Commit Graph

Select branches

Hide Pull Requests

abstractions_4

add_local_buffer

after_qol

algebraic_upat

all_bv_no_idx

all_sink

amd_isel

amd_refac

amd_smu_reset_patch

amd_sqtt

amd_uop

amd_uop_cleanups

assign_copy

assign_try_3

before2

better_spec

better_syn

bitcast_spec

broadcast_shape_2

bt_ei

buf_heur

bump_nv_2404

cache_reshape

call_inline

call_is_kernels

callify

cast_vec_as_cast

cat_mop

changes_from_rangify

check_clsetarg

check_dims

cifar_30

cifar_30_clean

clean_load

clean_spec

clone_tg

codegen_try_2

contig_on_after

continue_mat

deepsek

default_late_allreduce

delete_kernel_py

delete_ones

delete_slow_rangeify

delta_attention

delta_attention_scan

demote_op

disable_rules

dsp_search

dsp_search_merged

dtype_shape

earlier_gater

emu_cdna

emulate_ctx

end_all_ranges

endrange

er_prioity

execution_unit

failing_assign

fancy_usb

fast_llm

fix_fa_fwd_clang_22

fix_fuse

fix_llvm_cast

fix_range_merging_sd

fix_rangeify_tests

fix_rc

fix_sched_copy

fix_slice_store

fix_src_spec

fixup_fix_kernel

full_spec

fun_w_egraphs

gate-on-load-store

gate_rewrite

gemma4_gpt

gen_pdf_fast

gh-pages

gl_dims

gldims

glm_flash

grad_outerworld

grok_x86_novec

group_div_rules

hcq2_vram_copies

heur_4096

heur_group

hk_gemm

http_threading

hwtest_fixes

id_store

if_range_cleanups

ignore_oob

image_idx_xy

image_in_mc

image_no_vec

image_refactors_stack

index_to_slice

invalid_try_3

jinja_template

kernel4

kernel_is_call

kernelless

kimi_slop

kitten_matmul

late_Locals

late_add_load

late_allreduce

late_codegen

late_decomp_is_late

late_lower_idx

late_view

late_weakint

lds_double_buffering

less_wmma_arg

lil_dg_change

lil_rangeify_td

llama_trainer

llm_262k_minimal

llm_flash_attn

llm_qwen

llm_serve_fixes

load_view_late

local_cache

locals_have_device

loop_end_op

loop_is_rng

lowerer_hacks

mac_pytest

manual_kernel_3

master

mega_lowerer

min_new_lin

min_outer_range

minor_views

mixins7

mlperf_training_v4.1

mlperf_training_v5.0

mlperf_training_v6.0

modern_scan

more_apps

more_early_comps

more_folding_rules

more_kernelcount

more_pcode

more_rangeify_pm_tests

more_ren_clean

more_uop_gc_test

more_uop_mm

more_vmap

move_devec

move_dtype_decomps

move_expander

move_gates_to_load_store

move_mop_cleanup

move_schedule

move_simp_views

moveleftright

mt_typing

multi_range

multioutput

new_ck

new_devec

new_invalid_and_image

new_linearizer

new_shape

new_test

no_assign_2

no_buf_0

no_decimals

no_late_simplify_marg

no_merge_ends

no_merge_views

no_mstack_in_allreduce

no_one_ranges

no_sip_usbgpu

no_tsink_base

no_tuplize

no_user_ck_contig

no_vec_x86_4

no_weakint_variable

noop_index

novecimg

num_batches_tracked_shape

olmo3_support

one_compiler

one_hcopt

only_assign_buffers

only_loop_backends

only_reg_emu

only_reg_emu2

outer_range

outerworld_work

pad_invalid_work

pad_mixin

param_call_is_function

parse_sqtt_more

pci_iommu

port_nir

postrange_hcopts

precompile_backward

prefill_qwen_fast

print_special_ops

ps_clean

pyremu_try

python_prg

python_speed

q4k_support

qcom-openpilot-13ms

qcom_830

qcom_mmapeak

qcom_speed_codex

qol_string

qwen36_27b_amd_900

qwen36_27b_amd_minimal

qwen36_amd_bugfixes

qwen_mergable

r_cleanups

r_new_opt

ranged_call

rangeify_2

rangeify_const_folding

rdna3_vibes

rdna4_asm

rdna4_gemm

real_pyrender

recurse_sub

reg_coalese

relu_pattern

remove_assign

remove_cast_hack

remove_ck_contig

remove_define_var

remove_forced_re

remove_on_stack

remove_programspec

remove_pylint

remove_slow_tests

remove_vec_2

remove_vec_3

renderers_use_shapes_2

replace_if_with_range

reprocess_node_experiments

reshape_trait

resnet-ci

revert-11547-fix_mismatch_red

revert-11715-revert-11711-amd_llvm_tc_cast

revert-11809-no_ki_in_gpudims

revert-11811-better_div_nesting

revert-11876-memplan_vbig

revert-12104-simplify_valid_in_symbolic

revert-12308-bufferize_early

revert-12731-jitlink_init_check

revert-12813-after_cleanups

revert-13046-better_cleanup_arange_buffers

revert-13311-prof_link_sqtt

revert-13583-rocr_desc_2

revert-14734-tload

revert-17141-full_rewrite_to_sink

revert-17408-_broadcasted-weak

revert-9438-simple_matmul_np_rand

revert_wrong_ck_fix

rewrite_rangeify2

rtoposort

scan_assign

scan_op_8

sched_cache

sched_lin

sd_load_simple

self_type

set_testpath

shrink_in_render

simp_spec

simple_gidxs

simple_pool

simple_priority

simple_rangeify

simpler-sd-compile

simpler_end

simpler_fusion

simpler_image

simpler_postrange

simpler_reduce

slice_like_shrink

spec_3

spec_cleanup

spec_ren

speed_one_linear

split_buf_idx

split_decomp_pass

split_sub

stack_0

store_is_void

string_ranges

support_opts_in_contig

swizzle_in_kernel

sym_llm

sym_work

symbolic-chunk-split

szdiff_bot_test

tc_is_heur

tensor_graph_single_sink

test_amd_crash

test_fa

test_from_postopt

test_hgemm

test_unit

tiny_scan

tmr

tool_call_support

topoprop

ttomsa-new_linearizer

tuplegtuple

two_stage_remove

typed_checks

typing_work

unused_permute_arg

uop_prg

upd_test4

update_benchmark

update_benchmark_search

update_mlperf

user_failure

vecless_x86_glm

verify_sink_dims_clean

vf_ish

view_after_const

view_in_codegen

viz_cbroadcast

warp_fun

wgpu-f16-packed

working_postopt

x86_moves

x86_numel

v0.10.0

v0.10.1

v0.10.2

v0.10.3

v0.11.0

v0.12.0

v0.13.0

v0.4.0

v0.5.0

v0.6.0

v0.7.0

v0.8.0

v0.9.0

v0.9.1

v0.9.2

Mono Color

• e86f7a4aa3 deterministic geohot 2021-11-29 15:10:15 -05:00
• f909ab194f gelu with broken test geohot 2021-11-29 15:00:50 -05:00
• 9ce881f88c fix bug in getitem, drop int axis geohot 2021-11-29 14:01:24 -05:00
• c752033283 fix GPU OOM in test geohot 2021-11-29 13:05:59 -05:00
• 1eafa5580e layernorm with learnable parameters geohot 2021-11-29 13:03:57 -05:00
• c7f795ca1e added dot affine geohot 2021-11-29 12:55:56 -05:00
• 30eb3afbe1 add bias term to transformer geohot 2021-11-29 12:45:27 -05:00
• 99b6051467 add ff_dim to transformer geohot 2021-11-29 12:40:52 -05:00
• b4839eb6bb factor out binary_broadcast geohot 2021-11-29 03:49:12 -05:00
• ca154f8b59 add sub and a few refactors geohot 2021-11-29 03:35:23 -05:00
• 7b3c442f3b sum in metal geohot 2021-11-29 03:13:41 -05:00
• c6fb087695 fix old list behavior geohot 2021-11-29 01:49:21 -05:00
• 7c160c6cee metal: got relu working and add somewhat working geohot 2021-11-29 01:40:18 -05:00
• 29dee59368 cat: forward only not required geohot 2021-11-29 00:14:56 -05:00
• 06838481e8 use get data dtype geohot 2021-11-28 23:28:37 -05:00
• 3cdc77f526 add cat support geohot 2021-11-28 23:21:49 -05:00
• ce3d198bb7 less lines and fix default device geohot 2021-11-27 11:18:49 -05:00
• 7ae14179d3 refactor ops geohot 2021-11-27 11:12:23 -05:00
• 4320c45c4b oops, gpu not available geohot 2021-11-17 12:54:33 -08:00
• e2a8961a18 less lines, fix bug geohot 2021-11-17 12:52:17 -08:00
• e8ca3ad053 add Buffer abstraction for each device geohot 2021-11-17 12:46:07 -08:00
• c162e748f5 fix float64 warning on training geohot 2021-10-30 20:07:31 -07:00
• b0f14b4af8 move datasets into datasets geohot 2021-10-30 19:55:50 -07:00
• a95ef16c8c sub 1000 lines geohot 2021-10-30 19:48:24 -07:00
• 844540a5ed yolo in readme geohot 2021-10-30 19:47:34 -07:00
• ba28761894 move yolo into examples/yolo geohot 2021-10-30 19:46:00 -07:00
• 121d5a17ee use tinynn for Conv2d geohot 2021-10-30 19:40:44 -07:00
• 6bee5bdb7d add torch tests geohot 2021-10-30 18:58:45 -07:00
• 75260c6e49 unify more of cpu and torch geohot 2021-10-30 18:57:24 -07:00
• b1dae764d3 more similar between torch and numpy geohot 2021-10-30 18:46:59 -07:00
• d6cb373c79 import the torch ops that are the same geohot 2021-10-30 18:22:44 -07:00
• 51a49de299 tests pass for torch ops geohot 2021-10-30 18:00:12 -07:00
• 641b1dbb40 remove ane, start supporting ops_torch geohot 2021-10-30 17:47:00 -07:00
• 7d12482d80 refactor efficientnet loading geohot 2021-10-30 17:02:17 -07:00
• 7472a7ebe2 not forcing 3.9 for a stupid type geohot 2021-10-30 16:52:40 -07:00
• 114f6ca3fd more readme cleanup geohot 2021-10-30 16:51:25 -07:00
• fc6597a6d9 only resnet18, it's too slow otherwise geohot 2021-10-30 16:48:39 -07:00
• c05867dcbb header geohot 2021-10-30 16:41:05 -07:00
• e10dc90524 tpu docs geohot 2021-10-30 16:38:19 -07:00
• effd0dc833 update readme geohot 2021-10-30 16:34:00 -07:00
• f193eeed25 bump all to python 3.9 geohot 2021-10-30 16:15:41 -07:00
• 63f50cff45 move back again geohot 2021-10-30 16:13:29 -07:00
• 285621aeda Cherry backprop for conv2d (#281) Evan MaysandGitHub 2021-10-30 19:12:19 -04:00
• 3d646272d6 move back geohot 2021-10-30 16:12:12 -07:00
• ac8afd24fa refactor accel geohot 2021-10-30 16:10:59 -07:00
• 082089d1c7 install requires pillow geohot 2021-10-30 16:00:33 -07:00
• 8113eec4cf feat: add efficientnet test (#285) Sebastian KreftandGitHub 2021-10-30 19:53:51 -03:00
• d8e5758c89 needs pillow geohot 2021-10-27 10:14:02 -07:00
• 3358770182 chore(efficientnet): don't use eval when loading weights (#286) 3 people 2021-10-22 19:10:04 -03:00
• 5402fd7f05 Fix: (#279) Wenjun.ZhangandGitHub 2021-08-06 22:58:13 +08:00
• 9f732e697a Fix loading relu.hwx from the wrong path caused by moving the ane folder into the accel folder (#276) Zoli VeresandGitHub 2021-06-29 17:20:54 +03:00
• 2b7589db64 Added ResNet-{18, 34, 50, 101, 152} (#271) Guglielmo CamporeseandGitHub 2021-06-21 18:37:24 +02:00
• 89798d2f43 some flags geohot 2021-06-19 11:46:31 -07:00
• d81eae8288 debug cherry crash geohot 2021-06-19 11:41:20 -07:00
• 0eda3eb421 move ane and fpga to accel folder geohot 2021-06-19 11:26:20 -07:00
• d3f169b267 move good models to models, add a training step test geohot 2021-06-19 11:24:15 -07:00
• b48d4bad2e clean up print spam geohot 2021-06-19 10:30:51 -07:00
• 3a91d5434f Add dropout test (#265) Jacky LeeandGitHub 2021-06-19 08:49:13 -07:00
• ca0a38f2d5 more tpu notes geohot 2021-06-18 13:27:58 -07:00
• 027535d0b5 microcoded matmul geohot 2021-06-17 21:03:08 -07:00
• d1dd9b46f6 if i was giving systolic arrays thumbs like siskyl and ebert (RIP) i would give them 0 geohot 2021-06-17 19:48:58 -07:00
• 026e2ae6a7 three registers and a zero command geohot 2021-06-17 17:09:18 -07:00
• 2e71ae33f6 max op works geohot 2021-06-17 17:01:21 -07:00
• 9e12c1bbba cherry binop geohot 2021-06-17 16:50:40 -07:00
• fcdabea880 training mnist with cherry ops geohot 2021-06-17 16:45:35 -07:00
• 2affd226b3 speed up sum geohot 2021-06-17 16:38:34 -07:00
• e8eb7d1b7e max op geohot 2021-06-17 16:20:56 -07:00
• c1d469d440 sum op geohot 2021-06-17 16:19:35 -07:00
• d6517a8a7c ins geohot 2021-06-16 19:31:13 -07:00
• 29a08ba352 pytorch earlier geohot 2021-06-16 12:24:21 -07:00
• 4a07b71731 update business model geohot 2021-06-16 12:01:50 -07:00
• d29b16e5b4 more business notes geohot 2021-06-16 11:47:57 -07:00
• b1000d866e readme, plus reduce ops geohot 2021-06-16 11:21:06 -07:00
• ff3fdc58e5 risk -> cherry geohot 2021-06-16 09:59:48 -07:00
• 2f91c012eb build note geohot 2021-06-15 22:41:41 -07:00
• 0c02b66259 more geohot 2021-06-15 15:02:32 -07:00
• 1e62e45d67 better todo geohot 2021-06-15 10:30:16 -07:00
• 9ca4388695 debug geohot 2021-06-15 10:24:21 -07:00
• 3d44aab52c more geohot 2021-06-15 10:23:57 -07:00
• 4850d6eb43 update todo geohot 2021-06-15 10:22:39 -07:00
• 4e1edb3692 have tinygrad log the loads geohot 2021-06-14 18:35:14 -07:00
• 93f2e9769d little note geohot 2021-06-14 15:49:31 -07:00
• 611d81dcb4 Add asserts for non-zero indices (#264) Jacky LeeandGitHub 2021-06-13 21:14:46 -07:00
• 508ced114c readme geohot 2021-06-13 17:17:35 -07:00
• 2146860307 fixed OpenCL installation while running tests (#262) 3 people 2021-06-12 23:44:21 +05:30
• a89d12d735 wow, way faster geohot 2021-06-10 17:11:39 -07:00
• 10b1306525 binops geohot 2021-06-10 16:52:37 -07:00
• 4535d39baa comments and pow geohot 2021-06-10 09:03:40 -07:00
• 2075fdeb4f FPGA Based Accelerator for Tinygrad (#258) George HotzandGitHub 2021-06-07 17:45:09 -07:00
• 77ba198b57 Revert "Update README.md (#259)" (#260) George HotzandGitHub 2021-06-04 14:41:41 -07:00
• 5a69c5db6d Update README.md (#259) Gabriel RojasandGitHub 2021-06-04 17:41:07 -04:00
• ad756f6112 minor optimizations & cleaning (#257) Josh SmithandGitHub 2021-06-02 12:57:15 -04:00
• 74e874cc0d comment geohot 2021-05-26 18:06:55 -07:00
• 343c5f13c7 add output shape to DEBUG geohot 2021-05-26 17:42:38 -07:00
• b80cacb416 fix GPU efficientnet example geohot 2021-05-26 17:29:35 -07:00
• 1ae0e88627 nvidia notes geohot 2021-05-26 14:26:48 -07:00
• 2653d33292 vgg7 (image upscaling) implementation - not the best, but it works (#255) 20kdcandGitHub 2021-05-13 07:48:51 +01:00
• 81bf933a91 Improved __getitem__ (#254) SkoshandGitHub 2021-05-06 08:15:22 +03:00
• 78aa147b39 [WIP] YOLO working on tinygrad! (#245) SkoshandGitHub 2021-04-26 04:06:52 +03:00
• 155ec1f18e saving 50 LOC with automatic @staticmethod for forward and backward (#252) ziofilandGitHub 2021-04-25 21:04:16 -04:00