ollama

Tree: 1f371ea92f

Jesse Gross 1f371ea92f ml: Panic rather than return error on tensor allocation failure FromFloatSlice and FromIntSlice return an error if the shape doesn't match the passed data or if memory can't be allocated. Since these are inputs, the memory being allocated is system memory rather than VRAM. In many cases, the caller can't really handle the error and panics. Empty and Zeros directly panic if they can't allocate memory. This makes things consistent by panicing for the first two cases, removing a fair amount of error handling code. This is also consistent with how Go typically handles these situations.		11 months ago
..
imageproc	imageproc mllama refactor (#7537)	1 year ago
input	ollamarunner: Separate text and multimodal graphs	11 months ago
models	ml: Panic rather than return error on tensor allocation failure	11 months ago
testdata	gemma2 impl	1 year ago
bytepairencoding.go	model: handle multiple eos tokens (#10577)	11 months ago
bytepairencoding_test.go	model: handle multiple eos tokens (#10577)	11 months ago
model.go	ml: Panic rather than return error on tensor allocation failure	11 months ago
model_test.go	fs: move ml.Config to fs package	1 year ago
sentencepiece.go	model: handle multiple eos tokens (#10577)	11 months ago
sentencepiece_test.go	model: handle multiple eos tokens (#10577)	11 months ago
textprocessor.go	model: handle multiple eos tokens (#10577)	11 months ago
vocabulary.go	model: handle multiple eos tokens (#10577)	11 months ago