update: tokenizer

lib/nlp/getEmbedding.ts

@@ -1,38 +0,0 @@
-type EmbeddingDict = { [key: number]: Float32Array };
-
-function getEmbeddingLayer(buffer: Buffer): EmbeddingDict {
-    const dict: EmbeddingDict = {};
-
-    const entrySize = 514;
-    const numEntries = buffer.length / entrySize;
-
-    for (let i = 0; i < numEntries; i++) {
-        const offset = i * entrySize;
-        const key = buffer.readUInt16LE(offset);
-        const floatArray = new Float32Array(128);
-
-        for (let j = 0; j < 128; j++) {
-            floatArray[j] = buffer.readFloatLE(offset + 2 + j * 4);
-        }
-
-        dict[key] = floatArray;
-    }
-
-    return dict;
-}
-
-function getEmbedding(tokenIds: number[], embeddingDict: EmbeddingDict, contextSize: number) {
-  let result: number[] = [];
-  for (let i = 0; i < contextSize; i++) {
-    if (i < tokenIds.length) {
-      const tokenId = tokenIds[i];
-      result = result.concat(Array.from(embeddingDict[tokenId]))
-    }
-    else {
-      result = result.concat(new Array(128).fill(0))
-    }
-  }
-  return new Float32Array(result);
-}
-
-export {getEmbeddingLayer, getEmbedding};

lib/nlp/tokenizer.ts

@@ -1,56 +1,16 @@
-type TokenDict = { [key: string]: number };
+import { AutoTokenizer, env } from "@xenova/transformers";
 
-function tokenize(query: string, tokenDict: TokenDict): number[] {
+env.allowRemoteModels = false;
-	const tokenIds: number[] = [];
+env.localModelPath = "/transformers/";
-	let index = 0;
 
-	// Replace spaces with "▁"
+async function tokenize(text: string, model: string) {
-	query = "▁" + query.replace(/ /g, "▁");
+	const tokenizer = await AutoTokenizer.from_pretrained(model);
-	query = query.replace(/\n/g, "<0x0A>");
+	const { input_ids } = await tokenizer(text);
-
+	const tokenIds = [];
-	while (index < query.length) {
+	for (let id of input_ids.data) {
-		let bestToken = null;
+		tokenIds.push(parseInt(id));
-		let bestLength = 0;
-
-		// Step 2: Find the longest token that matches the beginning of the remaining query
-		for (const token in tokenDict) {
-			if (query.startsWith(token, index) && token.length > bestLength) {
-				bestToken = token;
-				bestLength = token.length;
 	}
-		}
-
-		if (bestToken) {
-			tokenIds.push(tokenDict[bestToken]);
-			index += bestLength;
-			continue;
-		}
-
-		// Step 3: Handle the case where no token matches
-		const char = query[index];
-		if (char.charCodeAt(0) <= 127) {
-			// If the character is ASCII, and it doesn't match any token, treat it as an unknown token
-			console.error(`Unknown token: ${char}`);
-            index++;
-            continue;
-		}
-
-		// If the character is non-ASCII, convert it to a series of bytes and match each byte
-		const bytes = new TextEncoder().encode(char);
-		for (const byte of bytes) {
-			const byteToken = `<0x${byte.toString(16).toUpperCase()}>`;
-			if (tokenDict[byteToken] === undefined) {
-                console.error(`Unknown byte token: ${byteToken}`);
-                index++;
-                continue;
-			}
-            tokenIds.push(tokenDict[byteToken]);
-		}
-		index++;
-	}
-
 	return tokenIds;
 }
 
 export default tokenize;
-export type { TokenDict };

package.json

@@ -17,6 +17,7 @@
     "@types/bun": "^1.1.6",
     "@types/express": "^4.17.21",
     "@types/tr46": "^5.0.0",
+    "@xenova/transformers": "^2.17.2",
     "cac": "^6.7.14",
     "chalk": "^5.3.0",
     "express": "^4.19.2",

public/transformers/Qwen/Qwen2.5-3B/config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

public/transformers/Qwen/Qwen2.5-3B/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}