fix: use block_in_place in llama tokenizer (#235)

This commit changes the `LLamaTokenizer` to use
`tokio::task::block_in_place` in an attempt avoid the following error:

```console
$ env RUST_BACKTRACE=1 \
 LLM_CHAIN_MODEL=models/llama-2-7b-chat.ggmlv3.q4_0.bin \
 cargo r --example few_shot

thread 'main' panicked at
'Cannot block the current thread from within a runtime. This happens
because a function attempted to block the current thread while the
thread is being used to drive asynchronous tasks.',
crates/llm-chain-llama/src/executor.rs:290:36
stack backtrace:
   0: rust_begin_unwind
             at /rustc/8ede3aae28fe6e4d52b38157d7bfe0d3bceef225/library/std/src/panicking.rs:593:5
   1: core::panicking::panic_fmt
             at /rustc/8ede3aae28fe6e4d52b38157d7bfe0d3bceef225/library/core/src/panicking.rs:67:14
   2: core::panicking::panic_display
             at /rustc/8ede3aae28fe6e4d52b38157d7bfe0d3bceef225/library/core/src/panicking.rs:150:5
   3: core::panicking::panic_str
             at /rustc/8ede3aae28fe6e4d52b38157d7bfe0d3bceef225/library/core/src/panicking.rs:134:5
   4: core::option::expect_failed
             at /rustc/8ede3aae28fe6e4d52b38157d7bfe0d3bceef225/library/core/src/option.rs:1932:5
   5: core::option::Option<T>::expect
             at /rustc/8ede3aae28fe6e4d52b38157d7bfe0d3bceef225/library/core/src/option.rs:898:21
   6: tokio::future::block_on::block_on
             at /home/danielbevenius/.cargo/registry/src/index.crates.io-6f17d22bba15001f/tokio-1.33.0/src/future/block_on.rs:6:21
   7: tokio::sync::mutex::Mutex<T>::blocking_lock
             at /home/danielbevenius/.cargo/registry/src/index.crates.io-6f17d22bba15001f/tokio-1.33.0/src/sync/mutex.rs:510:9
   8: <llm_chain_llama::executor::LLamaTokenizer as llm_chain::tokens::Tokenizer>::tokenize_str
             at ./llm-chain-llama/src/executor.rs:290:23
   9: <llm_chain_llama::executor::Executor as llm_chain::traits::Executor>::tokens_used
             at ./llm-chain-llama/src/executor.rs:233:31
```

Fixes: #211

Signed-off-by: Daniel Bevenius <[email protected]>

crates/llm-chain-llama/examples/few_shot.rs

@@ -5,10 +5,10 @@ use llm_chain::{chains::conversation::Chain, executor, parameters, prompt, step:
 ///
 /// This example can be seen as a "chain of thought"
 ///
-/// Usage: cargo run --example few-shot
+/// Usage: cargo run --example few_shot
 ///
 /// Make sure to have the env var 'LLM_CHAIN_MODEL' set
-#[tokio::main(flavor = "current_thread")]
+#[tokio::main(flavor = "multi_thread", worker_threads = 1)]
 async fn main() -> Result<(), Box<dyn std::error::Error>> {
     let opts = options!(
         NThreads: 4_usize,

crates/llm-chain-llama/src/executor.rs

@@ -287,14 +287,19 @@ impl<'a> LLamaTokenizer<'a> {
 
 impl Tokenizer for LLamaTokenizer<'_> {
     fn tokenize_str(&self, doc: &str) -> Result<TokenCollection, TokenizerError> {
-        let context = self.context.blocking_lock();
-        let tokenized = tokenize(&context, doc, true);
+        let tokenized = tokio::task::block_in_place(|| {
+            let context = self.context.blocking_lock();
+            tokenize(&context, doc, true)
+        });
         Ok(tokenized.into())
     }
 
     fn to_string(&self, tokens: TokenCollection) -> Result<String, TokenizerError> {
-        let context = self.context.blocking_lock();
-        let output = embedding_to_output(&context, &tokens.as_i32()?);
+        let tokens = &tokens.as_i32()?;
+        let output = tokio::task::block_in_place(|| {
+            let context = self.context.blocking_lock();
+            embedding_to_output(&context, tokens)
+        });
         Ok(output.to_string())
     }
 }