adding evaluation for apps

apps_eval_util.py

@@ -0,0 +1,545 @@
+ 
+import argparse
+import json
+import os
+import sys
+import io
+import faulthandler
+
+# used for debugging to time steps
+from datetime import datetime
+
+# to run the solution files we're using a timing based approach
+import signal
+
+import numpy as np
+# for capturing the stdout
+from io import StringIO
+from typing import get_type_hints
+from typing import List, Tuple
+# used for testing the code that reads from input
+from unittest.mock import patch, mock_open
+
+from pyext import RuntimeModule
+
+from enum import Enum
+class CODE_TYPE(Enum):
+    call_based = 0
+    standard_input = 1
+
+# stuff for setting up signal timer
+class TimeoutException(Exception):
+    pass
+def timeout_handler(signum, frame):
+    print("alarm went off")
+    #return
+    raise TimeoutException
+signal.signal(signal.SIGALRM, timeout_handler)
+timeout = 4  # seconds
+
+# used to capture stdout as a list
+# from https://stackoverflow.com/a/16571630/6416660
+# alternative use redirect_stdout() from contextlib
+class Capturing(list):
+    def __enter__(self):
+        self._stdout = sys.stdout
+        sys.stdout = self._stringio = StringIO()
+        # Make closing the StringIO a no-op
+        self._stringio.close = lambda x: 1
+        return self
+    def __exit__(self, *args):
+        self.extend(self._stringio.getvalue().splitlines())
+        del self._stringio    # free up some memory
+        sys.stdout = self._stdout
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(description="Utility for testing code generation.")
+    parser.add_argument("-v", "--verbosity-level", action="store", type=int,
+            help="")
+    parser.add_argument("-s", "--source",  type=str, default="leetcode", 
+            choices=["leetcode", "atcoder", "codewars",],
+            help="which data source to gather from.")
+    parser.add_argument("-d", "--data",  type=str, default="question",
+            choices=["question", "q", "solutions", "sol", "s", "starter", "tests", "t"],
+            help="which type of data to receive.")
+    parser.add_argument("-n", "--number",  type=int, default=0,
+            help="which problem to query.")
+
+    args = parser.parse_args()
+    return args
+
+
+def get_valid_problems(data_dir="leetcode"):
+    # these are unnecessary atm
+    if data_dir == "leetcode":
+        root = os.path.join(args.source, "data")
+    elif data_dir == "atcoder":
+        pass
+
+    root = os.path.join(data_dir, "data")
+    if os.path.exists(os.path.join(data_dir, "valid_problems.json")):
+        with open(os.path.join(data_dir, "valid_problems.json"), "r") as f:
+            return json.load(f)
+
+    # after we compute it once let's save it and load that instead
+    # TODO determine if might be better to reload each time
+    tmp = os.listdir(root)
+    valid_probs = []
+    for folder in tmp:
+        prob_path = os.path.join(root, folder)
+        files = os.listdir(prob_path)
+        #TODO add more validity checks
+        if "input_output.json" in files or "sols.json" in files:
+            valid_probs.append(prob_path)
+    valid_probs = sorted(valid_probs)
+    #with open(os.path.join(args.source,"valid_problems.json"), "w") as f:
+    #    json.dump(valid_probs, f)
+    return valid_probs
+
+
+def get_question(problem_list, prob_index):
+    root = problem_list[prob_index]
+    #print("get q", root)
+    if os.path.exists(os.path.join(root, "question.txt")):
+        with open(os.path.join(root, "question.txt")) as f:
+            question = f.readlines()
+    else:
+        print("question prompt not found")
+        question = ""
+    question = "".join(question)
+    return question
+
+
+def get_solutions(problem_list, prob_index):
+    root = problem_list[prob_index]
+    if os.path.exists(os.path.join(root, "solutions.json")):
+        with open(os.path.join(root, "solutions.json")) as f:
+            sols = json.load(f)
+    return sols
+
+
+def run_test(prob_path:str=None, problem_list:List[str]=None, prob_index:int=None, 
+        test:str=None, debug:bool=False):
+    """
+    if test is not None it'll try to run the code.
+    otherwise it'll just return an input and output pair.
+    """
+    if prob_path is None and problem_list is None:
+        print("please provide either prob_path or problem_list")
+        exit()
+
+    if debug:
+        print(f"start = {datetime.now().time()}")
+    if prob_path is not None:
+        root = prob_path
+    elif problem_list is not None:
+        root = problem_list[prob_index]
+
+    if os.path.exists(os.path.join(root, "input_output.json")):
+        with open(os.path.join(root, "input_output.json")) as f:
+            in_outs = json.load(f)
+            if debug:
+                print(f"test cases json = {in_outs['inputs']} {in_outs['outputs']}")
+            
+            if in_outs.get("fn_name") is None:
+                which_type = CODE_TYPE.standard_input  # Standard input
+                method_name = None
+            else:
+                which_type = CODE_TYPE.call_based  # Call-based
+                method_name = in_outs["fn_name"]
+    if debug:
+        print(f"loaded json = {datetime.now().time()}")
+ 
+    #else:
+    #    continue
+    if test is None:
+        return in_outs
+    elif test is not None:
+        results = []
+        sol = "import sys\nimport time\nimport itertools\nfrom itertools import accumulate, product, permutations, combinations\nimport collections\nfrom collections import Counter, OrderedDict, deque, defaultdict, ChainMap\nfrom functools import lru_cache\nimport math\nfrom math import sqrt, sin, cos, tan, ceil, fabs, floor, gcd, exp, log, log2\nimport fractions\nfrom typing import List, Tuple\nimport numpy as np\nimport random\nimport heapq\nfrom heapq import *\n"
+        if debug:
+            print(f"loading test code = {datetime.now().time()}")
+ 
+        if which_type == CODE_TYPE.call_based:
+            sol += test
+            if debug: # or True:
+                print(f"sol = {sol}")
+            signal.alarm(timeout)
+            try:
+                tmp_sol = RuntimeModule.from_string("tmp_sol", "", sol)
+                if "class Solution" not in test:
+                    tmp = tmp_sol
+                else:
+                    tmp = tmp_sol.Solution()
+                signal.alarm(0)
+            except Exception as e:
+                signal.alarm(0)
+                print(f"type 0 compilation error = {e}")
+                results.append(-2)
+                return results
+            signal.alarm(0)
+
+        elif which_type == CODE_TYPE.standard_input:
+            # sol
+            tmp_test = test.split("\n")
+
+            new_test = []
+            for x in tmp_test:
+                if (not x.startswith("from ")) and (not x.startswith("import ")):
+                    new_test.append("\t" + x + "\n")
+                else:
+                    new_test.append(x + "\n")
+            tmp_test = new_test
+            
+            new_test = ""
+            started = False
+            for i in tmp_test:
+                if i.startswith("\t") and not started:
+                    new_test += "stdin = sys.stdin\nstdout = sys.stdout\n"
+                    new_test += "def code():\n"
+                    new_test += i
+                    started = True
+                elif started and ((i.startswith("from ")) or (i.startswith("import "))): 
+                    new_test += "\t" + i
+                else:
+                    new_test += i
+            tmp_test = new_test
+
+            sol += tmp_test
+            if debug:
+                print(f"sol = {sol}")
+                # print(f"{o}") 
+            method_name = "code"
+            signal.alarm(timeout)
+            try:
+                tmp_sol = RuntimeModule.from_string("tmp_sol", "", sol)
+                tmp = tmp_sol
+                signal.alarm(0)
+            except Exception as e:
+                signal.alarm(0)
+                print(f"type 1 compilation error = {e}")
+                results.append(-2)
+                return results
+            signal.alarm(0)
+        if debug:
+            print(f"get method = {datetime.now().time()}")
+ 
+        try:
+            method = getattr(tmp, method_name)  # get_attr second arg must be str
+        except:
+            signal.alarm(0)
+            e = sys.exc_info()
+            print(f"unable to get function error = {e}")
+            return results
+
+        for index, inputs in enumerate(in_outs["inputs"]):
+            # JSON forces dictionaries to have string keys; this undoes this (assuming a singleton list)
+            try:
+                if isinstance(inputs[0], dict):
+                    inputs = [{int(k): v for k,v in inputs[0].items()}]
+            except:
+                True
+            try:
+                if isinstance(in_outs["outputs"][index], dict):
+                    in_outs["outputs"][index] = [{int(k): v for k,v in in_outs["outputs"][index].items()}]
+            except:
+                True
+            try:
+                if isinstance(in_outs["outputs"][index][0], dict):
+                    in_outs["outputs"][index] = [{int(k): v for k,v in in_outs["outputs"][index][0].items()}]
+            except:
+                True
+
+            if debug:
+                print(f"time: {datetime.now().time()} testing index = {index}  inputs = {inputs}, {type(inputs)}. type = {which_type}")
+            if which_type == CODE_TYPE.call_based:  # Call-based
+                signal.alarm(timeout)
+                faulthandler.enable()
+                try:
+                    # print("------------")
+                    # print(inputs)
+                    output = method(*inputs)
+
+                    # ground truth sequences are not tuples
+                    if isinstance(output, tuple):
+                        output = list(output)
+                    
+                    tmp_result = output == in_outs["outputs"][index]
+                    if isinstance(in_outs["outputs"][index], list) and in_outs["outputs"][index]:
+                        tmp_result = tmp_result or (output == in_outs["outputs"][index][0])
+
+                    # ground truth sequences are not tuples
+                    try:
+                        if isinstance(output[0], tuple):
+                            tmp_result = tmp_result or ([list(x) for x in output] == in_outs["outputs"][index][0])
+                    except:
+                        True
+                    results.append(tmp_result)
+
+                    # reset the alarm
+                    signal.alarm(0)
+                except Exception as e:
+                    signal.alarm(0)
+                    faulthandler.disable()
+                    print(f"Standard input runtime error or time limit exceeded error = {e}")
+                    results.append(-1)
+                    continue
+                faulthandler.disable()
+                signal.alarm(0)
+                if debug:
+                    print(f"outputs = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs}, {type(inputs)}, {output == [in_outs['outputs'][index]]}")
+            elif which_type == CODE_TYPE.standard_input:  # Standard input
+                faulthandler.enable()
+                signal.alarm(timeout)
+                passed = False
+
+                if isinstance(inputs, list):
+                    inputs = "\n".join(inputs)
+                if isinstance(in_outs['outputs'][index], list):
+                    in_outs['outputs'][index] = "\n".join(in_outs['outputs'][index])
+
+                with Capturing() as output:
+                    try:
+                        call_method(method, inputs)
+                        # reset the alarm
+                        signal.alarm(0)
+                        passed = True
+                    except Exception as e:
+                        # runtime error or took too long
+                        signal.alarm(0)
+                        print(f"Call-based runtime error or time limit exceeded error = {repr(e)}{e}")
+                        results.append(-1)
+                    signal.alarm(0)
+
+                if not passed:
+                    if debug:
+                        nl = "\n"
+                        if not isinstance(inputs, list):
+                            print(f"not passed output = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs.replace(nl,' new-line ')}, {type(inputs)}, {output == [in_outs['outputs'][index]]}")
+                        else:
+                            print(f"not passed output = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs}, {type(inputs)}, {output == [in_outs['outputs'][index]]}")
+                    continue
+
+                if passed and debug:
+                    print(f"==> output = {output}, test outputs = {in_outs['outputs'][index]}")
+
+                if custom_compare_(output, in_outs['outputs'][index]):
+                    tmp_result = True
+                    results.append(tmp_result)
+                    continue
+
+                # ground truth sequences are expressed as lists not tuples
+                if isinstance(output, tuple):
+                    output = list(output)
+
+                tmp_result = False
+                try:
+                    tmp_result = (output == [in_outs["outputs"][index]])
+                    if isinstance(in_outs["outputs"][index], list):
+                        tmp_result = tmp_result or (output == in_outs["outputs"][index])
+                        if isinstance(output[0], str):
+                            tmp_result = tmp_result or ([e.strip() for e in output] == in_outs["outputs"][index])
+                except Exception as e:
+                    print(f"Failed check1 exception = {e}")
+                    pass
+
+                if tmp_result == True:  
+                    results.append(tmp_result)
+                    continue
+
+                # try one more time without \n
+                if isinstance(in_outs["outputs"][index], list):
+                    for tmp_index, i in enumerate(in_outs["outputs"][index]):
+                        in_outs["outputs"][index][tmp_index] = i.split("\n")
+                        in_outs["outputs"][index][tmp_index] = [x.strip() for x in in_outs["outputs"][index][tmp_index] if x]
+                else:
+                    in_outs["outputs"][index] = in_outs["outputs"][index].split("\n")
+                    in_outs["outputs"][index] = list(filter(len, in_outs["outputs"][index]))
+                    in_outs["outputs"][index] = list(map(lambda x:x.strip(), in_outs["outputs"][index]))
+
+                try:
+                    tmp_result = (output == [in_outs["outputs"][index]])
+                    if isinstance(in_outs["outputs"][index], list):
+                        tmp_result = tmp_result or (output == in_outs["outputs"][index])
+                except Exception as e:
+                    print(f"Failed check2 exception = {e}")
+                    pass
+
+                if tmp_result == True:
+                    results.append(tmp_result)
+                    continue
+
+                # try by converting the output into a split up list too
+                if isinstance(output, list):
+                    output = list(filter(len, output))
+
+                if debug:
+                    nl = "\n"
+                    if not isinstance(inputs, list):
+                        print(f"output = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs.replace(nl,' new-line ')}, {type(inputs)}, {output == [in_outs['outputs'][index]]}") 
+                    else:
+                        print(f"output = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs}, {type(inputs)}, {output == [in_outs['outputs'][index]]}") 
+                
+                if tmp_result == True:
+                    results.append(tmp_result)
+                    continue
+
+                try:
+                    tmp_result = (output == [in_outs["outputs"][index]])
+                    if isinstance(in_outs["outputs"][index], list):
+                        tmp_result = tmp_result or (output == in_outs["outputs"][index])
+                except Exception as e:
+                    print(f"Failed check3 exception = {e}")
+                    pass
+
+                try:
+                    output_float = [float(e) for e in output]
+                    gt_float = [float(e) for e in in_outs['outputs'][index]]
+                    tmp_result = tmp_result or ((len(output_float) == len(gt_float)) and np.allclose(output_float, gt_float))
+                except Exception as e:
+                    pass
+                try:
+                    if isinstance(output[0], list):
+                        output_float = [float(e) for e in output[0]]
+                        gt_float = [float(e) for e in in_outs['outputs'][index][0]]
+                        tmp_result = tmp_result or ((len(output_float) == len(gt_float)) and np.allclose(output_float, gt_float))
+                except Exception as e:
+                    pass
+
+                if tmp_result == True:
+                    results.append(tmp_result)
+                    continue
+
+                # try by converting the stuff into split up list
+                if isinstance(in_outs["outputs"][index], list):
+                    for tmp_index, i in enumerate(in_outs["outputs"][index]):
+                        in_outs["outputs"][index][tmp_index] = set(i.split())
+                else:
+                    in_outs["outputs"][index] = set(in_outs["outputs"][index].split())
+
+                try:
+                    tmp_result = (output == in_outs["outputs"][index])
+                except Exception as e:
+                    print(f"Failed check4 exception = {e}")
+                    continue
+
+                if tmp_result == True:
+                    results.append(tmp_result)
+                    continue 
+
+                # try by converting the output into a split up list too
+                if isinstance(output, list):
+                    for tmp_index, i in enumerate(output):
+                        output[tmp_index] = i.split()
+                    output = list(filter(len, output))
+                    for tmp_index, i in enumerate(output):
+                        output[tmp_index] = set(i)    
+                else:
+                    output = output.split()
+                    output = list(filter(len, output))
+                    output = set(output)
+
+                try:
+                    tmp_result = (set(frozenset(s) for s in output) == set(frozenset(s) for s in in_outs["outputs"][index]))
+                except Exception as e:
+                    print(f"Failed check5 exception = {e}")
+
+
+                # if they are all numbers, round so that similar numbers are treated as identical
+                try:
+                    tmp_result = tmp_result or (set(frozenset(round(float(t),3) for t in s) for s in output) ==\
+                        set(frozenset(round(float(t),3) for t in s) for s in in_outs["outputs"][index]))
+                except Exception as e:
+                    print(f"Failed check6 exception = {e}")
+                
+                if tmp_result == True and debug:
+                    print("PASSED")
+ 
+                results.append(tmp_result)
+                
+                if debug:
+                    nl = "\n"
+                    if not isinstance(inputs, list):
+                        print(f"output = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs.replace(nl,' new-line ')}, {type(inputs)}, {output == [in_outs['outputs'][index]]}")
+                    else:
+                        print(f"output = {output}, test outputs = {in_outs['outputs'][index]}, inputs = {inputs}, {type(inputs)}, {output == [in_outs['outputs'][index]]}") 
+
+
+    return results
+
+def custom_compare_(output, ground_truth):
+    
+    if isinstance(output, list):
+        output_1 = "\n".join(output)
+        if stripped_string_compare(output_1, ground_truth):
+            return True
+
+    if isinstance(output, list):
+        output_2 = [o.lstrip().rstrip() for o in output]
+        output_2 = "\n".join(output_2)
+        if stripped_string_compare(output_2, ground_truth):
+            return True
+
+    return False
+
+def stripped_string_compare(s1, s2):
+    s1 = s1.lstrip().rstrip()
+    s2 = s2.lstrip().rstrip()
+    return s1 == s2
+
+def call_method(method, inputs):
+
+    if isinstance(inputs, list):
+        inputs = "\n".join(inputs)
+
+    inputs_line_iterator = iter(inputs.split("\n"))
+
+    # sys.setrecursionlimit(10000)
+
+    # @patch('builtins.input', side_effect=inputs.split("\n"))
+    @patch('builtins.open', mock_open(read_data=inputs))
+    @patch('sys.stdin', StringIO(inputs))
+    @patch('sys.stdin.readline', lambda *args: next(inputs_line_iterator))
+    @patch('sys.stdin.readlines', lambda *args: inputs.split("\n"))
+    @patch('sys.stdin.read', lambda *args: inputs)
+    # @patch('sys.stdout.write', print)
+    def _inner_call_method(_method):
+        try:
+            return _method()
+        except SystemExit as e:
+            pass
+        finally:
+            pass
+    return _inner_call_method(method) 
+
+def main(args):
+    print(args)
+    problem_list = sorted(get_valid_problems(args.source))
+    print(f"number of problems = {len(problem_list)}")
+    prob_index = args.number
+    print(f"problem is {problem_list[prob_index]}")
+
+    # This checks it correctly loaded. remove this later
+    assert prob_index < len(problem_list)
+
+    if args.data == "q" or args.data == "question":
+        tmp = get_question(problem_list, prob_index)
+        print("q", tmp)
+    elif args.data in ["solutions", "sol", "s",]:
+        tmp = get_solutions(problem_list, prob_index)
+        print("sol", tmp)
+    elif args.data == "starter":
+        tmp = get_starter(problem_list, prob_index)
+        print("starter", tmp)
+    elif args.data in ["test", "t"]:
+        # test it with sols
+        sols = get_solutions(problem_list, prob_index)
+        tmp = run_test(problem_list, prob_index, test=sols[0])
+
+        print("results = ", tmp)
+        print("-2 = compile error, -1 is runtime error, False failed test, True passed test")
+
+if __name__ == "__main__":
+    args = parse_args()
+    main(args)

eval_apps.py

@@ -0,0 +1,176 @@
+"""
+Run solutions from one problem.
+"""
+
+import io
+import json
+import logging
+import math
+import numpy as np
+import os
+import pprint
+import sys
+import testing_util as test_util
+import time
+
+# for timing debugging
+from datetime import datetime, date
+from tqdm import tqdm
+
+from typing import List
+
+def print_results(results, args):
+    res = []
+    per_prob_res = []
+    all_correct = []
+    for index in results:
+       res.extend(results[index])
+       per_prob_res.append(np.mean(results[index]))
+       all_correct.append(np.all(results[index]))
+    tmp_results = res
+    compile_errors = len(tmp_results[tmp_results==-2])
+    runtime_errors = len(tmp_results[tmp_results==-1])
+    failures = len(tmp_results[tmp_results==False])
+    successes = len(tmp_results[tmp_results==True])
+    total_testcases = len(res)
+    if args.debug:
+        print(f"number of compile errors = {compile_errors} avg = {compile_errors / total_testcases }")
+        print(f"number of runtime errors = {runtime_errors} avg = {runtime_errors / total_testcases}")
+        print(f"number of test cases run = {total_testcases}")
+
+    print(f"Test Case Average (average accuracy over problems) = {np.mean(per_prob_res)}")
+    print(f"Strict Accuracy (all test cases passed / total problems) = {np.mean(all_correct)}")
+
+
+def eval_and_save_problems(args):
+    with open(args.test_loc, "r") as f:
+        problems = sorted(json.load(f))
+
+    print(len(problems))
+    gpt_codes = {}
+    gpt_bleu = {}
+    gpt_codebleu = {}
+    results = {}
+    codes_loc = os.path.join(args.save, f"all_codes.json")
+    if not os.path.exists(codes_loc):
+        codes_loc = os.path.join(args.save, f"{args.start}-{args.end}_codes.json")
+
+    if os.path.exists(codes_loc):
+        results_loc = os.path.join(args.save, f"all_results.json") 
+    else:
+        results_loc = os.path.join(args.save, f"{args.start}-{args.end}_results.json") 
+    print(codes_loc, results_loc)
+
+    with open(codes_loc, "r") as f: 
+        gpt_codes = json.load(f)
+
+    if args.index:
+        problems = [problems[args.index]]
+    else:
+        if args.start > len(problems) or args.start < 0:
+            print(f"start index {args.start} > number of problems {len(problems)}")
+            return
+        start = args.start
+        if args.end is None or args.end > len(problems):
+            end = len(problems)
+        else:
+            end = args.end
+        problems = problems[start:end]
+
+    if args.stop_early:
+        problems = problems[:args.stop_early]
+
+    # main eval loop
+    for index, problem in enumerate(tqdm(problems)):
+        try:
+            if args.debug:
+                print(f"\n\nproblem path = {problem}")
+            output_str = gpt_codes[str(index+args.start)]
+        except:
+            print("CANNOT FIND OUTPUT_STR FOR", problem)
+            continue
+        prob_path = os.path.join(args.root, problem)
+
+        with open(os.path.join(prob_path, "solutions.json"), "r") as f:
+            sols = json.load(f)
+        
+        if not os.path.exists(args.save):
+            os.makedirs(args.save)
+
+        res = []
+        for o_idx, o in enumerate(output_str):
+            if args.debug:
+                print(f"\nTesting solution {o_idx}")
+            curr_res = [-2]
+            try:
+                curr_res = test_util.run_test(prob_path=prob_path, test=o, debug=args.debug)
+                fixed = []
+                for e in curr_res:
+                    if isinstance(e, np.ndarray):
+                       e = e.item(0)
+                    if isinstance(e, np.bool_):
+                        e = bool(e)
+                    fixed.append(e)
+                curr_res = fixed
+                if not np.all(curr_res):
+                    print(f"Results were not all True: {curr_res}")
+            except Exception as e:
+                print(f"test framework exception = {repr(e)}{e}\n")
+                break
+            finally:
+                assert isinstance(curr_res, list)
+                res.append(curr_res)
+
+        if args.debug:
+            print(f"\nHow to read results [-2] = compile error, [-1] = runtime error [False] = failed test case [True] = passed test case")
+            #print(f"results = {res}")
+ 
+        results[index+args.start+args.index] = res
+        
+        with open(results_loc, "w") as f:
+            try:
+                f.write(json.dumps(results))
+            except Exception as e:
+                import pdb; pdb.set_trace()
+                print("didn't save problem due to {e}")
+
+    return results
+
+
+def main(args):
+
+    argsdict = vars(args)
+    print(pprint.pformat(argsdict))
+
+    if args.print_results:
+        results = {}
+        codes_loc = os.path.join(args.save, f"all_codes.json")
+        if os.path.exists(codes_loc):
+            results_loc = os.path.join(args.save, f"all_results.json") 
+        else:
+            results_loc = os.path.join(args.save, f"{args.start}-{args.end}_results.json") 
+        with open(results_loc, "r") as f: 
+            results = json.load(f)
+    else:
+        results = eval_and_save_problems(args)
+
+    print_results(results, args)
+
+
+if __name__ == "__main__":
+    import argparse
+
+    parser = argparse.ArgumentParser(description="Testing a Language Model on Python Code")
+    parser.add_argument("-t","--test_loc", default="../data_split/test.json", type=str, help="path to the json containing problem paths to be evaluated.")
+    parser.add_argument("-r","--root", default="../", type=str, help="where the data is stored.")
+    parser.add_argument("-s","--start", default=0, type=int)
+    parser.add_argument("-e","--end", default=None, type=int, help="If you want to evaluate a subset of problems specify start and ending index. File with start and ending prefix must exist typically used with batch evaluation.")
+    parser.add_argument("-i", "--index", default=0, type=int)
+    parser.add_argument("-p", "--print_results", action="store_true", help="If you have already evaluated the results and only want to print them.")
+    parser.add_argument("-d", "--debug", action="store_true")
+    parser.add_argument("--save", type=str, default="./results", help="Where the evaluated data is loaded from and results saved to.")
+    parser.add_argument("--stop-early", default=None, type=int)
+ 
+    args = parser.parse_args()
+
+    main(args)

finetune_apps.sh

@@ -0,0 +1,31 @@
+#! /bin/bash
+./run_clm_apps.py \
+    --output_dir $HOME/gpt-code-clippy-apps-2 \
+    --model_name_or_path EleutherAI/gpt-neo-1.3B \
+    --dataset_name ./apps.py \
+    --do_train --do_eval \
+    --block_size="1024" \
+    --per_device_train_batch_size="2" \
+    --per_device_eval_batch_size="2" \
+    --preprocessing_num_workers="16" \
+    --learning_rate="2e-5" \
+    --warmup_steps="5000" \
+    --adam_beta1="0.9" \
+    --adam_beta2="0.98" \
+    --weight_decay="0.1" \
+    --overwrite_output_dir \
+    --num_train_epochs="5" \
+    --logging_steps="20" \
+    --eval_steps="1000" \
+    --push_to_hub="False" \
+    --report_to="wandb" \
+    --dtype="bfloat16" \
+    --skip_memory_metrics="False" \
+    --save_steps="1000" \
+    --save_strategy epoch \
+    --save_total_limit 2 \
+    --gradient_accumulation_steps 2 \
+    --adafactor \
+    # --resume_from_checkpoint $HOME/gpt-neo-125M-code-clippy/ckpt_201 \
+    # --max_train_samples="10000" \
+    # --max_eval_samples="1000"

partitions.py

@@ -1,86 +0,0 @@
-#!/usr/bin/env python
-# coding=utf-8
-# Copyright 2021 The Google Research Authors and The HuggingFace Team All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""Utilities for constructing PyTrees of PartitionSpecs."""
-
-# utils adapted from https://github.com/google-research/google-research/blob/master/flax_models/t5x/partitions.py
-
-import re
-
-from flax.core.frozen_dict import freeze
-from flax.traverse_util import flatten_dict, unflatten_dict
-from jax.experimental import PartitionSpec as P
-
-
-# Sentinels
-_unmatched = object()
-
-# For specifying empty leaf dict `{}`
-empty_dict = object()
-
-
-def _match(qs, ks):
-    """Return True if regexes in qs match any window of strings in tuple ks."""
-    # compile regexes and force complete match
-    qts = tuple(map(lambda x: re.compile(x + "$"), qs))
-    for i in range(len(ks) - len(qs) + 1):
-        matches = [x.match(y) for x, y in zip(qts, ks[i:])]
-        if matches and all(matches):
-            return True
-    return False
-
-
-def _replacement_rules(rules):
-    def replace(key, val):
-        for rule, replacement in rules:
-            if _match(rule, key):
-                return replacement
-        return val
-
-    return replace
-
-
-# PartitionSpec for GPTNeo
-# replicate the hidden dim and shard feed-forward and head dim
-def _get_partition_rules():
-    return [
-        # embeddings
-        (("transformer", "wpe", "embedding"), P("mp", None)),
-        (("transformer", "wte", "embedding"), P("mp", None)),
-        # atention
-        (("attention", "(q_proj|k_proj|v_proj)", "kernel"), P(None, "mp")),
-        (("attention", "out_proj", "kernel"), P("mp", None)),
-        (("attention", "out_proj", "bias"), None),
-        # mlp
-        (("mlp", "c_fc", "kernel"), P(None, "mp")),
-        (("mlp", "c_fc", "bias"), P("mp")),
-        (("mlp", "c_proj", "kernel"), P("mp", None)),
-        (("mlp", "c_proj", "bias"), None),
-        # layer norms
-        ((r"ln_\d+", "bias"), None),
-        ((r"\d+", r"ln_\d+", "scale"), None),
-        (("ln_f", "bias"), None),
-        (("ln_f", "scale"), None),
-    ]
-
-
-def set_partitions(in_dict):
-    rules = _get_partition_rules()
-    replace = _replacement_rules(rules)
-    initd = {k: _unmatched for k in flatten_dict(in_dict)}
-    result = {k: replace(k, v) for k, v in initd.items()}
-    assert _unmatched not in result.values(), "Incomplete partition spec."
-    return freeze(unflatten_dict(result))
-{"mode":"full","isActive":false}

run_clm_mp_apps.py

@@ -0,0 +1,636 @@
+#!/usr/bin/env python
+# coding=utf-8
+# Copyright 2021 The HuggingFace Team All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Pre-training/Fine-tuning the GPTNeo model for causal language modeling on a text file or a dataset using model parallelism.
+"""
+
+import logging
+import math
+import os
+import sys
+import time
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Callable, Optional
+
+import datasets
+import numpy as np
+from datasets import Dataset, load_dataset
+from tqdm import tqdm
+
+import jax
+import jax.numpy as jnp
+import optax
+import transformers
+from flax.core.frozen_dict import freeze, unfreeze
+from flax.training.common_utils import onehot, stack_forest
+from jax.experimental.maps import mesh
+from jax.experimental.pjit import pjit
+from partitions import set_partitions
+from transformers import (
+    CONFIG_MAPPING,
+    FLAX_MODEL_FOR_CAUSAL_LM_MAPPING,
+    AutoConfig,
+    AutoTokenizer,
+    FlaxAutoModelForCausalLM,
+    HfArgumentParser,
+    TrainingArguments,
+    is_tensorboard_available,
+)
+from transformers.testing_utils import CaptureLogger
+
+
+logger = logging.getLogger(__name__)
+
+MODEL_CONFIG_CLASSES = list(FLAX_MODEL_FOR_CAUSAL_LM_MAPPING.keys())
+MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)
+
+
+@dataclass
+class ModelArguments:
+    """
+    Arguments pertaining to which model/config/tokenizer we are going to fine-tune, or train from scratch.
+    """
+
+    model_name_or_path: Optional[str] = field(
+        default=None,
+        metadata={
+            "help": "The model checkpoint for weights initialization."
+            "Don't set if you want to train a model from scratch."
+        },
+    )
+    model_type: Optional[str] = field(
+        default=None,
+        metadata={"help": "If training from scratch, pass a model type from the list: " + ", ".join(MODEL_TYPES)},
+    )
+    config_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
+    )
+    tokenizer_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
+    )
+    cache_dir: Optional[str] = field(
+        default=None, metadata={"help": "Where do you want to store the pretrained models downloaded from s3"}
+    )
+    use_fast_tokenizer: bool = field(
+        default=True,
+        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
+    )
+    dtype: Optional[str] = field(
+        default="float32",
+        metadata={
+            "help": "Floating-point format in which the model weights should be initialized and trained. Choose one of `[float32, float16, bfloat16]`."
+        },
+    )
+    from_pt: Optional[bool] = field(
+        default=False,
+         metadata={"help": "Whether the model weights should be converted from pytorch."},
+    )
+
+
+@dataclass
+class DataTrainingArguments:
+    """
+    Arguments pertaining to what data we are going to input our model for training and eval.
+    """
+
+    dataset_name: Optional[str] = field(
+        default=None, metadata={"help": "The name of the dataset to use (via the datasets library)."}
+    )
+    dataset_config_name: Optional[str] = field(
+        default=None, metadata={"help": "The configuration name of the dataset to use (via the datasets library)."}
+    )
+    train_file: Optional[str] = field(default=None, metadata={"help": "The input training data file (a text file)."})
+    validation_file: Optional[str] = field(
+        default=None,
+        metadata={"help": "An optional input evaluation data file to evaluate the perplexity on (a text file)."},
+    )
+    max_train_samples: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": "For debugging purposes or quicker training, truncate the number of training examples to this "
+            "value if set."
+        },
+    )
+    max_eval_samples: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": "For debugging purposes or quicker training, truncate the number of evaluation examples to this "
+            "value if set."
+        },
+    )
+    overwrite_cache: bool = field(
+        default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
+    )
+    validation_split_percentage: Optional[int] = field(
+        default=5,
+        metadata={
+            "help": "The percentage of the train set used as validation set in case there's no validation split"
+        },
+    )
+    block_size: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": "Optional input sequence length after tokenization. "
+            "The training dataset will be truncated in block of this size for training. "
+            "Default to the model max input length for single sentence inputs (take into account special tokens)."
+        },
+    )
+    overwrite_cache: bool = field(
+        default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
+    )
+    preprocessing_num_workers: Optional[int] = field(
+        default=None,
+        metadata={"help": "The number of processes to use for the preprocessing."},
+    )
+
+    def __post_init__(self):
+        if self.dataset_name is None and self.train_file is None and self.validation_file is None:
+            raise ValueError("Need either a dataset name or a training/validation file.")
+        else:
+            if self.train_file is not None:
+                extension = self.train_file.split(".")[-1]
+                assert extension in ["csv", "json", "txt"], "`train_file` should be a csv, a json or a txt file."
+            if self.validation_file is not None:
+                extension = self.validation_file.split(".")[-1]
+                assert extension in ["csv", "json", "txt"], "`validation_file` should be a csv, a json or a txt file."
+
+
+def data_loader(rng: jax.random.PRNGKey, dataset: Dataset, batch_size: int, shuffle: bool = False):
+    """
+    Returns batches of size `batch_size` from truncated `dataset`, sharded over all local devices.
+    Shuffle batches if `shuffle` is `True`.
+    """
+    steps_per_epoch = len(dataset) // batch_size
+
+    if shuffle:
+        batch_idx = jax.random.permutation(rng, len(dataset))
+    else:
+        batch_idx = jnp.arange(len(dataset))
+
+    batch_idx = batch_idx[: steps_per_epoch * batch_size]  # Skip incomplete batch.
+    batch_idx = batch_idx.reshape((steps_per_epoch, batch_size))
+
+    for idx in batch_idx:
+        batch = dataset[idx]
+        batch = {k: jnp.array(v) for k, v in batch.items()}
+        yield batch
+
+
+def write_train_metric(summary_writer, train_metrics, train_time, step):
+    summary_writer.scalar("train_time", train_time, step)
+
+    train_metrics = stack_forest(train_metrics)
+    for key, vals in train_metrics.items():
+        tag = f"train_{key}"
+        for i, val in enumerate(vals):
+            summary_writer.scalar(tag, val, step - len(vals) + i + 1)
+
+
+def write_eval_metric(summary_writer, eval_metrics, step):
+    for metric_name, value in eval_metrics.items():
+        summary_writer.scalar(f"eval_{metric_name}", value, step)
+
+
+def create_learning_rate_fn(
+    train_ds_size: int, train_batch_size: int, num_train_epochs: int, num_warmup_steps: int, learning_rate: float
+) -> Callable[[int], jnp.array]:
+    """Returns a linear warmup, linear_decay learning rate function."""
+    steps_per_epoch = train_ds_size // train_batch_size
+    num_train_steps = steps_per_epoch * num_train_epochs
+    warmup_fn = optax.linear_schedule(init_value=0.0, end_value=learning_rate, transition_steps=num_warmup_steps)
+    decay_fn = optax.linear_schedule(
+        init_value=learning_rate, end_value=0, transition_steps=num_train_steps - num_warmup_steps
+    )
+    schedule_fn = optax.join_schedules(schedules=[warmup_fn, decay_fn], boundaries=[num_warmup_steps])
+    return schedule_fn
+
+
+def main():
+    # See all possible arguments in src/transformers/training_args.py
+    # or by passing the --help flag to this script.
+    # We now keep distinct sets of args, for a cleaner separation of concerns.
+
+    parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments))
+    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
+        # If we pass only one argument to the script and it's the path to a json file,
+        # let's parse it to get our arguments.
+        model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args, data_args, training_args = parser.parse_args_into_dataclasses()
+
+    if (
+        os.path.exists(training_args.output_dir)
+        and os.listdir(training_args.output_dir)
+        and training_args.do_train
+        and not training_args.overwrite_output_dir
+    ):
+        raise ValueError(
+            f"Output directory ({training_args.output_dir}) already exists and is not empty."
+            "Use --overwrite_output_dir to overcome."
+        )
+
+    # Make one log on every process with the configuration for debugging.
+    logging.basicConfig(
+        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
+        datefmt="%m/%d/%Y %H:%M:%S",
+        level=logging.INFO,
+    )
+    # Setup logging, we only want one process per machine to log things on the screen.
+    logger.setLevel(logging.INFO if jax.process_index() == 0 else logging.ERROR)
+    if jax.process_index() == 0:
+        datasets.utils.logging.set_verbosity_warning()
+        transformers.utils.logging.set_verbosity_info()
+    else:
+        datasets.utils.logging.set_verbosity_error()
+        transformers.utils.logging.set_verbosity_error()
+
+    # Set the verbosity to info of the Transformers logger (on main process only):
+    logger.info(f"Training/evaluation parameters {training_args}")
+
+    # Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below)
+    # or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/
+    # (the dataset will be downloaded automatically from the datasets Hub).
+    #
+    # For CSV/JSON files, this script will use the column called 'text' or the first column if no column called
+    # 'text' is found. You can easily tweak this behavior (see below).
+    if data_args.dataset_name is not None:
+        # Downloading and loading a dataset from the hub.
+        dataset = load_dataset(
+            data_args.dataset_name, data_args.dataset_config_name, cache_dir=model_args.cache_dir, keep_in_memory=False
+        )
+
+        if "validation" not in dataset.keys():
+            dataset["validation"] = load_dataset(
+                data_args.dataset_name,
+                data_args.dataset_config_name,
+                split=f"train[:{data_args.validation_split_percentage}%]",
+                cache_dir=model_args.cache_dir,
+            )
+            dataset["train"] = load_dataset(
+                data_args.dataset_name,
+                data_args.dataset_config_name,
+                split=f"train[{data_args.validation_split_percentage}%:]",
+                cache_dir=model_args.cache_dir,
+            )
+    else:
+        data_files = {}
+        if data_args.train_file is not None:
+            data_files["train"] = data_args.train_file
+        if data_args.validation_file is not None:
+            data_files["validation"] = data_args.validation_file
+        extension = data_args.train_file.split(".")[-1]
+        if extension == "txt":
+            extension = "text"
+        dataset = load_dataset(extension, data_files=data_files, cache_dir=model_args.cache_dir)
+    # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
+    # https://huggingface.co/docs/datasets/loading_datasets.html.
+
+    # Load pretrained config and tokenizer
+    if model_args.config_name:
+        config = AutoConfig.from_pretrained(model_args.config_name, cache_dir=model_args.cache_dir)
+    elif model_args.model_name_or_path:
+        config = AutoConfig.from_pretrained(model_args.model_name_or_path, cache_dir=model_args.cache_dir)
+    else:
+        config = CONFIG_MAPPING[model_args.model_type]()
+        logger.warning("You are instantiating a new config instance from scratch.")
+
+    if model_args.tokenizer_name:
+        tokenizer = AutoTokenizer.from_pretrained(
+            model_args.tokenizer_name, cache_dir=model_args.cache_dir, use_fast=model_args.use_fast_tokenizer
+        )
+    elif model_args.model_name_or_path:
+        tokenizer = AutoTokenizer.from_pretrained(
+            model_args.model_name_or_path, cache_dir=model_args.cache_dir, use_fast=model_args.use_fast_tokenizer
+        )
+    else:
+        raise ValueError(
+            "You are instantiating a new tokenizer from scratch. This is not supported by this script."
+            "You can do it from another script, save it, and load it from here, using --tokenizer_name."
+        )
+
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = tokenizer.eos_token
+
+    if training_args.do_train:
+        column_names = dataset["train"].column_names
+    else:
+        column_names = dataset["validation"].column_names
+    text_column_name = "text" if "text" in column_names else column_names[0]
+
+
+    if data_args.block_size is None:
+        block_size = tokenizer.model_max_length
+        if block_size > config.max_position_embeddings:
+            logger.warning(
+                f"The tokenizer picked seems to have a very large `model_max_length` ({tokenizer.model_max_length}). "
+                "Picking 1024 instead. You can change that default value by passing --block_size xxx."
+            )
+            block_size = 1024
+    else:
+        if data_args.block_size > tokenizer.model_max_length:
+            logger.warning(
+                f"The block_size passed ({data_args.block_size}) is larger than the maximum length for the model"
+                f"({tokenizer.model_max_length}). Using block_size={tokenizer.model_max_length}."
+            )
+        block_size = min(data_args.block_size, tokenizer.model_max_length)
+
+        
+    def tokenize_function(examples):
+        toks = tokenizer(examples["question"],
+                         examples["answer"], 
+                         max_length=block_size,
+                         padding="max_length",
+                         truncation=True, 
+                         return_token_type_ids=True,
+                        #  return_tensors="np",
+                         )
+        labels = toks["input_ids"].copy()
+        toks["labels"] = labels
+        return toks
+
+
+    lm_datasets = dataset.map(
+        tokenize_function,
+        batched=True,
+        num_proc=data_args.preprocessing_num_workers,
+        remove_columns=column_names,
+        load_from_cache_file=not data_args.overwrite_cache,
+    )
+
+
+    if training_args.do_train:
+        if "train" not in lm_datasets:
+            raise ValueError("--do_train requires a train dataset")
+        train_dataset = lm_datasets["train"]
+        if data_args.max_train_samples is not None:
+            train_dataset = train_dataset.select(range(data_args.max_train_samples))
+
+    if training_args.do_eval:
+        if "validation" not in lm_datasets:
+            raise ValueError("--do_eval requires a validation dataset")
+        eval_dataset = lm_datasets["validation"]
+        if data_args.max_eval_samples is not None:
+            eval_dataset = eval_dataset.select(range(data_args.max_eval_samples))
+
+    # Enable tensorboard only on the master node
+    has_tensorboard = is_tensorboard_available()
+    if has_tensorboard and jax.process_index() == 0:
+        try:
+            from flax.metrics.tensorboard import SummaryWriter
+
+            summary_writer = SummaryWriter(log_dir=Path(training_args.output_dir))
+        except ImportError as ie:
+            has_tensorboard = False
+            logger.warning(
+                f"Unable to display metrics through TensorBoard because some package are not installed: {ie}"
+            )
+    else:
+        logger.warning(
+            "Unable to display metrics through TensorBoard because the package is not installed: "
+            "Please run pip install tensorboard to enable."
+        )
+
+    # Initialize our training
+    rng = jax.random.PRNGKey(training_args.seed)
+    rng, dropout_rng = jax.random.split(rng)
+
+    # Store some constant
+    num_epochs = int(training_args.num_train_epochs)
+    train_batch_size = int(training_args.per_device_train_batch_size) * jax.device_count()
+    eval_batch_size = int(training_args.per_device_eval_batch_size) * jax.device_count()
+    steps_per_epoch = len(train_dataset) // train_batch_size
+    total_train_steps = steps_per_epoch * num_epochs
+
+    # TODO: weights should be initialized in pjitted fun, this won't work for REALLY large models
+    # TODO: when loading from pre-trained model we need to make sure the vocab is divisible by num_partitions
+    # GPT2's vocab is odd, we need to resize it for fine-tuning
+    model = FlaxAutoModelForCausalLM.from_pretrained(
+        model_args.model_name_or_path, seed=training_args.seed, dtype=getattr(jnp, model_args.dtype), from_pt=model_args.from_pt,
+    )
+
+    # Create learning rate schedule
+    linear_decay_lr_schedule_fn = create_learning_rate_fn(
+        len(train_dataset),
+        train_batch_size,
+        training_args.num_train_epochs,
+        training_args.warmup_steps,
+        training_args.learning_rate,
+    )
+
+    if training_args.adafactor:
+        optimizer = optax.adafactor(
+            learning_rate=linear_decay_lr_schedule_fn,
+        )
+    else:
+        optimizer = optax.adamw(
+            learning_rate=linear_decay_lr_schedule_fn,
+            b1=training_args.adam_beta1,
+            b2=training_args.adam_beta2,
+            eps=training_args.adam_epsilon,
+            weight_decay=training_args.weight_decay,
+        )
+
+    def get_initial_state(params):
+        state = optimizer.init(params)
+        return tuple(state), params
+
+    # Get PartitionSpec for model params
+    param_spec = set_partitions(unfreeze(model.params))
+
+    # Get the PyTree for opt_state, we don't actually initialize the opt_state yet.
+    params_shapes = jax.tree_map(lambda x: x.shape, model.params)
+    state_shapes = jax.eval_shape(get_initial_state, params_shapes)
+
+    # get PartitionSpec for opt_state, this is very specific to adamw
+    # TODO: optax returns different state for different optimizers, how can we handle this generically ?
+    # or maybe we don't since in our examples we just use adamw or adafactor
+    def get_opt_spec(x):
+        if isinstance(x, dict):
+            return param_spec
+        return None
+
+    opt_state_spec, param_spec = jax.tree_map(
+        get_opt_spec, state_shapes, is_leaf=lambda x: isinstance(x, (dict, optax.EmptyState))
+    )
+
+    # pjit the get_initial_state function to shard params and init
+    # optimizer state in sharded way
+    p_get_initial_state = pjit(
+        get_initial_state,
+        in_axis_resources=None,
+        out_axis_resources=(opt_state_spec, param_spec),
+    )
+
+    # hack: move the inital params to CPU to free up device memory
+    # TODO: allow loading weights on CPU in pre-trained model
+    model.params = jax.tree_map(lambda x: np.asarray(x), model.params)
+
+    # mesh defination
+    mesh_devices = np.array(jax.devices()).reshape(1, jax.local_device_count())
+
+    # actually initialize the opt_state
+    with mesh(mesh_devices, ("dp", "mp")):
+        opt_state, params = p_get_initial_state(freeze(model.params))
+
+    # cross-entropy with z loss
+    def loss_fn(logits, labels, z_loss=0):
+        shift_logits = logits[..., :-1, :]
+        shift_labels = labels[..., 1:]
+
+        shift_labels = onehot(shift_labels, shift_logits.shape[-1])
+
+        shift_logits = shift_logits - jax.lax.stop_gradient(shift_logits.max(axis=-1, keepdims=True))
+        log_z = jnp.log(jnp.sum(jnp.exp(shift_logits), axis=-1, keepdims=True))
+        log_softmax = shift_logits - log_z
+        loss = -jnp.sum(shift_labels * log_softmax, axis=-1)
+
+        loss += (1e-4 * jnp.square(log_z.squeeze(-1))) * z_loss
+
+        return loss.mean()
+
+    # Define gradient update step fn
+    # TODO: try to use TrainState instead of passing params and opt_state individually
+    def train_step(params, opt_state, dropout_rng, batch, step):
+        dropout_rng, new_dropout_rng = jax.random.split(dropout_rng)
+
+        def compute_loss(params):
+            labels = batch.pop("labels")
+            # TODO: mask question in loss_func
+            token_type_ids = batch.pop('token_type_ids')
+            logits = model(**batch, params=params, dropout_rng=dropout_rng, train=True)[0]
+            loss = loss_fn(logits, labels, z_loss=1.0)
+            return loss
+
+        grad_fn = jax.value_and_grad(compute_loss)
+        loss, grads = grad_fn(params)
+
+        updates, new_opt_state = optimizer.update(grads, opt_state, params)
+        new_params = optax.apply_updates(params, updates)
+
+        metrics = {"loss": loss, "learning_rate": linear_decay_lr_schedule_fn(step)}
+        return new_params, tuple(new_opt_state), new_dropout_rng, metrics, step + 1
+
+    # Define eval fn
+    def eval_step(input_ids, labels, params):
+        logits = model(input_ids=input_ids, params=params, train=False)[0]
+        loss = loss_fn(logits, labels)
+        # metrics
+        return {"loss": loss}
+
+    p_train_step = pjit(
+        train_step,
+        in_axis_resources=(param_spec, opt_state_spec, None, None, None),
+        out_axis_resources=(param_spec, opt_state_spec, None, None, None),
+        donate_argnums=(0, 1),
+    )
+
+    p_eval_step = pjit(
+        eval_step,
+        in_axis_resources=(None, None, param_spec),
+        out_axis_resources=None,
+    )
+
+    logger.info("***** Running training *****")
+    logger.info(f"  Num examples = {len(train_dataset)}")
+    logger.info(f"  Num Epochs = {num_epochs}")
+    logger.info(f"  Instantaneous batch size per device = {training_args.per_device_train_batch_size}")
+    logger.info(f"  Total train batch size (w. parallel & distributed) = {train_batch_size}")
+    logger.info(f"  Total optimization steps = {total_train_steps}")
+
+    train_time = 0
+    train_metrics = []
+    epochs = tqdm(range(num_epochs), desc=f"Epoch ... (1/{num_epochs})", position=0)
+    global_step = 0
+    # we are not doing 2D parallelism (yet!), this just does model parallelism
+    with mesh(mesh_devices, ("dp", "mp")):
+        for _ in epochs:
+            # ======================== Training ================================
+            train_start = time.time()
+
+            # Create sampling rng
+            rng, input_rng = jax.random.split(rng)
+
+            # Generate an epoch by shuffling sampling indices from the train dataset
+            train_metrics = []
+            train_loader = data_loader(input_rng, train_dataset, train_batch_size, shuffle=True)
+            steps_per_epoch = len(train_dataset) // train_batch_size
+
+            # train
+            for _ in tqdm(range(steps_per_epoch), desc="Training...", position=1, leave=False):
+                batch = next(train_loader)
+
+                params, opt_state, dropout_rng, train_metric, global_step = p_train_step(
+                    params,
+                    opt_state,
+                    dropout_rng,
+                    batch,
+                    global_step,
+                )
+                train_metrics.append(train_metric)
+
+                cur_step = global_step
+
+                if cur_step % training_args.logging_steps == 0 and cur_step > 0:
+                    # Save metrics
+                    train_time += time.time() - train_start
+                    if has_tensorboard and jax.process_index() == 0:
+                        write_train_metric(summary_writer, train_metrics, train_time, cur_step)
+
+                    epochs.write(
+                        f"Step... ({cur_step} | Loss: {train_metric['loss']}, Learning Rate: {train_metric['learning_rate']})"
+                    )
+
+                    train_metrics = []
+
+                if cur_step % training_args.eval_steps == 0 and cur_step > 0:
+                    # ======================== Evaluating ==============================
+                    eval_metrics = []
+                    eval_loader = data_loader(input_rng, eval_dataset, eval_batch_size)
+                    eval_steps = len(eval_dataset) // eval_batch_size
+
+                    for _ in tqdm(range(eval_steps), desc="Evaluating...", position=2, leave=False):
+                        batch = next(eval_loader)
+                        metrics = p_eval_step(batch["input_ids"], batch["labels"], params)
+                        eval_metrics.append(metrics)
+
+                    # normalize eval metrics
+                    eval_metrics = stack_forest(eval_metrics)
+                    eval_metrics = jax.tree_map(jnp.mean, eval_metrics)
+
+                    try:
+                        eval_metrics["perplexity"] = math.exp(eval_metrics["loss"])
+                    except OverflowError:
+                        eval_metrics["perplexity"] = float("inf")
+
+                    logger.info(
+                        f"Step... ({cur_step} | Eval loss: {eval_metrics['loss']} | Eval Perplexity: {eval_metrics['perplexity']}"
+                    )
+
+                if cur_step % training_args.save_steps == 0 and cur_step > 0:
+                    # save checkpoint after each epoch and push checkpoint to the hub
+                    if jax.process_index() == 0:
+                        params = jax.device_get(params)
+                        model.save_pretrained(
+                            training_args.output_dir,
+                            params=params,
+                            push_to_hub=training_args.push_to_hub,
+                            commit_message=f"Saving weights and logs of step {cur_step}",
+                        )
+
+
+if __name__ == "__main__":
+    main()

run_clm_streaming.sh

@@ -1,16 +1,16 @@
 #! /bin/bash
 ./run_clm_streaming_flax_v2.py \
-    --output_dir $HOME/gpt-neo-125M-test \
-    --model_name_or_path="EleutherAI/gpt-neo-125M" \
+    --output_dir $HOME/gpt-neo-13B-test \
+    --model_name_or_path EleutherAI/gpt-neo-1.3B \
     --dataset_name $HOME/gpt-code-clippy/code_clippy.py \
-    --data_dir /home/shared/code-clippy-dataset/merged-data \
+    --data_dir /home/arto/exdata/merged-data \
     --text_column_name="text" \
     --do_train --do_eval \
-    --block_size="2048" \
-    --per_device_train_batch_size="8" \
-    --per_device_eval_batch_size="16" \
+    --block_size="1024" \
+    --per_device_train_batch_size="1" \
+    --per_device_eval_batch_size="2" \
     --preprocessing_num_workers="8" \
-    --learning_rate="6e-4" \
+    --learning_rate="1e-4" \
     --max_steps 500 \
     --warmup_steps 150 \
     --decay_steps 250 \
@@ -26,9 +26,9 @@
     --skip_memory_metrics="False" \
     --save_steps="50" \
     --save_total_limit 2 \
-    --gradient_accumulation_steps 8 \
+    --gradient_accumulation_steps 1 \
     --report_to="wandb" \
-    --run_name="testing-mini" \
+    --run_name="testing" \
     --max_eval_samples 100 \
     --save_optimizer true \
     # --adafactor \