repo stringclasses 15
values | fix_commit stringlengths 40 40 | buggy_commit stringlengths 40 40 | message stringlengths 3 64.3k | files listlengths 1 300 | timestamp timestamp[s]date 2013-03-13 20:45:00 2026-04-11 07:48:46 |
|---|---|---|---|---|---|
huggingface/transformers | b46bd8b9d2ac991c0c04674957ebc0a65fb3f42b | ef177a5e1cdf0ca53e24e6d76e813198f7300dc4 | Fix error when streaming to gradio with non-string tool arguments (#32360)
Fix error when streaming agent run to gradio with non-string tool arguments | [
{
"path": "src/transformers/agents/monitoring.py",
"patch": "@@ -34,7 +34,7 @@ def pull_message(step_log: dict):\n yield ChatMessage(\n role=\"assistant\",\n metadata={\"title\": f\"🛠️ Used tool {step_log['tool_call']['tool_name']}\"},\n- content=content,\n+ ... | 2024-07-31T16:44:53 |
vercel/next.js | b0d64d647aae2d896740b28c59178cab6544deff | 9664c8d054956eb6da3bb61366f6f690d2d9f51d | docs: clarify that `matcher` needs to be constant values (#38597)
Ref: #38461, #37346
As [introduced](https://github.com/vercel/next.js/pull/37177/files#diff-7ec7f47987bd890b5b47f6cf101398cd22fa37f874348d70036f85309a040c92
) in #37177, we statically analyze the `matcher` values, so we should make it clear.
## Bug
... | [
{
"path": "docs/advanced-features/middleware.md",
"patch": "@@ -86,6 +86,8 @@ export const config = {\n }\n ```\n \n+> **Note:** The `matcher` values need to be constants so they can be statically analyzed at build-time. Dynamic values such as variables will be ignored.\n+\n ### Conditional Statements\n \n ... | 2022-07-13T14:36:10 |
golang/go | 4d048194cd0323e1deffce96e88e8a672a08732d | 5ba1c3f290ef8f3c373c560d3608325b07f44f7a | runtime: support new callbackasm1 calling convention on windows/arm
This updates the callbacks implementation on windows/arm for the
changes made in CL 258938. At the time, that was left as a TODO.
At the same time, it also extends the previous support for only 4
arguments to also support additional arguments on the ... | [
{
"path": "src/runtime/sys_windows_arm.s",
"patch": "@@ -314,48 +314,42 @@ TEXT runtime·externalthreadhandler(SB),NOSPLIT|NOFRAME,$0\n GLOBL runtime·cbctxts(SB), NOPTR, $4\n \n TEXT runtime·callbackasm1(SB),NOSPLIT|NOFRAME,$0\n-\t// TODO(austin): This needs to be converted to match changes\n-\t// in cgocall... | 2020-11-18T20:38:57 |
nodejs/node | f3cd53751ba3f917a0996a8f38c991242a8fbc76 | ef49f55e9376cdc2ef99f3e4cd3021d3a253a1c9 | src: refactor and harden `ProcessEmitWarning()`
- Handle exceptions when getting `process.emitWarning` or when
calling it properly
- Add `Maybe<bool>` to the return type, like the V8 API uses it
to indicate failure conditions
- Update call sites to account for that and clean up/return to JS
when encountering an ... | [
{
"path": "src/env.h",
"patch": "@@ -133,6 +133,7 @@ class ModuleWrap;\n V(dns_txt_string, \"TXT\") \\\n V(domain_string, \"domain\") \\\n V(emit_string, \"emit\") ... | 2017-12-01T23:04:56 |
huggingface/transformers | 5f1fcc299cb00c1edce5eb1efb8bacdde2365690 | b75ad56620431984a44a962c98136c8571b4fca9 | [Idefics2] - Fix FA2 call for Perceiver layer (#32275)
* Fix FA2 call for Perciever layer
* [run_slow] idefics2
* [run_slow] idefics2
* [run_slow] idefics2
* Fix up
* [run_slow] idefics2
* [run_slow] idefics2
* [run_slow] idefics2 | [
{
"path": "src/transformers/models/idefics2/modeling_idefics2.py",
"patch": "@@ -894,7 +894,7 @@ def forward(\n attention_mask,\n q_len,\n dropout=dropout_rate,\n- sliding_window=False,\n+ sliding_window=None,\n is_causal=self.is_causal,\... | 2024-07-31T13:51:04 |
golang/go | 5ba1c3f290ef8f3c373c560d3608325b07f44f7a | ff2824d4b3391c644b264273f98a47302e2d4ab6 | cmd/go/internal/modload: remove SetBuildList
For the last remaining call site (in cmd/go/internal/work, added for
the new 'go install pkg@version' codepath in CL 254365), use
EditBuildList instead.
SetBuildList assumes that the caller has enough information to produce
a complete, coherent build list. With lazy loadin... | [
{
"path": "src/cmd/go/internal/modload/buildlist.go",
"patch": "@@ -73,13 +73,6 @@ func Selected(path string) (version string) {\n \treturn \"\"\n }\n \n-// SetBuildList sets the module build list.\n-// The caller is responsible for ensuring that the list is valid.\n-// SetBuildList does not retain a refere... | 2020-11-18T04:16:30 |
vercel/next.js | 9664c8d054956eb6da3bb61366f6f690d2d9f51d | b8ed30648d38d5c98107bdbd1123c32d78537df7 | Add missing closing curly bracket in compiler option example (#38586)
This PR adds a missing closing curly bracket in the compiler option example of the [Styled Components](https://nextjs.org/docs/advanced-features/compiler#styled-components).
## Bug
- [ ] Related issues linked using `fixes #number`
- [ ] Integrati... | [
{
"path": "docs/advanced-features/compiler.md",
"patch": "@@ -66,6 +66,7 @@ module.exports = {\n transpileTemplateLiterals?: boolean,\n // Not supported yet.\n pure?: boolean,\n+ },\n },\n }\n ```",
"additions": 1,
"deletions": 0,
"language": "Markdown"
}
] | 2022-07-13T14:16:00 |
nodejs/node | dc925400d78a01216ae017f4e336eb114c41bca9 | 2c77cab4402ffcfc47367b14e0008258a70fd469 | doc: fix link for https api change
The PR number included for this api addition was originally incorrect.
PR-URL: https://github.com/nodejs/node/pull/17630
Refs: https://github.com/nodejs/node/pull/14903
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Colin... | [
{
"path": "doc/api/https.md",
"patch": "@@ -151,7 +151,7 @@ Global instance of [`https.Agent`][] for all HTTPS client requests.\n added: v0.3.6\n changes:\n - version: REPLACEME\n- pr-url: https://github.com/nodejs/node/pull/6569\n+ pr-url: https://github.com/nodejs/node/pull/14903\n description... | 2017-12-12T09:18:02 |
huggingface/transformers | b75ad56620431984a44a962c98136c8571b4fca9 | 7f552e28e0aca00ce60868c7620f7463eab60e14 | Llama 3.1: Fix incorrect `inv_freq` assignment (#32330)
fix 💩 | [
{
"path": "src/transformers/modeling_rope_utils.py",
"patch": "@@ -328,14 +328,14 @@ def _compute_llama3_parameters(\n wavelen = 2 * math.pi / inv_freq\n # wavelen < high_freq_wavelen: do nothing\n # wavelen > low_freq_wavelen: divide by factor\n- inv_freq_new = torch.where(wavelen > low_freq... | 2024-07-31T10:12:46 |
nodejs/node | 2c77cab4402ffcfc47367b14e0008258a70fd469 | 031d01bba5856ca9ebeb0646b902723b3cd4d9a3 | doc: note that randomBytes throws when passed null
Fixes: https://github.com/nodejs/node/issues/16778
PR-URL: https://github.com/nodejs/node/pull/17594
Fixes: https://github.com/nodejs/node/issues/16778
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Colin... | [
{
"path": "doc/api/crypto.md",
"patch": "@@ -1792,6 +1792,11 @@ be passed instead of a public key.\n ### crypto.randomBytes(size[, callback])\n <!-- YAML\n added: v0.5.8\n+changes:\n+ - version: v9.0.0\n+ pr-url: https://github.com/nodejs/node/pull/16454\n+ description: Passing `null` as the `callbac... | 2017-12-10T20:12:21 |
golang/go | 0bb6115dd6246c047335a75ce4b01a07c291befd | 96b943a483dca715ea0164644e1192052105881a | internal/fmtsort: sort the unsafe pointers in map
Currently storing keys that contain unsafe.
Pointer in a map could result inruntime panic when printing the map.
The root cause is that unsafe.Pointer is not comparable.
Fixes #42622.
Change-Id: Ie3bae7ee4945041843b66514de6227212a3da73e
GitHub-Last-Rev: d12d41302e611... | [
{
"path": "src/internal/fmtsort/sort.go",
"patch": "@@ -130,7 +130,7 @@ func compare(aVal, bVal reflect.Value) int {\n \t\tdefault:\n \t\t\treturn -1\n \t\t}\n-\tcase reflect.Ptr:\n+\tcase reflect.Ptr, reflect.UnsafePointer:\n \t\ta, b := aVal.Pointer(), bVal.Pointer()\n \t\tswitch {\n \t\tcase a < b:",
... | 2020-11-16T17:29:33 |
huggingface/transformers | 7f552e28e0aca00ce60868c7620f7463eab60e14 | a3264332cfb5ab8675ddb42740a75aeee1782a74 | Gemma2 and flash-attention (#32188)
* enable flash-attn & static cache
* this works, not the prev
* fix for sliding window layers
* not needed anymore | [
{
"path": "src/transformers/models/gemma2/modeling_gemma2.py",
"patch": "@@ -327,6 +327,11 @@ def forward(\n }\n key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)\n \n+ if attention_mask is not None:\n+ seq_len ... | 2024-07-31T05:33:38 |
ggml-org/llama.cpp | e4fed9d08de1d33ab51748880d38b62b9968dc2e | 5dd102539b97b3cc92c503063a2a244b3c7d57bd | ggml-webgpu: address quantization precision and backend lifecycle managment (#21521)
* ggml(webgpu): fix the busy-polls in Emscripten in the waitAny after #20618, and remove the busy webgpu log
* Merge with upstream
* Fix GET_ROWS packed integer NaN when using f16 as memory buffer in shader quants
* Update Unary w... | [
{
"path": "ggml/src/ggml-webgpu/ggml-webgpu-shader-lib.hpp",
"patch": "@@ -1115,6 +1115,32 @@ class ggml_webgpu_shader_lib {\n std::string type_upper = type_str;\n std::transform(type_upper.begin(), type_upper.end(), type_upper.begin(), ::toupper);\n \n+ ... | 2026-04-10T17:52:01 |
ollama/ollama | 33a65e3ba3ad5666d6ba8430efbccfa6d642d1de | e70610ef0680184f1947b0ed23f9e174ebcb8eb6 | error | [
{
"path": "llm/server.go",
"patch": "@@ -560,6 +560,9 @@ func (s *llmServer) WaitUntilRunning(ctx context.Context) error {\n \t\t\tif s.status != nil && s.status.LastErrMsg != \"\" {\n \t\t\t\tmsg = s.status.LastErrMsg\n \t\t\t}\n+\t\t\tif strings.Contains(msg, \"unknown model\") {\n+\t\t\t\treturn fmt.Erro... | 2024-07-01T23:04:13 |
rust-lang/rust | 7e8494f0a511d9374d96fb741efebb3ea71957fd | 0a679514d41443329c183110a4142e6813bc4307 | Don't return an error from get_or_default_sysroot
All callers unwrap the result. | [
{
"path": "compiler/rustc_codegen_ssa/src/back/link.rs",
"patch": "@@ -1286,8 +1286,7 @@ fn link_sanitizer_runtime(\n if path.exists() {\n sess.target_tlib_path.dir.clone()\n } else {\n- let default_sysroot =\n- filesearch::get_or_default_sysroot().expec... | 2025-03-07T16:09:20 |
huggingface/transformers | a3264332cfb5ab8675ddb42740a75aeee1782a74 | 6e2d04e429dc4ce240c99bd14b7b84550b79fd73 | LLaVA-NeXT: fix anyres shapes (#32314)
fix | [
{
"path": "src/transformers/models/llava_next/modeling_llava_next.py",
"patch": "@@ -660,7 +660,7 @@ def pack_image_features(self, image_features, image_sizes, image_newline=None):\n height = width = self.config.vision_config.image_size // self.config.vision_config.patch_size\n ... | 2024-07-31T05:01:12 |
nodejs/node | 031d01bba5856ca9ebeb0646b902723b3cd4d9a3 | 3b9803838cf60d74d1df053b2bd6affcceeef009 | doc: clearify promisify behavior for bad arguments
Currently the documentation states that promisify() will result in
undefined behavior if bad arguments are passed. This is not
necessarily the case, since the behavior is well defined, but just
not useful.
PR-URL: https://github.com/nodejs/node/pull/17593
Fixes: http... | [
{
"path": "doc/api/util.md",
"patch": "@@ -553,8 +553,10 @@ If there is an `original[util.promisify.custom]` property present, `promisify`\n will return its value, see [Custom promisified functions][].\n \n `promisify()` assumes that `original` is a function taking a callback as its\n-final argument in all ... | 2017-12-10T18:50:19 |
golang/go | 96b943a483dca715ea0164644e1192052105881a | 35693d037f9d1c30d6de1fafd08e8c923a415ab8 | go/types: report an error for invalid constant values
The parser reports syntactic errors in constant literals.
The go/constant package produces an "unknown" value for
syntactically correct numeric constants that are too small
or too large. Check for the unknown value and report an
error rather than silently continuin... | [
{
"path": "src/go/types/expr.go",
"patch": "@@ -1073,7 +1073,11 @@ func (check *Checker) exprInternal(x *operand, e ast.Expr, hint Type) exprKind {\n \tcase *ast.BasicLit:\n \t\tx.setConst(e.Kind, e.Value)\n \t\tif x.mode == invalid {\n-\t\t\tcheck.invalidAST(e, \"invalid literal %v\", e.Value)\n+\t\t\t// T... | 2020-11-19T01:05:02 |
ggml-org/llama.cpp | fb38d6f278a4a391b8db1a16ee5946db3766459c | 0893f50f2dc14fcc046e10d4f76a1ac7a62c0490 | common : fix when loading a cached HF models with unavailable API (#21670)
Signed-off-by: Adrien Gallouët <angt@huggingface.co> | [
{
"path": "common/download.cpp",
"patch": "@@ -283,6 +283,13 @@ static int common_download_file_single_online(const std::string & url,\n static const int max_attempts = 3;\n static const int retry_delay_seconds = 2;\n \n+ const bool file_exists = std::filesystem::exists(path);\n+\n+... | 2026-04-10T14:37:46 |
rust-lang/rust | 1066ee09ed106a19cd902dfb1d4d44fdf472cb56 | 714c64c7a140bc5c7accbf16770d11ddf7d9e742 | fix: `redundant_clone` FP on enum cast | [
{
"path": "clippy_utils/src/mir/mod.rs",
"patch": "@@ -76,7 +76,7 @@ impl<'tcx> Visitor<'tcx> for V<'_> {\n }\n if matches!(\n ctx,\n- PlaceContext::NonMutatingUse(NonMutatingUseContext::Move)\n+ PlaceContext::NonMutat... | 2025-03-12T13:55:31 |
huggingface/transformers | 6e2d04e429dc4ce240c99bd14b7b84550b79fd73 | 026a173a64372e9602a16523b8fae9de4b0ff428 | Fix slow GemmaTokenizer and improve SPM slow -> fast conversion process (#32191)
* Remove user-defined tokens which can be obtained through merges
* Remove debug line
* formatting
* Refactor spm slow -> fast converter
* revert unnecessary refactor
* set comprehension
* remove test files
* Use `voc... | [
{
"path": "src/transformers/convert_slow_tokenizer.py",
"patch": "@@ -53,6 +53,25 @@ def _get_prepend_scheme(add_prefix_space: bool, original_tokenizer) -> str:\n return prepend_scheme\n \n \n+def generate_merges(vocab, vocab_scores):\n+ reverse = vocab_scores is not None\n+ vocab_scores = dict(vo... | 2024-07-30T21:36:38 |
nodejs/node | 9236dfe1ef3becfc02a18770ec5e74bf3b01bd99 | 89b374623f969897a2e2499f646fa69c2fc1119c | src: make url host a proper C++ class
- Gives `URLHost` a proper destructor that clears memory
depending on the type of the host (This fixes a memory leak)
- Hide the host type enums and class layout as implementation details
- Make the `Parse` methods members of `URLHost`
- Turn `WriteHost` into a `ToString()` meth... | [
{
"path": "src/node_url.cc",
"patch": "@@ -55,27 +55,71 @@ const char kEOL = -1;\n const char16_t kUnicodeReplacementCharacter = 0xFFFD;\n \n // https://url.spec.whatwg.org/#concept-host\n-union url_host_value {\n- std::string domain;\n- uint32_t ipv4;\n- uint16_t ipv6[8];\n- std::string opaque;\n- ~ur... | 2017-12-05T13:23:38 |
golang/go | 35693d037f9d1c30d6de1fafd08e8c923a415ab8 | 5b0ec1a6ac0e644c89940e0fe5f79863ad2eafaa | cmd/compile: fix miscompilation during inlining
When inlining a function call expression, it's possible that the
function callee subexpression has side effects that need to be
preserved. This used to not be an issue, because inlining wouldn't
recognize these as inlinable anyway. But golang.org/cl/266199 extended
the i... | [
{
"path": "src/cmd/compile/internal/gc/inl.go",
"patch": "@@ -963,6 +963,21 @@ func mkinlcall(n, fn *Node, maxCost int32, inlMap map[*Node]bool) *Node {\n \n \tninit := n.Ninit\n \n+\t// For normal function calls, the function callee expression\n+\t// may contain side effects (e.g., added by addinit during\... | 2020-11-18T20:50:46 |
ggml-org/llama.cpp | d7ff074c87ecacd57d5760e2f678866ba9fe7149 | 3f8752b5592fdef6fe58fb23a8d468f603095154 | common : enable reasoning budget sampler for gemma4 (#21697)
* fix: enable reasoning budget sampler for gemma4
Add thinking_start_tag and thinking_end_tag to
common_chat_params_init_gemma4(). Without these, the reasoning
budget sampler never activates for gemma4.
Make the newline after "thought" optional in the PEG ... | [
{
"path": "common/chat.cpp",
"patch": "@@ -1083,7 +1083,9 @@ static common_chat_params common_chat_params_init_gemma4(const common_chat_templ\n \n data.prompt = common_chat_template_direct_apply_impl(tmpl, inputs);\n data.format = COMMON_CHAT_FORMAT_PEG_GEMMA4;\n- data.suppo... | 2026-04-10T09:49:14 |
rust-lang/rust | eea0db2590a79b58d1354c86954382e2952f7749 | d2b328e6b3a4cd42b85a289ce3285ca0ac461e7c | Fix rustc test suite | [
{
"path": "scripts/test_rustc_tests.sh",
"patch": "@@ -166,6 +166,18 @@ index 30387af428c..f7895b12961 100644\n \"/build-root\",\n )\n .normalize(r#\"\"[^\"]*\\/symbols.o\"\"#, \"\\\\\"/symbols.o\\\\\"\")\n+diff --git a/src/tools/compiletest/src/runtest/run_make.rs... | 2025-03-12T13:18:52 |
ollama/ollama | cff3f44f4a4097de864d70d9a95f31c62e8ecdfa | 3518aaef3318b47c63d3df9ef3bdd96dff7541ae | Fix case for NumCtx | [
{
"path": "server/sched.go",
"patch": "@@ -23,7 +23,7 @@ type LlmRequest struct {\n \tctx context.Context //nolint:containedctx\n \tmodel *Model\n \topts api.Options\n-\torigNumCTX int // Track the initial ctx request\n+\torigNumCtx int // Track the initial ctx req... | 2024-07-01T16:43:59 |
ggml-org/llama.cpp | 3f8752b5592fdef6fe58fb23a8d468f603095154 | 7b69125331d7a69c6ea6349f33f506247bc66127 | docs : fix broken link to ggml-openvino in OPENVINO.md (#21709) | [
{
"path": "docs/backend/OPENVINO.md",
"patch": "@@ -3,7 +3,7 @@\n > [!NOTE]\n > Performance and memory optimizations, accuracy validation, broader quantization coverage, broader operator and model support are work in progress.\n \n-[OpenVINO](https://docs.openvino.ai/) is an open-source toolkit for optimizi... | 2026-04-10T07:50:08 |
nodejs/node | 89b374623f969897a2e2499f646fa69c2fc1119c | 2050bab09c7b3917caa92133299da0f7c07bd38d | src: move url internals into anonymous namespace
This helps because `static` doesn’t work for C++ classes,
but refactoring `url_host` into a proper C++ class seems the
most reasonable soluation for the memory leak fixed by the next commit.
PR-URL: https://github.com/nodejs/node/pull/17470
Reviewed-By: Timothy Gu <tim... | [
{
"path": "src/node_url.cc",
"patch": "@@ -46,11 +46,13 @@ using v8::Value;\n \n namespace url {\n \n+namespace {\n+\n // https://url.spec.whatwg.org/#eof-code-point\n-static const char kEOL = -1;\n+const char kEOL = -1;\n \n // Used in ToUSVString().\n-static const char16_t kUnicodeReplacementCharacter = 0... | 2017-12-05T12:37:38 |
golang/go | 5b0ec1a6ac0e644c89940e0fe5f79863ad2eafaa | b4f3d52f6a90aa520799f836e5951d5cf65f7fe4 | cmd/compile: fix panic in field tracking logic
Within the frontend, we generally don't guarantee uniqueness of
anonymous types. For example, each struct type literal gets
represented by its own types.Type instance.
However, the field tracking code was using the struct type as a map
key. This broke in golang.org/cl/25... | [
{
"path": "src/cmd/compile/internal/gc/main.go",
"patch": "@@ -89,6 +89,7 @@ var debugtab = []struct {\n \t{\"dwarfinl\", \"print information about DWARF inlined function creation\", &Debug_gendwarfinl},\n \t{\"softfloat\", \"force compiler to emit soft-float code\", &Debug_softfloat},\n \t{\"defer\", \"pri... | 2020-11-18T20:08:59 |
rust-lang/rust | 7bfe2136e4ffa5b5f6c85f0544a5d7ef8ab2b705 | 57a4736e9f4b7e8089b2db60583607f3b550c862 | Delegation: one more ICE fix for `MethodCall` generation | [
{
"path": "compiler/rustc_ast_lowering/src/delegation.rs",
"patch": "@@ -330,6 +330,7 @@ impl<'hir> LoweringContext<'_, 'hir> {\n .unwrap_or_default()\n && delegation.qself.is_none()\n && !has_generic_args\n+ && !args.is_empty()\n {\n let as... | 2025-03-12T12:59:21 |
ollama/ollama | cb42e607c5cf4d439ad4d5a93ed13c7d6a09fc34 | 2aa91a937ba199ae5832c71ecc10221cc6420fa8 | llm: speed up gguf decoding by a lot (#5246)
Previously, some costly things were causing the loading of GGUF files
and their metadata and tensor information to be VERY slow:
* Too many allocations when decoding strings
* Hitting disk for each read of each key and value, resulting in a
not-okay amount of ... | [
{
"path": "llm/ggla.go",
"patch": "@@ -53,7 +53,7 @@ func (llm *ggla) Tensors() Tensors {\n \treturn llm.tensors\n }\n \n-func (llm *ggla) decode(rs io.ReadSeeker) error {\n+func (llm *ggla) decode(rs io.ReadSeeker) (retErr error) {\n \tvar r uint32\n \tif err := binary.Read(rs, binary.LittleEndian, &r); er... | 2024-06-25T04:47:52 |
nodejs/node | 727339e9c276748d7b4efa066b72e05a1d60a4b9 | 88a9e2df859c9de8a3b2d0cca9862ee00478c863 | repl: fix util.inspect() coloring regression
The `Object.assign()` calls introduced in commit 90a4390 ("repl: show
proxies as Proxy objects") mutated their first argument, causing the
`{ colors: true }` option from the REPL to leak over into the global
`util.inspect.defaultOptions` object.
Refs: https://github.com/no... | [
{
"path": "lib/repl.js",
"patch": "@@ -103,7 +103,7 @@ function hasOwnProperty(obj, prop) {\n // This is the default \"writer\" value if none is passed in the REPL options.\n const writer = exports.writer = (obj) => util.inspect(obj, writer.options);\n writer.options =\n- Object.assign(util.inspect.defau... | 2017-12-09T11:43:40 |
golang/go | b63db7f72446753de0f5bb78b629dbe58fb15cda | ae76f6e96216f352cc5021a4c8a7d879c4cb6873 | runtime: give test child time to block
The child in TestPanicSystemstack prints "x\n" and then blocks on a
lock. Receiving those bytes only indicates that the child is _about to
block_. Since we don't have a way to know when it is fully blocked,
sleep a bit to give it time to block. This makes us less likely to lose
t... | [
{
"path": "src/runtime/crash_unix_test.go",
"patch": "@@ -232,13 +232,20 @@ func TestPanicSystemstack(t *testing.T) {\n \t}\n \tdefer pr.Close()\n \n-\t// Wait for \"x\\nx\\n\" to indicate readiness.\n+\t// Wait for \"x\\nx\\n\" to indicate almost-readiness.\n \tbuf := make([]byte, 4)\n \t_, err = io.ReadFu... | 2020-11-09T21:37:05 |
ggml-org/llama.cpp | d6f3030047f85a98b009189e76f441fe818ea44d | 009a1133268d040a7c574a7b9c95413b0be369a9 | ggml: backend-agnostic tensor parallelism (experimental) (#19378)
* ggml: backend-agnostic tensor parallelism
* support for GPT-OSS, Qwen 3 MoE
* partial Vulkan fix
* add support for 4/8 GPUs
* unconditional peer access
* re-use buffers + ggml contexts
* fix output pattern
* NCCL support
* GGML: HIP: add RCCL ... | [
{
"path": "common/arg.cpp",
"patch": "@@ -2348,19 +2348,21 @@ common_params_context common_params_parser_init(common_params & params, llama_ex\n }\n ).set_env(\"LLAMA_ARG_N_GPU_LAYERS\"));\n add_opt(common_arg(\n- {\"-sm\", \"--split-mode\"}, \"{none,layer,row}\",\n+ {\"-sm\", ... | 2026-04-09T14:42:19 |
huggingface/transformers | 516af4bb63538edc448f814e3690dd5171c4f311 | 62c60a30181a65e1a3a7f19c3055a240a6a21335 | fixes #32329 : The Torch code is correct - to get an average of 10% o… (#32335)
fixes #32329 : The Torch code is correct - to get an average of 10% of the total, we want to take 50% of the remainder after we've already masked 80% with [MASK] in the previous step. | [
{
"path": "src/transformers/data/data_collator.py",
"patch": "@@ -751,7 +751,7 @@ def tf_mask_tokens(\n inputs = tf.where(indices_replaced, mask_token_id, inputs)\n \n # 10% of the time, we replace masked input tokens with random word\n- indices_random = self.tf_bernoulli(input_shape,... | 2024-07-30T17:21:45 |
nodejs/node | 88a9e2df859c9de8a3b2d0cca9862ee00478c863 | 93656f4366bcb6084cae63aeb97a63d3634bac2b | src: fix inspector nullptr deref on abrupt exit
Fix a nullptr dereference on abrupt termination when async call stack
recording is enabled.
Bug discovered while trying to write a regression test for the bug fix
in commit df79b7d821 ("src: fix missing handlescope bug in inspector".)
PR-URL: https://github.com/nodejs/... | [
{
"path": "src/inspector_agent.cc",
"patch": "@@ -325,10 +325,12 @@ class NodeInspectorClient : public V8InspectorClient {\n }\n \n void maxAsyncCallStackDepthChanged(int depth) override {\n- if (depth == 0) {\n- env_->inspector_agent()->DisableAsyncHook();\n- } else {\n- env_->inspector... | 2017-12-08T14:33:41 |
ggml-org/llama.cpp | 009a1133268d040a7c574a7b9c95413b0be369a9 | c8ac02fa1b9f8a154e110b655f7f7e4907796b0c | ggml : check return value of CUB calls used in argsort and top-k (they all return cudaError_t) (#21676)
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com> | [
{
"path": "ggml/src/ggml-cuda/argsort.cu",
"patch": "@@ -60,24 +60,24 @@ void argsort_f32_i32_cuda_cub(ggml_cuda_pool & pool,\n \n if (order == GGML_SORT_ORDER_ASC) {\n if (nrows == 1) {\n- DeviceRadixSort::SortPairs(nullptr, temp_storage_bytes, temp_keys, temp_keys, // keys (in-plac... | 2026-04-09T13:17:11 |
golang/go | ae76f6e96216f352cc5021a4c8a7d879c4cb6873 | ee1b51294ab50179b60d6c548ddded5b91cd0cbb | runtime: use clock_gettime instead of gettimeofday on darwin
clock_gettime has higher resolution than gettimeofday and is available
since macOS 10.12. Go 1.15 already requires at least macOS 10.12 and
thus clock_gettime can be used unconditionally (also see
https://golang.org/doc/go1.15#darwin)
Fixes #25633
Change-I... | [
{
"path": "src/runtime/sys_darwin.go",
"patch": "@@ -303,9 +303,9 @@ func nanotime_trampoline()\n //go:nosplit\n //go:cgo_unsafe_args\n func walltime1() (int64, int32) {\n-\tvar t timeval\n+\tvar t timespec\n \tlibcCall(unsafe.Pointer(funcPC(walltime_trampoline)), unsafe.Pointer(&t))\n-\treturn int64(t.tv_s... | 2020-11-18T10:08:43 |
huggingface/transformers | 62c60a30181a65e1a3a7f19c3055a240a6a21335 | 16271080333ad52be5349fb31d789fb232b68760 | fixes to properly shard FSDP across cpu and meta for cpu_efficient_loading for prequantized 4bit (#32276) | [
{
"path": "src/transformers/modeling_utils.py",
"patch": "@@ -932,6 +932,8 @@ def _load_state_dict_into_meta_model(\n )\n )\n ):\n+ if is_fsdp_enabled():\n+ param_device = \"cpu\" if is_local_dist_rank_0() else \"meta\"\n # For backwa... | 2024-07-30T16:55:59 |
ollama/ollama | 9929751cc8b415e7b83d5151742dad734e8b5efc | 17b7186cd759337fa98b626e82de150f3789b040 | Disable concurrency for AMD + Windows
Until ROCm v6.2 ships, we wont be able to get accurate free memory
reporting on windows, which makes automatic concurrency too risky.
Users can still opt-in but will need to pay attention to model sizes otherwise they may thrash/page VRAM or cause OOM crashes.
All other platforms ... | [
{
"path": "envconfig/config.go",
"patch": "@@ -85,13 +85,13 @@ func AsMap() map[string]EnvVar {\n \t\t\"OLLAMA_HOST\": {\"OLLAMA_HOST\", Host, \"IP Address for the ollama server (default 127.0.0.1:11434)\"},\n \t\t\"OLLAMA_KEEP_ALIVE\": {\"OLLAMA_KEEP_ALIVE\", KeepAlive, \"The duration t... | 2024-06-19T20:35:38 |
ggml-org/llama.cpp | c8ac02fa1b9f8a154e110b655f7f7e4907796b0c | 4ef9301e4de5ea8f5ebf20b107cf30de0070bf2a | requirements : update transformers to 5.5.1 (#21617)
* requirements : update transformers to 5.5.0
This commit updates the transformers dependency to version 5.5.0.
The motivation for this is that transformers 5.5.0 includes support for
Gemma4 and is required to be able to convert Gemma4 models. This is also
causing... | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -1229,15 +1229,15 @@ def get_vocab_base(self) -> tuple[list[str], list[int], str]:\n \n from transformers import AutoTokenizer\n tokenizer = AutoTokenizer.from_pretrained(self.dir_model)\n- vocab_size = self.hparams.get(\"vocab_size\", l... | 2026-04-09T10:36:29 |
rust-lang/rust | e4d3b0a0724adc2f8e6e1bfdacc1d58eb9926202 | cfc743226e86e9a772fa9a7126364ec1bd6aab89 | Update `produce_ice` so its error can be checked in ui tests | [
{
"path": "clippy_lints/src/utils/internal_lints/produce_ice.rs",
"patch": "@@ -1,6 +1,6 @@\n use rustc_ast::ast::NodeId;\n use rustc_ast::visit::FnKind;\n-use rustc_lint::{EarlyContext, EarlyLintPass};\n+use rustc_lint::{EarlyContext, EarlyLintPass, LintContext};\n use rustc_session::declare_lint_pass;\n u... | 2025-03-12T10:30:46 |
nodejs/node | 8d5b3de8cb3467ad9f29232f0ecbbd2fd6f99438 | e17dba7a4553081a4ae84e4f27bf89e589e27fed | test: simplify common.expectsError
The mustCall is actually only necessary in case it is used as
callback. Otherwise it works as a must call on its own.
PR-URL: https://github.com/nodejs/node/pull/17616
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Jon... | [
{
"path": "test/common/index.js",
"patch": "@@ -735,7 +735,7 @@ exports.expectsError = function expectsError(fn, settings, exact) {\n settings = fn;\n fn = undefined;\n }\n- const innerFn = exports.mustCall(function(error) {\n+ function innerFn(error) {\n assert.strictEqual(error.code, setti... | 2017-12-11T08:34:57 |
golang/go | ee1b51294ab50179b60d6c548ddded5b91cd0cbb | d3072b838366e7cc6b9af9767d3648f6e62bae8b | runtime: use pipe2 syscall for Pipe in tests
On FreeBSD >= 11 with a kernel built with COMPAT_FREEBSD11 but not
COMPAT_FREEBSD10, the pipe syscall is not available. Thus, tests using
runtime.pipe fail with ENOSYS. As suggested by Ian, fix this by calling
pipe2(0) in these tests and fall back to pipe() in case of ENOSY... | [
{
"path": "src/runtime/export_pipe2_test.go",
"patch": "@@ -0,0 +1,15 @@\n+// Copyright 2020 The Go Authors. All rights reserved.\n+// Use of this source code is governed by a BSD-style\n+// license that can be found in the LICENSE file.\n+\n+// +build freebsd linux netbsd openbsd solaris\n+\n+package runti... | 2020-11-18T07:53:00 |
huggingface/transformers | 16271080333ad52be5349fb31d789fb232b68760 | bd54ed2ed7f578e4122f3e6d536fbe3c9bc76de1 | fix: Added missing raise keyword for few exceptions (#32333)
Fixed raising of few exceptions. | [
{
"path": "examples/research_projects/movement-pruning/masked_run_glue.py",
"patch": "@@ -98,7 +98,7 @@ def regularization(model: nn.Module, mode: str):\n elif mode == \"l0\":\n regu += torch.sigmoid(param - 2 / 3 * np.log(0.1 / 1.1)).sum() / param.numel()\n else:\n- ... | 2024-07-30T16:53:03 |
ggml-org/llama.cpp | ddf03c6d9a347c3b7b7577a643d23aaa8d03a7c8 | 26229755c50543af8b95f8334c4d217b4b41117d | common : fix ambiguous grammar rule in gemma4 (#21661)
* common : fix ambiguous grammar rule in gemma4
* cont : fix missing comma... | [
{
"path": "common/chat.cpp",
"patch": "@@ -1124,7 +1124,7 @@ static common_chat_params common_chat_params_init_gemma4(const common_chat_templ\n p.rule(\"gemma4-bool\", p.json_bool());\n p.rule(\"gemma4-null\", p.json_null());\n p.rule(\"gemma4-number\", p.json_number());\... | 2026-04-09T10:25:07 |
rust-lang/rust | cb260791a43efb40abf22e7413249cbdf3615d8c | d2b52c5c48ea024fc277feddcc92a60cc92be13c | fix incorrect tracing log
Previous information is not correct.
Signed-off-by: onur-ozkan <work@onurozkan.dev> | [
{
"path": "src/bootstrap/src/core/build_steps/compile.rs",
"patch": "@@ -194,11 +194,7 @@ impl Step for Std {\n trace!(?compiler_to_use);\n \n if compiler_to_use != compiler {\n- trace!(\n- ?compiler_to_use,\n- ?compiler,\n- \"compiler ... | 2025-03-12T07:29:13 |
nodejs/node | e17dba7a4553081a4ae84e4f27bf89e589e27fed | b961d9fd83c963657c2305ed13ff447573eac852 | net: remove Socket.prototype.listen
The function was never documented and now throws a TypeError if used.
PR-URL: https://github.com/nodejs/node/pull/13735
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com> | [
{
"path": "lib/net.js",
"patch": "@@ -380,15 +380,6 @@ Socket.prototype.read = function(n) {\n return this.read(n);\n };\n \n-\n-// FIXME(joyeecheung): this method is neither documented nor tested\n-Socket.prototype.listen = function() {\n- debug('socket.listen');\n- this.on('connection', arguments[0]);... | 2017-12-06T16:48:37 |
golang/go | d3072b838366e7cc6b9af9767d3648f6e62bae8b | b194b5151fdac0c33da0b7359c97f92781ab3b5e | cmd/go: in 'go get', only load retractions for resolved versions
Previously, 'go get' loaded retractions for every module in the build
list, which took a long time and usually wasn't helpful.
This rolls forward CL 269019, which was reverted in CL 270521. The new
revision adds a call to modload.ListModules at the end ... | [
{
"path": "src/cmd/go/internal/modget/get.go",
"patch": "@@ -419,20 +419,7 @@ func runGet(ctx context.Context, cmd *base.Command, args []string) {\n \t\t\tpkgPatterns = append(pkgPatterns, q.pattern)\n \t\t}\n \t}\n-\tif len(pkgPatterns) > 0 {\n-\t\t// We skipped over missing-package errors earlier: we want... | 2020-11-10T20:48:37 |
huggingface/transformers | bd54ed2ed7f578e4122f3e6d536fbe3c9bc76de1 | e68ec18ce224af879f22d904c7505a765fb77de3 | Alternative agent plan (#32295)
* new agent plan
* plan type assertion
* style corrections
* better prompt naming
* make fixup | [
{
"path": "src/transformers/agents/agents.py",
"patch": "@@ -17,7 +17,7 @@\n import json\n import logging\n import re\n-from typing import Any, Callable, Dict, List, Optional, Tuple, Union\n+from typing import Any, Callable, Dict, List, Literal, Optional, Tuple, Union\n \n from .. import is_torch_available\... | 2024-07-30T16:48:18 |
ggml-org/llama.cpp | 26229755c50543af8b95f8334c4d217b4b41117d | 057dba336e60103be2b2e95714ccab5fcb3bae6d | common : simplify autoparser tagged parser rules (#21216)
* common : simplify autoparser tagged parser rules
* cont : remove upper limit on optional args
* cont : revert changes to parsing at the end
* cont : undo arbitrary ordering of optional args
* cont : fix uninitialized required parameters
* revert to simpl... | [
{
"path": "common/chat-auto-parser-generator.cpp",
"patch": "@@ -332,58 +332,36 @@ common_peg_parser analyze_tools::build_tool_parser_tag_tagged(parser_build_conte\n const auto & inputs = ctx.inputs;\n bool force_tools = inputs.tool_choice == COMMON_CHAT_TOOL_CHOICE_REQUIRED;\n \n+ a... | 2026-04-09T10:24:20 |
nodejs/node | 800ce94e5cdaa8dba7cf69b4c560a66a51ae0ec8 | 3674bee884bd86a7eb778345a410b2d8ab473a84 | tools,test: throw if common.PORT used in parallel tests
common.PORT should not be used in parallelized tests. (There can be a
port collision if another tests requests an arbitrary open port from the
operating system and ends up getting common.PORT before a test that uses
common.PORT uses the port.) In such a situation... | [
{
"path": "test/common/index.js",
"patch": "@@ -40,7 +40,16 @@ const noop = () => {};\n // gets tools to ignore it by default or by simple rules, especially eslint.\n let tmpDirName = '.tmp';\n \n-exports.PORT = +process.env.NODE_COMMON_PORT || 12346;\n+Object.defineProperty(exports, 'PORT', {\n+ get: () =... | 2017-12-09T02:33:36 |
ggml-org/llama.cpp | 057dba336e60103be2b2e95714ccab5fcb3bae6d | 501aeed18f54716eacc7bd6397376b7829602084 | model: fix multimodal padding token for gemma3n/gemma4 (#21625)
* model: fix multimodal padding token for gemma3n/gemma4
* nits | [
{
"path": "src/models/gemma3n-iswa.cpp",
"patch": "@@ -250,27 +250,29 @@ ggml_tensor * llm_build_gemma3n_iswa::calc_magnitude(ggml_tensor * x) {\n ggml_tensor * llm_build_gemma3n_iswa::build_inp_per_layer() {\n auto inp = std::make_unique<llm_graph_input_embd>(n_embd);\n ggml_tensor * inp_per_layer;... | 2026-04-09T10:18:23 |
golang/go | 64ef84881f607c2d3a0c35762853af8f3bc4ac26 | 399b5d14d4114b326096127df9a89342eb1fc2ad | cmd/go: fix retract interval syntax in 'go help mod edit'
For #24031
Change-Id: I70461431aac24c9465b9bdab082bcc34343a53a8
Reviewed-on: https://go-review.googlesource.com/c/go/+/270557
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com... | [
{
"path": "src/cmd/go/alldocs.go",
"patch": "@@ -1155,7 +1155,7 @@\n //\n // The -retract=version and -dropretract=version flags add and drop a\n // retraction on the given version. The version may be a single version\n-// like \"v1.2.3\" or a closed interval like \"[v1.1.0-v1.1.9]\". Note that\n+// like \"... | 2020-11-16T23:31:22 |
huggingface/transformers | e68ec18ce224af879f22d904c7505a765fb77de3 | 2fbbcf5007509c66b02924ce6dcff66f58e7f58c | Docs: formatting nits (#32247)
* doc formatting nits
* ignore non-autodocs
* Apply suggestions from code review
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* Update src/transformers/models/esm/modeling_esm.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply... | [
{
"path": "docs/source/en/model_sharing.md",
"patch": "@@ -77,7 +77,7 @@ Then use `notebook_login` to sign-in to the Hub, and follow the link [here](http\n \n To ensure your model can be used by someone working with a different framework, we recommend you convert and upload your model with both PyTorch and ... | 2024-07-30T14:49:14 |
nodejs/node | ec6c063279b4cd4d85dcc0ac7b223c10982b6e77 | 73ad3f9bea3993b486621aaf9e61484dc37741d4 | doc: fix modules.md export example
Arrow functions cannot be called with the new keyword,
convert to ES6 classes instead.
PR-URL: https://github.com/nodejs/node/pull/17579
Refs: https://github.com/nodejs/node/pull/17364
Reviewed-By: Rich Trott <rtrott@gmail.com> | [
{
"path": "doc/api/modules.md",
"patch": "@@ -38,7 +38,7 @@ In this example, the variable `PI` is private to `circle.js`.\n The `module.exports` property can be assigned a new value (such as a function\n or object).\n \n-Below, `bar.js` makes use of the `square` module, which exports a constructor:\n+Below,... | 2017-12-09T18:27:31 |
ggml-org/llama.cpp | 501aeed18f54716eacc7bd6397376b7829602084 | 0ec191e1d761a4daec873dfd73278ee356944fe5 | mtmd: support dots.ocr (#17575)
* convert gguf
* clip impl
* fix conversion
* wip
* corrections
* update docs
* add gguf to test script | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -3777,7 +3777,14 @@ def set_vocab(self):\n self._set_vocab_qwen()\n \n \n-@ModelBase.register(\"Qwen2Model\", \"Qwen2ForCausalLM\", \"Qwen2AudioForConditionalGeneration\", \"KORMoForCausalLM\", \"AudioFlamingo3ForConditionalGeneration\")\n+@ModelBase.r... | 2026-04-09T10:16:38 |
rust-lang/rust | bd385f3064f558d4bba19a0447f8b08208915dc9 | 9cab8c25dc3c7c1cac7475568161b3b4fc34a88f | Fix panic handler for cygwin | [
{
"path": "library/std/src/sys/personality/gcc.rs",
"patch": "@@ -248,7 +248,10 @@ cfg_if::cfg_if! {\n }\n \n cfg_if::cfg_if! {\n- if #[cfg(all(windows, any(target_arch = \"aarch64\", target_arch = \"x86_64\"), target_env = \"gnu\"))] {\n+ if #[cfg(any(\n+ ... | 2025-03-12T07:48:05 |
huggingface/transformers | 2fbbcf5007509c66b02924ce6dcff66f58e7f58c | 084b5094eb490319719cc11cb05b751e0b419d49 | Fix M4T for ASR pipeline (#32296)
* tentative fix
* do the same for M4T | [
{
"path": "src/transformers/models/seamless_m4t/modeling_seamless_m4t.py",
"patch": "@@ -3154,6 +3154,7 @@ def generate(\n \"\"\"\n text_decoder_input_ids = kwargs.pop(\"decoder_input_ids\", None)\n # overwrite text_decoder_input_ids if tgt_lang is passed. The latter gets priority ov... | 2024-07-30T14:00:13 |
golang/go | a14e7bf6d42d9a8b0d698c0a47422c12e38b3f6c | bcfaeca58c791ada53fae18fffa26936bc245423 | go/ast: document invalid comment end positions with CRLF line endings
We've decided that issues like golang/go#41197 are unfixable, so
instead, document the bug.
Fixes golang/go#41197
Change-Id: I5649027f6e2445eec765516f2f642db0d601ea20
Reviewed-on: https://go-review.googlesource.com/c/go/+/270938
Trust: Rebecca Sta... | [
{
"path": "src/go/ast/ast.go",
"patch": "@@ -57,6 +57,11 @@ type Decl interface {\n // Comments\n \n // A Comment node represents a single //-style or /*-style comment.\n+//\n+// The Text field contains the comment text without carriage returns (\\r) that\n+// may have been present in the source. Because a ... | 2020-11-17T23:28:38 |
ollama/ollama | e835ef183691db1cc7da30cfc61fb4b96b321e80 | c7c2f3bc228714988c650d4c238a08510f58e8cb | fix: quantization with template | [
{
"path": "server/images.go",
"patch": "@@ -414,17 +414,22 @@ func CreateModel(ctx context.Context, name model.Name, modelFileDir, quantizatio\n \t\t\t\t\t\t\treturn err\n \t\t\t\t\t\t}\n \n-\t\t\t\t\t\tlayers, err := parseFromFile(ctx, temp, \"\", fn)\n+\t\t\t\t\t\tlayer, err := NewLayer(temp, baseLayer.Me... | 2024-06-21T20:30:43 |
ggml-org/llama.cpp | 0ec191e1d761a4daec873dfd73278ee356944fe5 | 243532e5568704ced8b08136e70247f9ceecdbf7 | vocab: add gemma4 tokenizer tests, fix edge case (#21534)
* YATF (Yet Another Tokenizer Fix) for Gemma 4. With tests!
* Remove unnecessary hash from update script.
* minor: move constant | [
{
"path": "models/ggml-vocab-gemma-4.gguf.inp",
"patch": "@@ -0,0 +1,111 @@\n+ied 4 ½ months\n+__ggml_vocab_test__\n+Äpfel\n+__ggml_vocab_test__\n+\n+__ggml_vocab_test__\n+ \n+__ggml_vocab_test__\n+ \n+__ggml_vocab_test__\n+ \n+__ggml_vocab_test__\n+\t\n+__ggml_vocab_test__\n+\n+\n+__ggml_vocab_test__\n+... | 2026-04-09T09:41:14 |
huggingface/transformers | 20528f067cf9204cea5178ce0f837245e146e159 | 934fe1504e6d5e87e01d96305f4d97faa63cf4c1 | Cast epochs_trained to int when resuming training (#32286)
* fix epochs_trained as int when resuming training
* refactor
---------
Co-authored-by: teddyferdinan <teddy.ferdinan@pwr.edu.pl> | [
{
"path": "src/transformers/trainer.py",
"patch": "@@ -2155,7 +2155,7 @@ def _inner_training_loop(\n self.state = TrainerState.load_from_json(os.path.join(resume_from_checkpoint, TRAINER_STATE_NAME))\n self.compare_trainer_and_checkpoint_args(self.args, self.state)\n self... | 2024-07-30T09:25:54 |
nodejs/node | 73ad3f9bea3993b486621aaf9e61484dc37741d4 | e51fb90a6db53588ab2b884e4309d4eea9e37bbd | inspector: Fix crash for WS connection
Attaching WS session will now include a roundtrip onto the main thread
to make sure there is no other session (e.g. JS bindings)
This change also required refactoring WS socket implementation to better
support scenarios like this.
Fixes: https://github.com/nodejs/node/issues/16... | [
{
"path": "src/inspector_agent.cc",
"patch": "@@ -548,10 +548,6 @@ void Agent::Connect(InspectorSessionDelegate* delegate) {\n client_->connectFrontend(delegate);\n }\n \n-bool Agent::IsConnected() {\n- return io_ && io_->IsConnected();\n-}\n-\n void Agent::WaitForDisconnect() {\n CHECK_NE(client_, nul... | 2017-11-11T00:01:00 |
golang/go | bcfaeca58c791ada53fae18fffa26936bc245423 | 041a4e4c34e21d769de35c54a86b32cdb0475f65 | time: in NewTicker, document that the 1st tick comes after d
Fixes #42245
Change-Id: I3b298ab6be65569389873d68bd3c6e49cf892c69
Reviewed-on: https://go-review.googlesource.com/c/go/+/265818
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Rob Pike <r@golang.org>
Trust: Alberto Donizetti <alb.donizetti@gmai... | [
{
"path": "src/time/tick.go",
"patch": "@@ -13,11 +13,12 @@ type Ticker struct {\n \tr runtimeTimer\n }\n \n-// NewTicker returns a new Ticker containing a channel that will send the\n-// time with a period specified by the duration argument.\n-// It adjusts the intervals or drops ticks to make up for slow ... | 2020-10-28T09:48:17 |
ggml-org/llama.cpp | 243532e5568704ced8b08136e70247f9ceecdbf7 | 5e9c6354638858f4c9aa0a70fd4a41a73f894dcc | jinja : support ensure_ascii=true, string repetition and int/float self-filtering (#21623)
* feat: jinja engine improvements for reka-edge
Port three Jinja engine improvements needed for the reka-edge model:
1. Python-style string repetition ("ab" * 3 → "ababab")
2. ensure_ascii=true support for tojson filter (escape... | [
{
"path": "common/jinja/runtime.cpp",
"patch": "@@ -251,6 +251,23 @@ value binary_expression::execute_impl(context & ctx) {\n return res;\n }\n \n+ // Python-style string repetition\n+ // TODO: support array/tuple repetition (e.g., [1, 2] * 3 → [1, 2, 1, 2, 1, 2])\n+ if (op.value == \"*... | 2026-04-09T09:28:33 |
ollama/ollama | 7e7749224c57ea4d7ae98e4d07dcb00e192a5c7c | c7c2f3bc228714988c650d4c238a08510f58e8cb | Fix use_mmap parsing for modelfiles
Add the new tristate parsing logic for the code path for modelfiles,
as well as a unit test. | [
{
"path": "api/types.go",
"patch": "@@ -608,6 +608,19 @@ func FormatParams(params map[string][]string) (map[string]interface{}, error) {\n \t\t} else {\n \t\t\tfield := valueOpts.FieldByName(opt.Name)\n \t\t\tif field.IsValid() && field.CanSet() {\n+\t\t\t\tif reflect.PointerTo(field.Type()) == reflect.Type... | 2024-06-21T19:27:19 |
huggingface/transformers | 934fe1504e6d5e87e01d96305f4d97faa63cf4c1 | 3e8106d2533cbd890ddd1e919bd62132cd4718c3 | Fix GGUF dequantize for `gguf==0.9.1` (#32298)
* fix gguf dequantize for gguf==0.9.1
* fix old version
* make style | [
{
"path": "src/transformers/integrations/ggml.py",
"patch": "@@ -217,13 +217,13 @@ def _gguf_parse_value(_value, data_type):\n return _value\n \n \n-def dequantize_q4_k(data):\n+def dequantize_q4_k(data, n_bytes: int):\n # C implementation\n # https://github.com/ggerganov/ggml/blob/fca1caafea7de... | 2024-07-30T09:01:00 |
nodejs/node | e51fb90a6db53588ab2b884e4309d4eea9e37bbd | 094bfaf769f5b6bc8ca6fbcc289cdf5e0f07073f | tools: prefer common.expectsError in tests
Add lint rule to validate that common.expectsError(fn, err) is being
used instead of assert.throws(fn, common.expectsError(err));
PR-URL: https://github.com/nodejs/node/pull/17557
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Colin Ihrig <cjihrig@gmail.c... | [
{
"path": "test/.eslintrc.yaml",
"patch": "@@ -10,6 +10,7 @@ rules:\n # Custom rules in tools/eslint-rules\n prefer-assert-iferror: error\n prefer-assert-methods: error\n+ prefer-common-expectserror: error\n prefer-common-mustnotcall: error\n crypto-check: error\n inspector-check: error",
"... | 2017-12-08T21:06:02 |
golang/go | 041a4e4c34e21d769de35c54a86b32cdb0475f65 | 05082c90d5b35935ccc27acb070e00702df91a3a | go/types: add test case for incorrect map index expression
The existing code for map index expressions checked the
wrong variable (x rather than key) to see if the index
assignment was correct. Since x.mode was always valid in
that case, type-checking didn't follow the error exit in
case of an incorrect map index expr... | [
{
"path": "src/go/types/expr.go",
"patch": "@@ -1357,9 +1357,7 @@ func (check *Checker) exprInternal(x *operand, e ast.Expr, hint Type) exprKind {\n \t\t\tvar key operand\n \t\t\tcheck.expr(&key, e.Index)\n \t\t\tcheck.assignment(&key, typ.key, \"map index\")\n-\t\t\tif x.mode == invalid {\n-\t\t\t\tgoto Er... | 2020-11-17T06:23:48 |
ggml-org/llama.cpp | 9949ad08f6453b61603e87236fad1ac8cb8609dd | 3ee9da0e4f2382dbfc65cb8fa48e9eeacda4f6d5 | fix: Model Selector choice sync (#21628) | [
{
"path": "tools/server/public/index.html",
"patch": "@@ -18,7 +18,7 @@\n \t\t<div style=\"display: contents\">\n \t\t\t<script>\n \t\t\t\t{\n-\t\t\t\t\t__sveltekit_nl4lme = {\n+\t\t\t\t\t__sveltekit_6n4hpv = {\n \t\t\t\t\t\tbase: new URL('.', location).pathname.slice(0, -1)\n \t\t\t\t\t};\n ",
"additio... | 2026-04-09T07:46:27 |
huggingface/transformers | 3e8106d2533cbd890ddd1e919bd62132cd4718c3 | f0bc49e7f61f74f055c47ad40e6010f57eed0b0b | Docs: fix GaLore optimizer code example (#32249)
Docs: fix GaLore optimizer example
Fix incorrect usage of GaLore optimizer in Transformers trainer code example.
The GaLore optimizer uses low-rank gradient updates to reduce memory usage. GaLore is quite popular and is implemented by the authors in [https://githu... | [
{
"path": "docs/source/en/trainer.md",
"patch": "@@ -278,7 +278,7 @@ args = TrainingArguments(\n max_steps=100,\n per_device_train_batch_size=2,\n optim=\"galore_adamw\",\n- optim_target_modules=[\"attn\", \"mlp\"]\n+ optim_target_modules=[r\".*.attn.*\", r\".*.mlp.*\"]\n )\n \n model_id =... | 2024-07-30T07:19:24 |
nodejs/node | 094bfaf769f5b6bc8ca6fbcc289cdf5e0f07073f | 4df7c190bf449e3d0d8877f45bb7490d7a801118 | test: replace assert.throws w/ common.expectsError
PR-URL: https://github.com/nodejs/node/pull/17557
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com> | [
{
"path": "test/async-hooks/test-embedder.api.async-resource.js",
"patch": "@@ -17,12 +17,12 @@ common.expectsError(\n code: 'ERR_INVALID_ARG_TYPE',\n type: TypeError,\n });\n-assert.throws(() => {\n+common.expectsError(() => {\n new AsyncResource('invalid_trigger_id', { triggerAsyncId: null });... | 2017-12-08T21:41:38 |
ggml-org/llama.cpp | 3ee9da0e4f2382dbfc65cb8fa48e9eeacda4f6d5 | 75511a8d7e77d978cf3e541ff75177fb1ac576d6 | server : fix grammar commandline args (#21543)
Co-authored-by: AUTOMATIC <-> | [
{
"path": "tools/server/server-task.cpp",
"patch": "@@ -384,6 +384,8 @@ task_params server_task::params_from_json_cmpl(\n throw std::runtime_error(std::string(\"\\\"json_schema\\\": \") + e.what());\n }\n } else {\n+ params.sampling.grammar = defaults.sampling.grammar;\n+\n ... | 2026-04-09T07:16:54 |
golang/go | 05082c90d5b35935ccc27acb070e00702df91a3a | 01df2febf5f00b1dcba1843093ef99b338b23546 | cmd/compile: clean up buggy DWARF inlined info PC ranges
Repair the code that generates PC ranges for DWARF inlined routine
instances to insure that if II Y is a child of II X within the inline
tree, X's ranges include the ranges from Y. This is similar to what
we're already doing for DWARF scopes.
Updates #33188.
C... | [
{
"path": "src/cmd/compile/internal/gc/dwinl.go",
"patch": "@@ -8,6 +8,7 @@ import (\n \t\"cmd/internal/dwarf\"\n \t\"cmd/internal/obj\"\n \t\"cmd/internal/src\"\n+\t\"fmt\"\n \t\"strings\"\n )\n \n@@ -170,12 +171,32 @@ func assembleInlines(fnsym *obj.LSym, dwVars []*dwarf.Var) dwarf.InlCalls {\n \t\taddRan... | 2020-08-17T18:17:07 |
huggingface/transformers | 7f5d644e69068825bb5b6e84cdc56b3d3a9bd04f | 3fbaaaa64d1ef3d8327adb577994d3d11277c77a | [pipeline] fix padding for 1-d tensors (#31776)
* [pipeline] fix padding for 1-d tensors
* add test
* make style
* Update tests/pipelines/test_pipelines_automatic_speech_recognition.py
Co-authored-by: Kamil Akesbi <45195979+kamilakesbi@users.noreply.github.com>
* Update tests/pipelines/test_pipelines_au... | [
{
"path": "src/transformers/pipelines/base.py",
"patch": "@@ -90,6 +90,9 @@ def _pad(items, key, padding_value, padding_side):\n # Others include `attention_mask` etc...\n shape = items[0][key].shape\n dim = len(shape)\n+ if dim == 1:\n+ # We have a list of 1-dim to... | 2024-07-29T13:24:42 |
ollama/ollama | 4ebb66c6623d85f4fb69db0406ddd05bdc2d893d | 23e899f32d9f7b3bbe0b902a95c23be5a1254409 | reformat error check | [
{
"path": "gpu/assets.go",
"patch": "@@ -77,19 +77,20 @@ func cleanupTmpDirs() {\n \t\t\tcontinue\n \t\t}\n \t\traw, err := os.ReadFile(filepath.Join(d, \"ollama.pid\"))\n-\t\tif err == nil {\n-\t\t\tpid, err := strconv.Atoi(string(raw))\n-\t\t\tif err == nil {\n-\t\t\t\tif proc, err := os.FindProcess(pid);... | 2024-06-20T16:23:43 |
ggml-org/llama.cpp | 8a65a7a8ee70061ba180b6634157a8323f293087 | 8a132faaa06d904bbeb3ebfbc8f7e2ed921b8ded | ci: drop v5 `all:` composition from labeler.yml (#21627)
actions/labeler@v6 removed the `all:` / `any:` composition keys.
The `server/webui` and `server` entries used `all:` to combine
`any-glob-to-any-file` with negated `all-globs-to-all-files`,
which now errors on every PR with:
Unknown config options were unde... | [
{
"path": ".github/labeler.yml",
"patch": "@@ -75,21 +75,13 @@ android:\n - examples/llama.android/**\n server/webui:\n - changed-files:\n- - all:\n- - any-glob-to-any-file:\n- - tools/server/webui/**\n- - tools/server/public/**\n- -... | 2026-04-09T06:20:19 |
nodejs/node | 4df7c190bf449e3d0d8877f45bb7490d7a801118 | f3aaaa52b738e20192468a5157721c0a62027a16 | benchmark: fix http/simple.js benchmark
autocannon appears to have trouble recognizing URLs that contain true
or false within them. Use 0 or 1 instead to represent the same.
PR-URL: https://github.com/nodejs/node/pull/17583
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Colin Ihrig <cjihrig@gmail.... | [
{
"path": "benchmark/fixtures/simple-http-server.js",
"patch": "@@ -34,7 +34,7 @@ module.exports = http.createServer(function(req, res) {\n const arg = params[2];\n const n_chunks = parseInt(params[3], 10);\n const resHow = params.length >= 5 ? params[4] : 'normal';\n- const chunkedEnc = params.lengt... | 2017-12-09T22:32:48 |
golang/go | 01df2febf5f00b1dcba1843093ef99b338b23546 | 0968d2d599189229145b1000cec55d9df47fbc98 | cmd/go: allow querying other versions of the main module
'go mod download' and a few other commands can now query specific
versions of the main module.
'go get' still reports an error when attempting to update the main
module.
Fixes #42524
Change-Id: Ia93ef8f5f34443e938667c48a0db432200108c63
Reviewed-on: https://go... | [
{
"path": "src/cmd/go/internal/modcmd/download.go",
"patch": "@@ -88,12 +88,11 @@ func runDownload(ctx context.Context, cmd *base.Command, args []string) {\n \t\targs = []string{\"all\"}\n \t} else if modload.HasModRoot() {\n \t\tmodload.LoadModFile(ctx) // to fill Target\n-\t\ttargetAtLatest := modload.Tar... | 2020-11-16T21:27:19 |
huggingface/transformers | 3fbaaaa64d1ef3d8327adb577994d3d11277c77a | 7ffe25f2b935dcaf65079b04c5f91c8a42a99e28 | Whisper tokenizer word level timestamps (#32197)
* fix _fix_key in PreTrainedModel
* fix _find_longest_common_sequence
* add test
* remove result.json
* nit
* update test | [
{
"path": "src/transformers/models/whisper/tokenization_whisper.py",
"patch": "@@ -1174,7 +1174,22 @@ def _find_longest_common_sequence(sequences, token_timestamp_sequences=None):\n \"There is a bug within whisper `decode_asr` function, please report it. Dropping to prevent bad inference... | 2024-07-29T10:19:52 |
ollama/ollama | fedf71635ec77644f8477a86c6155217d9213a11 | 97c59be6536c1f7c4779c0a2ecd34efb878dfb13 | Extend api/show and ollama show to return more model info (#4881)
* API Show Extended
* Initial Draft of Information
Co-Authored-By: Patrick Devine <pdevine@sonic.net>
* Clean Up
* Descriptive arg error messages and other fixes
* Second Draft of Show with Projectors Included
* Remove Chat Template
... | [
{
"path": "api/types.go",
"patch": "@@ -253,6 +253,7 @@ type ShowRequest struct {\n \tModel string `json:\"model\"`\n \tSystem string `json:\"system\"`\n \tTemplate string `json:\"template\"`\n+\tVerbose bool `json:\"verbose\"`\n \n \tOptions map[string]interface{} `json:\"options\"`\n \n@@ -262,14 ... | 2024-06-19T21:19:02 |
ggml-org/llama.cpp | 2dcb7f74edde532831c5ee9a9b6c403f2b68854f | 660600081fb7b9b769ded5c805a2d39a419f0a0d | fix: free ctx_copy in ggml_opt_free to plug per-training-session leak (#21592)
* fix: free ctx_copy in ggml_opt_free to plug per-training-session leak
ggml_opt_alloc populates opt_ctx->ctx_copy via a free+init pair every
time the allocated graph shape changes. The last ctx_copy from the
final ggml_opt_alloc call surv... | [
{
"path": "ggml/src/ggml-opt.cpp",
"patch": "@@ -589,6 +589,7 @@ void ggml_opt_free(ggml_opt_context_t opt_ctx) {\n ggml_backend_buffer_free(opt_ctx->buf_cpu);\n ggml_free(opt_ctx->ctx_static);\n ggml_free(opt_ctx->ctx_cpu);\n+ ggml_free(opt_ctx->ctx_copy);\n delete opt_ctx;\n }\n ",
... | 2026-04-08T15:40:15 |
golang/go | 0968d2d599189229145b1000cec55d9df47fbc98 | 3e56bad13b0d287cd77472763fec5e75d9846de6 | cmd/go/internal/modget: clarify error for 'go get' without arguments
If the current directory doesn't contain a package, 'go get' will say
that without additional detail.
If there were no arguments, errors will start with "go get:" instead
of "go get .:".
Fixes #39080
Change-Id: I47366f2a27bce17bd8b79344ad15b8b934a... | [
{
"path": "src/cmd/go/internal/modget/get.go",
"patch": "@@ -505,6 +505,12 @@ func parseArgs(ctx context.Context, rawArgs []string) []*query {\n \t\t\tcontinue\n \t\t}\n \n+\t\t// If there were no arguments, CleanPatterns returns \".\". Set the raw\n+\t\t// string back to \"\" for better errors.\n+\t\tif le... | 2020-05-20T17:51:53 |
nodejs/node | 8d23afc807303a9afc1851802e1d88f0d3a84c29 | 51a7d97625f601d8d4b32784739eaad757ebc73e | doc: add link to debugger in process.md
PR-URL: https://github.com/nodejs/node/pull/17522
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Jon Moss <me@jonathanmo... | [
{
"path": "doc/api/process.md",
"patch": "@@ -379,7 +379,7 @@ terminal programs.\n \n It is important to take note of the following:\n \n-* `SIGUSR1` is reserved by Node.js to start the debugger. It's possible to\n+* `SIGUSR1` is reserved by Node.js to start the [debugger][]. It's possible to\n install a... | 2017-12-07T10:11:41 |
huggingface/transformers | 7ffe25f2b935dcaf65079b04c5f91c8a42a99e28 | 49928892d6491ff5a49c12cbc34695f6fa7ac0ed | Generate: end-to-end compilation (#30788)
* mvp
* added test (a few models need fixes)
* fix a few test cases
* test nits
* harder test 😈
* revert changes in stablelm
* test with improved condition
* add todo
* tmp commit
* merged with main
* nits
* add todo
* final corrections
... | [
{
"path": "docs/source/en/llm_optims.md",
"patch": "@@ -18,59 +18,109 @@ Basic inference is slow because LLMs have to be called repeatedly to generate th\n This guide will show you how to use the optimization techniques available in Transformers to accelerate LLM inference.\n \n > [!TIP]\n-> Hugging Face al... | 2024-07-29T09:52:13 |
ggml-org/llama.cpp | e9fd96283d75b55f46f198a68507825f400127d4 | 3ba12fed0a50af94bd9cfdea6f0b59e5aba8ed4a | Propose fix a couple of typos (#21581)
Signed-off-by: John E <jeis4wpi@outlook.com> | [
{
"path": "tests/test-chat.cpp",
"patch": "@@ -3454,7 +3454,7 @@ static void test_template_output_peg_parsers(bool detailed_debug) {\n },\n \"replaceAll\": {\n \"type\": \"boolean\",\n- \"d... | 2026-04-08T14:29:03 |
golang/go | 0ae3b7cb742c586df9b68d9eac042b32148abf9c | 869e2957b9f66021581b839cadce6cb48ad46114 | cmd/compile: fix rules regression with shifts on PPC64
Some rules for PPC64 were checking for a case
where a shift followed by an 'and' of a mask could
be lowered, depending on the format of the mask. The
function to verify if the mask was valid for this purpose
was not checking if the mask was 0 which we don't want t... | [
{
"path": "src/cmd/compile/internal/ssa/rewrite.go",
"patch": "@@ -1427,10 +1427,11 @@ func DecodePPC64RotateMask(sauxint int64) (rotate, mb, me int64, mask uint64) {\n \treturn\n }\n \n-// This verifies that the mask occupies the\n-// rightmost bits.\n+// This verifies that the mask is a set of\n+// consec... | 2020-11-16T14:40:45 |
huggingface/transformers | 49928892d6491ff5a49c12cbc34695f6fa7ac0ed | 6494479f1de9fe16e9c6f89e52eb0cf81f864a7c | fix(docs): Fixed a link in docs (#32274)
Fixed a link in docs. | [
{
"path": "docs/source/en/perf_torch_compile.md",
"patch": "@@ -98,7 +98,7 @@ Below you can find the list of the models we benchmarked.\n - [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224)\n - [microsoft/beit-base-patch16-224-pt22k-ft22k](https://huggingface.co/microsoft/bei... | 2024-07-29T09:50:43 |
ggml-org/llama.cpp | 87f4744a805f7837b95ce7ff94fc60eb7a3fbcc8 | 85d482e6b6706648070f620797e54f1a6a0ff3d8 | examples : disable cb_eval callback for --save-logits (#21553)
This commit updates the debug example to not create the
base_callback_data.
The motivation for this is when using `--save-logits`, which is used by
examples/model-conversion scripts, we often don't care about the tensor
outputs and they just add noise to ... | [
{
"path": "examples/debug/debug.cpp",
"patch": "@@ -9,6 +9,7 @@\n #include <vector>\n #include <filesystem>\n #include <fstream>\n+#include <optional>\n #include <regex>\n \n static void print_usage(int /*argc*/, char ** argv) {\n@@ -222,7 +223,10 @@ int main(int argc, char ** argv) {\n llama_backend_in... | 2026-04-08T12:10:33 |
ollama/ollama | 52ce350b7aecd4bce9c42fe4aac1d85e47a6d774 | 2abebb2cbe9aad5d4c4710fa33dd7cc078ca894b | Fix bad symbol load detection
pointer deref's weren't correct on a few libraries, which explains
some crashes on older systems or miswired symlinks for discovery libraries. | [
{
"path": "gpu/gpu_info_cudart.c",
"patch": "@@ -40,7 +40,7 @@ void cudart_init(char *cudart_lib_path, cudart_init_resp_t *resp) {\n \n for (i = 0; l[i].s != NULL; i++) {\n *l[i].p = LOAD_SYMBOL(resp->ch.handle, l[i].s);\n- if (!l[i].p) {\n+ if (!*(l[i].p)) {\n char *msg = LOAD_ERR();\n ... | 2024-06-19T15:39:07 |
golang/go | 869e2957b9f66021581b839cadce6cb48ad46114 | 97700baf8bd5c0fdbfe38eedc80d3c612805cbda | cmd/go: update 'go help mod init'
'go help mod init' now mentions that the module path can be derived
from the directory within GOPATH. We no longer mention version
control, since that's now ignored.
Fixes #36775
Change-Id: Ia5559ecb537fccd838eeab84517e76aa01989292
Reviewed-on: https://go-review.googlesource.com/c/g... | [
{
"path": "src/cmd/go/alldocs.go",
"patch": "@@ -1230,12 +1230,17 @@\n //\n // \tgo mod init [module]\n //\n-// Init initializes and writes a new go.mod to the current directory,\n-// in effect creating a new module rooted at the current directory.\n-// The file go.mod must not already exist.\n-// If possib... | 2020-10-12T16:03:43 |
huggingface/transformers | 6494479f1de9fe16e9c6f89e52eb0cf81f864a7c | 535fe78b9f1d148684723e51f00645351880c47a | make `p_mask` a numpy array before passing to `select_starts_ends` (#32076)
* fix
* bug fix
* refine
* fix | [
{
"path": "src/transformers/pipelines/document_question_answering.py",
"patch": "@@ -378,7 +378,7 @@ def preprocess(\n # p_mask: mask with 1 for token than cannot be in the answer (0 for token which can be in an answer)\n # We put 0 on the tokens from the context and 1 everywhere els... | 2024-07-29T09:29:11 |
ggml-org/llama.cpp | 85d482e6b6706648070f620797e54f1a6a0ff3d8 | ae65fbdf33b4ccc7a5c8fb4815fe86c1e7b278f5 | parser: fix MiniMax handling (#21573) | [
{
"path": "common/chat.cpp",
"patch": "@@ -1963,7 +1963,7 @@ static common_chat_params common_chat_templates_apply_jinja(const struct common_\n params.add_generation_prompt = true;\n std::string gen_prompt = common_chat_template_direct_apply_impl(tmpl, params);\n auto diff ... | 2026-04-08T10:47:25 |
huggingface/transformers | 535fe78b9f1d148684723e51f00645351880c47a | a2ad9d5ad53f68c1ad268f7f46538eac6f5b631b | Repo: remove exceptions in `check_docstrings` (#32259)
remove exceptions | [
{
"path": "utils/check_docstrings.py",
"patch": "@@ -69,8 +69,6 @@\n \"InputExample\",\n \"InputFeatures\",\n # Signature is *args/**kwargs\n- # \"PretrainedConfig\", #ignored but could be fixed\n- # \"GenerationConfig\", #ignored but could be fixed\n \"TFSequenceSummary\",\n \"TFB... | 2024-07-29T09:06:05 |
golang/go | 97700baf8bd5c0fdbfe38eedc80d3c612805cbda | 38367d098ed4d97539de5e43e03bce985fc56d8e | cmd/go: in 'go get', only load retractions for resolved versions
Previously, 'go get' loaded retractions for every module in the build
list, which took a long time and usually wasn't helpful.
Fixes #42185
Change-Id: I64294585db141106b63ec74aafa0d266b7536ef2
Reviewed-on: https://go-review.googlesource.com/c/go/+/2690... | [
{
"path": "src/cmd/go/internal/modget/get.go",
"patch": "@@ -419,20 +419,7 @@ func runGet(ctx context.Context, cmd *base.Command, args []string) {\n \t\t\tpkgPatterns = append(pkgPatterns, q.pattern)\n \t\t}\n \t}\n-\tif len(pkgPatterns) > 0 {\n-\t\t// We skipped over missing-package errors earlier: we want... | 2020-11-10T20:48:37 |
ggml-org/llama.cpp | 09343c01986667b26d414733950c9d38e91c3065 | 97508acb17ff933c67edb3a2f0997bc19a6abb98 | model : support step3-vl-10b (#21287)
* feat: support step3-vl-10b
* use fused QKV && mapping tensor in tensor_mapping.py
* guard hardcoded params and drop crop metadata
* get understand_projector_stride from global config
* img_u8_resize_bilinear_to_f32 move in step3vl class
* Apply suggestions from code review
... | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -2219,10 +2219,10 @@ def set_gguf_parameters(self):\n self.image_size = self.find_vparam([\"image_size\"])\n self.gguf_writer.add_vision_image_size(self.image_size)\n self.gguf_writer.add_vision_patch_size(self.find_vparam([... | 2026-04-08T07:51:31 |
nodejs/node | 4426d75ede638aaeb6961c22ab1cfc78ccb12056 | 9f55eac346dea4fba6ace9becc1e01d893f56c5c | doc: simplify and clarify FIPS text in BUILDING.md
PR-URL: https://github.com/nodejs/node/pull/17538
Fixes: https://github.com/nodejs/node/issues/17536
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Jon Moss <me@jonathanmoss.me> | [
{
"path": "BUILDING.md",
"patch": "@@ -347,17 +347,13 @@ as `deps/icu` (You'll have: `deps/icu/source/...`)\n \n ## Building Node.js with FIPS-compliant OpenSSL\n \n-NOTE: Windows is not yet supported\n+It is possible to build Node.js with the\n+[OpenSSL FIPS module](https://www.openssl.org/docs/fipsnotes.h... | 2017-12-07T22:49:57 |
ollama/ollama | 380e06e5bea06ae8ded37f47c37bd5d604194d3e | 21adf8b6d2b3edd085f079b6e98766f7d7bdc8cf | types/model: remove Digest
The Digest type in its current form is awkward to work with and presents
challenges with regard to how it serializes via String using the '-'
prefix.
We currently only use this in ollama.com, so we'll move our specific
needs around digest parsing and validation there. | [
{
"path": "types/model/name.go",
"patch": "@@ -4,7 +4,6 @@ package model\n \n import (\n \t\"cmp\"\n-\t\"encoding/hex\"\n \t\"errors\"\n \t\"fmt\"\n \t\"log/slog\"\n@@ -371,57 +370,3 @@ func cutPromised(s, sep string) (before, after string, ok bool) {\n \t}\n \treturn cmp.Or(before, MissingPart), cmp.Or(aft... | 2024-06-18T20:29:38 |
huggingface/transformers | a2ad9d5ad53f68c1ad268f7f46538eac6f5b631b | 5019aabfacf7599b9a6b4e7a1adc1fb5c9017727 | fix: Fixed wrong argument passed to `convert_blip_checkpoint` function call (#32262)
Removed one wrong argument passed to convert_blip_checkpoint function call. | [
{
"path": "src/transformers/models/blip/convert_blip_original_pytorch_to_hf.py",
"patch": "@@ -188,4 +188,4 @@ def convert_blip_checkpoint(pytorch_dump_folder_path, config_path=None):\n parser.add_argument(\"--config_path\", default=None, type=str, help=\"Path to hf config.json of model to convert\")\n ... | 2024-07-29T08:43:09 |
Subsets and Splits
Assembly Language GitHub Issues
Retrieves a sample of assembly-language related commits with their details, but doesn't provide meaningful analysis or patterns beyond basic filtering.
Swift Compiler Issues Analysis
Retrieves all training data for the Swift programming language repository, providing basic filtering but offering limited analytical insight beyond identifying relevant code examples.