped)
fs::prefetch(default_model_shm, sizeof(decoder));
if(!cache_mapped)
memcpy(default_model, data(default_model_shm), map_size);
if(!cache_mapped && !cache_shared)
default_model_shm = {};
2022-06-20 03:59:29 +02:00
char pbuf[48];
log::info
{
log, "model(%p) %s cached model `%s' shared:%b params:%b moments:%b align:%u %s",
default_model,
cache_mapped?
"mapped"_sv:
"loaded"_sv,
cache_path,
bool(cache_shared),
2022-06-20 03:59:29 +02:00
has_params,
has_moments,
map_opts.alignment,
pretty(pbuf, iec(map_size)),
};
2021-03-05 02:03:33 +01:00
return true;
}
2021-03-05 02:03:33 +01:00
void
ircd::gpt::model::init_from_json(const string_view &cache_path,
const string_view &model_path)
{
util::timer stopwatch;
auto decoder(std::make_unique<model::decoder>());
memset(decoder.get(), 0x0, sizeof(model::decoder));
// Load the top level files, vocab etc
for(size_t i(0); i < 4; ++i)
init_from_json_handle(*decoder, manifest[i], 0);
// Load the transformer files by layer
const size_t layers {12};
for(size_t i(0); i < layers; ++i)
for(size_t j(0); j < 12; ++j)
init_from_json_handle(*decoder, manifest_h[j], i);
const const_buffer src
{
reinterpret_cast<char *>(decoder.get()), sizeof(model::decoder)
};
2021-03-05 02:03:33 +01:00
const auto wrote
{
fs::write(cache_path, src)
};
2021-03-05 02:03:33 +01:00
char pbuf[2][48];
log::info
{
log, "model(%p) parsed `%s' cached %s to `%s' in %s",
decoder.get(),
model_path,
pretty(pbuf[0], iec(size(wrote))),
cache_path,
stopwatch.pretty(pbuf[1]),
};
}
void
ircd::gpt::model::init_from_json_handle(decoder &d,
const init_handler &handler,
const size_t &layer)
{
const auto &[fmt, func]
{
handler
};
char namebuf[128];
const string_view path_part[2]
{
model::path, fmt::sprintf
2021-03-05 02:03:33 +01:00
{
namebuf, fmt, layer
}
};
2021-03-05 02:03:33 +01:00
const auto path
{
fs::path(fs::path_scratch, path_part)
};
const fs::fd::opts fd_opts
{
.mode = std::ios::in,
.sequential = true,
};
const fs::fd fd
{
path, fd_opts
};
// mmap of the file
const fs::map map
{
fd, size(fd), fs::map::opts{fd_opts},
};
// Each file is a JSON array at the top level.
const json::array matrix
{
map
};
// Readable name for logging
const auto &name
{
path_part[1]
};
if(likely(func))
func(d, name, layer, matrix);
// Check for interrupt after long operation
ctx::interruption_point();
2021-03-05 02:03:33 +01:00
log::info
2021-03-05 02:03:33 +01:00
{
log, "model(%p) loaded layer:%zu :%s",
&d,
layer,
name,
2021-03-05 02:03:33 +01:00
};
}
bool
ircd::gpt::model::init_dataset(const string_view &path)
{
if(!fs::is_reg(path))
return false;
const auto size
{
fs::size(path)
};
const fs::fd::opts fd_opts
{
.mode = std::ios::in,
};
const fs::fd fd
{
path, fd_opts,
};
fs::map::opts map_opts{fd_opts};
map_opts.huge2mb = bool(cache_hugepage);
default_dataset_shm = fs::map
{
fd, size, map_opts
};
default_dataset = string_view
(
default_dataset_shm
);
size_t checkpoint(0);
default_data.resize(260000); //TODO: XXX
2022-06-20 03:59:29 +02:00
fs::prefetch(default_dataset_shm, size);
ircd::tokens(default_dataset, '\n', [&checkpoint]
(const string_view &line)
{
default_data.at(checkpoint++) = line;
});
char pbuf[48];
log::info
{
log, "dataset(%p) mapped `%s' %s @%lu",
data(default_dataset_shm),
path,
pretty(pbuf, iec(size)),
checkpoint,
};
2022-06-20 03:59:29 +02:00
fs::evict(default_dataset_shm, size);
return true;
}
2021-03-05 02:03:33 +01:00
void
ircd::gpt::model::init_wpe_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &mat)
{
size_t i(0);
for(const json::array vec : mat)
{
size_t j(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.embed.pos[i].elem[j++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(j == sizeof(d.embed.pos[i]) / sizeof(float));
2021-03-05 02:03:33 +01:00
++i;
}
}
void
ircd::gpt::model::init_wte_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &mat)
{
size_t i(0);
for(const json::array vec : mat)
{
size_t j(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.embed.token[i].elem[j++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(j == sizeof(d.embed.token[i]) / sizeof(float));
2021-03-05 02:03:33 +01:00
++i;
}
}
void
ircd::gpt::model::init_f_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.embed.norm.weight.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.embed.norm.weight) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_f_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.embed.norm.bias.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.embed.norm.bias) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_ffnn_fc_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &mat)
{
size_t i(0);
for(const json::array vec : mat)
{
size_t j(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].ffnn.fcon_weight[i].fcon[j++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(j == sizeof(d.layer[layer].ffnn.fcon_weight[i]) / sizeof(float));
2021-03-05 02:03:33 +01:00
++i;
}
always_assert
(
2022-06-20 03:59:29 +02:00
i == sizeof(d.layer[layer].ffnn.fcon_weight)
/ sizeof(d.layer[layer].ffnn.fcon_weight[0])
2021-03-05 02:03:33 +01:00
);
}
void
ircd::gpt::model::init_h_ffnn_fc_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].ffnn.fcon_bias.fcon[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.layer[layer].ffnn.fcon_bias) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_ffnn_proj_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &mat)
{
size_t i(0);
for(const json::array vec : mat)
{
size_t j(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].ffnn.proj_weight[i].elem[j++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
always_assert(j == sizeof(d.layer[layer].ffnn.proj_weight[i]) / sizeof(float));
++i;
}
always_assert
(
i == sizeof(d.layer[layer].ffnn.proj_weight)
/ sizeof(d.layer[layer].ffnn.proj_weight[0])
);
}
void
ircd::gpt::model::init_h_ffnn_proj_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].ffnn.proj_bias.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
always_assert(i == sizeof(d.layer[layer].ffnn.proj_bias) / sizeof(float));
}
void
ircd::gpt::model::init_h_ln_1_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].attn.norm.weight.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.layer[layer].attn.norm.weight) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_ln_1_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].attn.norm.bias.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.layer[layer].attn.norm.bias) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_ln_2_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].ffnn.norm.weight.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.layer[layer].ffnn.norm.weight) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_ln_2_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].ffnn.norm.bias.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.layer[layer].ffnn.norm.bias) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_attn_attn_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &mat)
{
size_t i(0);
for(const json::array vec : mat)
{
size_t j(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].attn.fcon_weight[i].fcon[j++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(j == sizeof(d.layer[layer].attn.fcon_weight[i]) / sizeof(float));
2021-03-05 02:03:33 +01:00
++i;
}
always_assert
(
2022-06-20 03:59:29 +02:00
i == sizeof(d.layer[layer].attn.fcon_weight)
/ sizeof(d.layer[layer].attn.fcon_weight[0])
2021-03-05 02:03:33 +01:00
);
}
void
ircd::gpt::model::init_h_attn_attn_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].attn.fcon_bias.fcon[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
2022-06-20 03:59:29 +02:00
always_assert(i == sizeof(d.layer[layer].attn.fcon_bias) / sizeof(float));
2021-03-05 02:03:33 +01:00
}
void
ircd::gpt::model::init_h_attn_proj_weight(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &mat)
{
size_t i(0);
for(const json::array vec : mat)
{
size_t j(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].attn.proj_weight[i].elem[j++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
always_assert(j == sizeof(d.layer[layer].attn.proj_weight[i]) / sizeof(float));
++i;
}
always_assert
(
i == sizeof(d.layer[layer].attn.proj_weight)
/ sizeof(d.layer[layer].attn.proj_weight[0])
);
}
void
ircd::gpt::model::init_h_attn_proj_bias(decoder &d,
const string_view &name,
const size_t &layer,
const json::array &vec)
{
size_t i(0);
for(const auto &elem : vec)
2022-06-20 03:59:29 +02:00
d.layer[layer].attn.proj_bias.elem[i++] = lex_cast<float>(elem);
2021-03-05 02:03:33 +01:00
always_assert(i == sizeof(d.layer[layer].attn.proj_bias) / sizeof(float));
}