Added iterative search to prevent instructions from being echoed, add ignore eos, add no-mmap, fixed 1 character echo too much bug

Mug · Mug · commit 0cccb41a8fa5 · 2023-04-10T16:35:38.000+02:00
diff --git a/examples/low_level_api/common.py b/examples/low_level_api/common.py
@@ -40,6 +40,7 @@ class GptParams:
     instruct: bool = False
     ignore_eos: bool = False
     perplexity: bool = False
+    use_mmap: bool = True
     use_mlock: bool = False
     mem_test: bool = False
     verbose_prompt: bool = False
@@ -110,7 +111,9 @@ def gpt_params_parse(argv = None, params: Optional[GptParams] = None):
         dest="use_color"
     )
     parser.add_argument("--mlock", action="store_true",help="force system to keep model in RAM rather than swapping or compressing",dest="use_mlock")
+    parser.add_argument("--no-mmap", action="store_false",help="do not memory-map model (slower load but may reduce pageouts if not using mlock)",dest="use_mmap")
     parser.add_argument("--mtest", action="store_true",help="compute maximum memory usage",dest="mem_test")
+    parser.add_argument("--verbose-prompt", action="store_true",help="print prompt before generation",dest="verbose_prompt")
     parser.add_argument(
         "-r",
         "--reverse-prompt",
diff --git a/examples/low_level_api/low_level_api_chat_cpp.py b/examples/low_level_api/low_level_api_chat_cpp.py
@@ -26,6 +26,25 @@
 CONSOLE_COLOR_PROMPT = ANSI_COLOR_YELLOW
 CONSOLE_COLOR_USER_INPUT = ANSI_BOLD + ANSI_COLOR_GREEN
 
+# Iterative search
+# Actively searches and prevents a pattern from being returned
+class IterSearch:
+	def __init__(self, pattern):
+		self.pattern = list(pattern)
+		self.buffer = []
+
+	def __call__(self, char):
+		self.buffer += [char]
+
+		if (self.pattern[:len(self.buffer)] == self.buffer):
+			if (len(self.buffer) >= len(self.pattern)):
+				self.buffer.clear()
+			return []
+
+		_tmp = self.buffer[:]
+		self.buffer.clear()
+		return _tmp
+
 # A LLaMA interactive session
 class LLaMAInteract:
 	def __init__(self, params: GptParams) -> None:
@@ -69,6 +88,7 @@ def __init__(self, params: GptParams) -> None:
 		self.lparams.seed = self.params.seed
 		self.lparams.memory_f16 = self.params.memory_f16
 		self.lparams.use_mlock = self.params.use_mlock
+		self.lparams.use_mmap = self.params.use_mmap
 
 		self.ctx = llama_cpp.llama_init_from_file(self.params.model.encode("utf8"), self.lparams)
 		if (not self.ctx):
@@ -114,7 +134,9 @@ def __init__(self, params: GptParams) -> None:
 		# in instruct mode, we inject a prefix and a suffix to each input by the user
 		if (self.params.instruct):
 			self.params.interactive_start = True
-			self.first_antiprompt.append(self._tokenize(self.params.instruct_inp_prefix.strip(), False))
+			_ptn = self._tokenize(self.params.instruct_inp_prefix.strip(), False)
+			self.first_antiprompt.append(_ptn)
+			self.antiecho = IterSearch(_ptn)
 
 		# enable interactive mode if reverse prompt or interactive start is specified
 		if (len(self.params.antiprompt) != 0 or self.params.interactive_start):
@@ -217,7 +239,9 @@ def generate(self):
 			if len(self.embd_inp) <= self.input_consumed:
 				# out of user input, sample next token
 
-				#TODO: self.params.ignore_eos
+				if (self.params.ignore_eos):
+					logits = llama_cpp.llama_get_logits(self.ctx)
+					logits[llama_cpp.llama_token_eos()] = llama_cpp.c_float(0)
 
 				_arr = self.last_n_tokens[-min(self.params.repeat_last_n, self.n_past):]
 				id = llama_cpp.llama_sample_top_p_top_k(
@@ -263,7 +287,11 @@ def generate(self):
 			# display tokens
 			if self.output_echo:
 				for id in self.embd:
-					yield id
+					if self.params.instruct:
+						for r in self.antiecho(id):
+							yield r
+					else:
+						yield id
 
 			# reset color to default if we there is no pending user input
 			if (self.params.input_echo and len(self.embd_inp) == self.input_consumed):
@@ -279,7 +307,7 @@ def generate(self):
 						break
 
 				# if we are using instruction mode, and we have processed the initial prompt
-				if (self.n_past > 0 and self.params.interactive_start):
+				if (self.params.interactive_start):
 					break
 
 			# end of text token
