Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-14 22:38:58 +01:00
Code Issues Packages Projects Releases Wiki Activity

batched.swift : fix build

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov 2024-02-23 16:15:37 +02:00
parent 8772658b11
commit fff1e8a54a
No known key found for this signature in database
GPG Key ID: BF970631944C16B7
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
examples/batched.swift/Sources/main.swift
View File

@ -79,7 +79,7 @@ batch.n_tokens = Int32(tokens.count)
for (i, token) in tokens.enumerated() { for (i, token) in tokens.enumerated() {
batch.token[i] = token batch.token[i] = token
batch.pos[i] = Int32(i) batch.pos[i] = Float(i)
batch.n_seq_id[i] = 1 batch.n_seq_id[i] = 1
// batch.seq_id[i][0] = 0 // batch.seq_id[i][0] = 0
// TODO: is this the proper way to do this? // TODO: is this the proper way to do this?
@ -98,7 +98,7 @@ if llama_decode(context, batch) != 0 {
} }
for i in 1 ..< n_parallel { for i in 1 ..< n_parallel {
llama_kv_cache_seq_cp(context, 0, Int32(i), 0, batch.n_tokens) llama_kv_cache_seq_cp(context, 0, Int32(i), 0, Float(batch.n_tokens))
} }
if n_parallel > 1 { if n_parallel > 1 {
@ -125,8 +125,8 @@ while n_cur <= n_len {
continue continue
} }
var n_vocab = llama_n_vocab(model) let n_vocab = llama_n_vocab(model)
var logits = llama_get_logits_ith(context, i_batch[i]) let logits = llama_get_logits_ith(context, i_batch[i])
var candidates: [llama_token_data] = .init(repeating: llama_token_data(), count: Int(n_vocab)) var candidates: [llama_token_data] = .init(repeating: llama_token_data(), count: Int(n_vocab))
@ -173,7 +173,7 @@ while n_cur <= n_len {
// push this new token for next evaluation // push this new token for next evaluation
batch.token[Int(batch.n_tokens)] = new_token_id batch.token[Int(batch.n_tokens)] = new_token_id
batch.pos[Int(batch.n_tokens)] = n_cur batch.pos[Int(batch.n_tokens)] = Float(n_cur)
batch.n_seq_id[Int(batch.n_tokens)] = 1 batch.n_seq_id[Int(batch.n_tokens)] = 1
if let seq_id = batch.seq_id[Int(batch.n_tokens)] { if let seq_id = batch.seq_id[Int(batch.n_tokens)] {
seq_id[0] = Int32(i) seq_id[0] = Int32(i)