From a8b59aa473d52437568429cca51af5831edea22a Mon Sep 17 00:00:00 2001 From: invo-coder19 Date: Wed, 18 Feb 2026 11:08:56 +0530 Subject: [PATCH] feat: update SYMBA_HEP and SYMBA_REG implementations --- .../model/helpers/mamba.py | 3 --- .../model/mamba_encdec.py | 5 ----- .../src/SineKAN_MoE/model.py | 1 - .../2022-08-14-QED-DataPreparation/dev/ampl_raw.py | 1 - .../dev/implement_hybrid_prefix_and_tokens_combiner.py | 7 +------ .../scripts/DataPreparation_parallel.py | 1 - .../source/ExpressionsTokensCombiner.py | 8 -------- .../2022-09-15-QCD-DataPreparation/dev/ampl_raw.py | 1 - .../dev/implement_hybrid_prefix_and_tokens_combiner.py | 7 +------ .../scripts/DataPreparation_parallel.py | 1 - .../source/ExpressionsTokensCombiner.py | 8 -------- .../source/ExpressionsTokensCombiner.py | 8 -------- .../source/ExpressionsTokensCombiner.py | 8 -------- .../data_preprocessing/ampl_tree/source/ampl_to_tree.py | 8 -------- .../data_preprocessing/tree/sympy_to_tree.py | 7 ------- .../DPOFineTuning/Tokenizers.py | 3 --- .../SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/model.py | 3 --- SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/Tokenizers.py | 3 --- SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/model.py | 3 --- .../SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/Tokenizers.py | 3 --- .../SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/model.py | 3 --- .../algorithms/gp/prepare_dataset.py | 4 +--- .../algorithms/hybrid/custom_gp.py | 1 - .../algorithms/hybrid/hybrid.py | 3 --- .../algorithms/xval_transformers/engine/trainer.py | 2 -- .../algorithms/xval_transformers/model/seq2seq.py | 4 ---- .../src/embeddings/t_net_embeddings.py | 4 +--- 27 files changed, 4 insertions(+), 106 deletions(-) diff --git a/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/helpers/mamba.py b/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/helpers/mamba.py index 72dcf3e..69bc4fa 100644 --- a/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/helpers/mamba.py +++ b/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/helpers/mamba.py @@ -239,7 +239,6 @@ def forward( hidden_states = self.embedding.forward(input_ids) residual = None for idx, layer in enumerate(self.layers): - # print('layer: ',idx) hidden_states, residual = layer( hidden_states, context=context, @@ -295,12 +294,10 @@ def __init__( factory_kwargs = {"device": device, "dtype": dtype} super().__init__() - # print(vocab_size) # if vocab_size % pad_vocab_size_multiple != 0: # vocab_size += pad_vocab_size_multiple - ( # vocab_size % pad_vocab_size_multiple # ) - # print(vocab_size) self.backbone = MixerModel( d_model=d_model, diff --git a/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/mamba_encdec.py b/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/mamba_encdec.py index 73a86f8..6768a11 100644 --- a/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/mamba_encdec.py +++ b/SYMBA_HEP/SYMBAHEP_Hybrid_SSM_Prasanth_Naidu/model/mamba_encdec.py @@ -138,7 +138,6 @@ def forward( input_ids=context_tokens, mask=source_attention_mask, ) - # print(source_vec.dtype, source_attention_mask.dtype) cache = self.allocate_inference_cache( batch_size=b, max_seqlen=300 + l + 1, # source + BOS @@ -153,10 +152,6 @@ def forward( # batch, seqlen, dim = self.decoder.backbone.embedding.forward(input_ids).shape # conv_state, ssm_state = self.decoder.backbone.layers[0].mixer._get_states_from_cache(inference_params, b) # inference_params = None - # print(conv_state.type(),input_ids.type(), source_vec.type()) - # print(source_attention_mask.type(), target_attention_mask.type()) - # print(position_ids.type()) - # print(num_last_tokens) out = self.decoder.forward( input_ids, diff --git a/SYMBA_HEP/SYMBAHEP_Sym_KAN_TP_Transformer_by_Ayush_Mishra/src/SineKAN_MoE/model.py b/SYMBA_HEP/SYMBAHEP_Sym_KAN_TP_Transformer_by_Ayush_Mishra/src/SineKAN_MoE/model.py index 9d66a39..a58baba 100644 --- a/SYMBA_HEP/SYMBAHEP_Sym_KAN_TP_Transformer_by_Ayush_Mishra/src/SineKAN_MoE/model.py +++ b/SYMBA_HEP/SYMBAHEP_Sym_KAN_TP_Transformer_by_Ayush_Mishra/src/SineKAN_MoE/model.py @@ -290,7 +290,6 @@ def forward(self , src, tgt): dec_output = decod_embed for decoder in self.decoder: dec_output = decoder(dec_output , enc_output , src_mask , tgt_mask ) - # print("Decoder output " , dec_output) out = self.ff(dec_output) # out = self.ff(dec_output) diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/ampl_raw.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/ampl_raw.py index 115f7c1..ddac25f 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/ampl_raw.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/ampl_raw.py @@ -59,7 +59,6 @@ # break # sqampls_prefix = [] - # print("Loading squared amplitudes") # for exp in tqdm(sqampl_raw): # # simplified = sp.factor(exp) # worked best for simplification # # prefix = sympy_to_prefix(simplified) diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py index c931cee..7b954fa 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py @@ -95,22 +95,17 @@ def _queue_mgr(func_str: str, q_in: mp.Queue, q_out: mp.Queue, timeout: int, pid proc = mp.Process(target=_lemmiwinks, args=(func_str, (x,), {}, q_worker,)) proc.start() try: - # print(f'[{pid}]: {positioning}: getting') res = q_worker.get(timeout=timeout) - # print(f'[{pid}]: {positioning}: got') q_out.put((positioning, res)) except mpq.Empty: q_out.put((positioning, sp.sympify(x))) - # print(f'[{pid}]: {positioning}: timed out ({timeout}s)') with open(timeout_logfile, "a") as f: f.write("Timed out after "+str(timeout)+" seconds. Argument:" + x + "\n") finally: try: proc.terminate() - # print(f'[{pid}]: {positioning}: terminated') except: pass - # print(f'[{pid}]: completed!') def killer_pmap(func: Callable, iterable: Iterable, cpus: Optional[int] = None, timeout: int = 10*60, @@ -140,7 +135,7 @@ def killer_pmap(func: Callable, iterable: Iterable, cpus: Optional[int] = None, mp.Process(target=_queue_mgr, args=(dill.dumps(func), q_in, q_out, timeout, pid, timeout_logfile)) for pid in range(cpus) ] - # print(f'Started {len(processes)} processes') + for proc in processes: proc.start() diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/scripts/DataPreparation_parallel.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/scripts/DataPreparation_parallel.py index e00b86a..12096c6 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/scripts/DataPreparation_parallel.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/scripts/DataPreparation_parallel.py @@ -165,7 +165,6 @@ def get_unique_indices(l): print("Resuming calculations, reading progress from "+progress_file) with open(progress_file) as f: progess_file_contents = [line for line in f.readlines()] - # print(progess_file_contents[-6:]) batch_resume = int(progess_file_contents[-7].split(":")[1]) + 1 index_resume = int(progess_file_contents[-3].split(":")[1]) batch_size_resume = int(progess_file_contents[-2].split(":")[1]) diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py index ad1644c..c501a33 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-08-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py @@ -13,14 +13,6 @@ def fix_i_match(matchobj): """ match = matchobj.group(0) return match.replace("i", "I") - # if int(match[1]) % 2 != 0: - # print("asdf") - # exponent = int(match[1]) // 2 - # m, m_name = match[0].split("_") - # if exponent == 1: - # return m+"2"+m_name - # else: - # return m+"2"+m_name + "**" + str(exponent) def combine_m_match(matchobj): diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/ampl_raw.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/ampl_raw.py index 115f7c1..ddac25f 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/ampl_raw.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/ampl_raw.py @@ -59,7 +59,6 @@ # break # sqampls_prefix = [] - # print("Loading squared amplitudes") # for exp in tqdm(sqampl_raw): # # simplified = sp.factor(exp) # worked best for simplification # # prefix = sympy_to_prefix(simplified) diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py index c931cee..7b954fa 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/dev/implement_hybrid_prefix_and_tokens_combiner.py @@ -95,22 +95,17 @@ def _queue_mgr(func_str: str, q_in: mp.Queue, q_out: mp.Queue, timeout: int, pid proc = mp.Process(target=_lemmiwinks, args=(func_str, (x,), {}, q_worker,)) proc.start() try: - # print(f'[{pid}]: {positioning}: getting') res = q_worker.get(timeout=timeout) - # print(f'[{pid}]: {positioning}: got') q_out.put((positioning, res)) except mpq.Empty: q_out.put((positioning, sp.sympify(x))) - # print(f'[{pid}]: {positioning}: timed out ({timeout}s)') with open(timeout_logfile, "a") as f: f.write("Timed out after "+str(timeout)+" seconds. Argument:" + x + "\n") finally: try: proc.terminate() - # print(f'[{pid}]: {positioning}: terminated') except: pass - # print(f'[{pid}]: completed!') def killer_pmap(func: Callable, iterable: Iterable, cpus: Optional[int] = None, timeout: int = 10*60, @@ -140,7 +135,7 @@ def killer_pmap(func: Callable, iterable: Iterable, cpus: Optional[int] = None, mp.Process(target=_queue_mgr, args=(dill.dumps(func), q_in, q_out, timeout, pid, timeout_logfile)) for pid in range(cpus) ] - # print(f'Started {len(processes)} processes') + for proc in processes: proc.start() diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/scripts/DataPreparation_parallel.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/scripts/DataPreparation_parallel.py index e00b86a..12096c6 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/scripts/DataPreparation_parallel.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/scripts/DataPreparation_parallel.py @@ -165,7 +165,6 @@ def get_unique_indices(l): print("Resuming calculations, reading progress from "+progress_file) with open(progress_file) as f: progess_file_contents = [line for line in f.readlines()] - # print(progess_file_contents[-6:]) batch_resume = int(progess_file_contents[-7].split(":")[1]) + 1 index_resume = int(progess_file_contents[-3].split(":")[1]) batch_size_resume = int(progess_file_contents[-2].split(":")[1]) diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/source/ExpressionsTokensCombiner.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/source/ExpressionsTokensCombiner.py index ad1644c..c501a33 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/source/ExpressionsTokensCombiner.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-09-15-QCD-DataPreparation/source/ExpressionsTokensCombiner.py @@ -13,14 +13,6 @@ def fix_i_match(matchobj): """ match = matchobj.group(0) return match.replace("i", "I") - # if int(match[1]) % 2 != 0: - # print("asdf") - # exponent = int(match[1]) // 2 - # m, m_name = match[0].split("_") - # if exponent == 1: - # return m+"2"+m_name - # else: - # return m+"2"+m_name + "**" + str(exponent) def combine_m_match(matchobj): diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-10-06-QED-DataPreparation/source/ExpressionsTokensCombiner.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-10-06-QED-DataPreparation/source/ExpressionsTokensCombiner.py index ad1644c..c501a33 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-10-06-QED-DataPreparation/source/ExpressionsTokensCombiner.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-10-06-QED-DataPreparation/source/ExpressionsTokensCombiner.py @@ -13,14 +13,6 @@ def fix_i_match(matchobj): """ match = matchobj.group(0) return match.replace("i", "I") - # if int(match[1]) % 2 != 0: - # print("asdf") - # exponent = int(match[1]) // 2 - # m, m_name = match[0].split("_") - # if exponent == 1: - # return m+"2"+m_name - # else: - # return m+"2"+m_name + "**" + str(exponent) def combine_m_match(matchobj): diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-11-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-11-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py index ad1644c..c501a33 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-11-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/2022-11-14-QED-DataPreparation/source/ExpressionsTokensCombiner.py @@ -13,14 +13,6 @@ def fix_i_match(matchobj): """ match = matchobj.group(0) return match.replace("i", "I") - # if int(match[1]) % 2 != 0: - # print("asdf") - # exponent = int(match[1]) // 2 - # m, m_name = match[0].split("_") - # if exponent == 1: - # return m+"2"+m_name - # else: - # return m+"2"+m_name + "**" + str(exponent) def combine_m_match(matchobj): diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/ampl_tree/source/ampl_to_tree.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/ampl_tree/source/ampl_to_tree.py index 02c1aa4..727f1e9 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/ampl_tree/source/ampl_to_tree.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/ampl_tree/source/ampl_to_tree.py @@ -591,14 +591,6 @@ def is_index(s: str): if __name__ == "__main__": - # with open(sqampls_file) as f: - # ampls = f.readlines() - # - # print(ampls[0]) - # print(ampls[100]) - # print(ampls[-1]) - # print(ampls[-20]) - with open(ampls_raw_file) as f: ampls_raw = f.readlines(100000) ampls_raw = [a[:-1] for a in ampls_raw] diff --git a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/tree/sympy_to_tree.py b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/tree/sympy_to_tree.py index 20f7d91..b900806 100644 --- a/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/tree/sympy_to_tree.py +++ b/SYMBA_HEP/SYMBAHEP_Transformers_Abdulhakim_Alnuqaydan+Marco_Knipfer/data_preprocessing/tree/sympy_to_tree.py @@ -125,13 +125,6 @@ def tree_to_sympy(tree, expression=None): else: node = tree._label op = operators_inv[node] - # num_args = operators_nargs[node] - # if num_args != len(tree): - # print("num args not len(tree):") - # ic(num_args) - # ic(len(tree)) - # ic(tree) - # assert num_args == len(tree) return op(*[tree_to_sympy(t) for t in tree]) return 0 diff --git a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/Tokenizers.py b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/Tokenizers.py index e4c9c91..a26c7d8 100644 --- a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/Tokenizers.py +++ b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/Tokenizers.py @@ -282,15 +282,12 @@ def decode_float(self,seq): if x[0] not in ["-", "+", "E", "N"]: return np.nan try: -# print(val) sign = 1 if val[0] == "+" else -1 mant = "" for x in val[1:-1]: mant += x[1:] mant = int(mant) -# print(mant) exp = int(val[-1][1:]) -# print(exp) value = sign * mant * (10 ** exp) value = float(value) except Exception: diff --git a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/model.py b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/model.py index b7ad56d..6a1511e 100644 --- a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/model.py +++ b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/DPOFineTuning/model.py @@ -89,11 +89,8 @@ def __init__(self, self.register_buffer('pos_embedding_1', self.pos_embedding) def forward(self, token_embedding): -# print(token_embedding.shape) token_embedding = token_embedding.to('cuda') self.pos_embedding = self.pos_embedding.to('cuda') -# token_embedding = token_embedding -# self.pos_embedding = self.pos_embedding return self.dropout(token_embedding + self.pos_embedding[:,:token_embedding.size(1), :]) diff --git a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/Tokenizers.py b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/Tokenizers.py index d357626..d556373 100644 --- a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/Tokenizers.py +++ b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/Tokenizers.py @@ -260,15 +260,12 @@ def decode_float(self,seq): if x[0] not in ["-", "+", "E", "N"]: return np.nan try: -# print(val) sign = 1 if val[0] == "+" else -1 mant = "" for x in val[1:-1]: mant += x[1:] mant = int(mant) -# print(mant) exp = int(val[-1][1:]) -# print(exp) value = sign * mant * (10 ** exp) value = float(value) except Exception: diff --git a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/model.py b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/model.py index 2fb776b..c9ac44d 100644 --- a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/model.py +++ b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/PIGP/model.py @@ -89,11 +89,8 @@ def __init__(self, self.register_buffer('pos_embedding_1', self.pos_embedding) def forward(self, token_embedding): -# print(token_embedding.shape) token_embedding = token_embedding.to('cuda') self.pos_embedding = self.pos_embedding.to('cuda') -# token_embedding = token_embedding -# self.pos_embedding = self.pos_embedding return self.dropout(token_embedding + self.pos_embedding[:,:token_embedding.size(1), :]) diff --git a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/Tokenizers.py b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/Tokenizers.py index 7936627..d137b3d 100644 --- a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/Tokenizers.py +++ b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/Tokenizers.py @@ -282,15 +282,12 @@ def decode_float(self,seq): if x[0] not in ["-", "+", "E", "N"]: return np.nan try: -# print(val) sign = 1 if val[0] == "+" else -1 mant = "" for x in val[1:-1]: mant += x[1:] mant = int(mant) -# print(mant) exp = int(val[-1][1:]) -# print(exp) value = sign * mant * (10 ** exp) value = float(value) except Exception: diff --git a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/model.py b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/model.py index a5b9ed9..bb6815f 100644 --- a/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/model.py +++ b/SYMBA_REG/SYMBREG_DPO+PIGP_Samyak_Jha/SymbolicDPO/model.py @@ -88,11 +88,8 @@ def __init__(self, self.register_buffer('pos_embedding_1', self.pos_embedding) def forward(self, token_embedding): -# print(token_embedding.shape) token_embedding = token_embedding.to('cuda') self.pos_embedding = self.pos_embedding.to('cuda') -# token_embedding = token_embedding -# self.pos_embedding = self.pos_embedding return self.dropout(token_embedding + self.pos_embedding[:,:token_embedding.size(1), :]) diff --git a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/gp/prepare_dataset.py b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/gp/prepare_dataset.py index 168bacc..b0de290 100644 --- a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/gp/prepare_dataset.py +++ b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/gp/prepare_dataset.py @@ -27,12 +27,10 @@ def get_args_parser(): # """ # total = len(list(iterable)) # # Progress Bar Printing Function -# print(total) # def printProgressBar (iteration): # percent = ("{0:." + str(decimals) + "f}").format(100 * (iteration / float(total))) # filledLength = int(length * iteration // total) # bar = fill * filledLength + '-' * (length - filledLength) -# print(f'\rProgress |{bar}| {percent}% Complete', end = printEnd) # # Initial Call # printProgressBar(0) # # Update Progress Bar @@ -40,7 +38,7 @@ def get_args_parser(): # yield item # printProgressBar(i + 1) # # Print New Line on Complete -# print() + def main(args): df = pd.read_csv(args.dataframe_path) diff --git a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/custom_gp.py b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/custom_gp.py index 8fec218..dfa50c6 100644 --- a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/custom_gp.py +++ b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/custom_gp.py @@ -153,7 +153,6 @@ def get_initial_population(pop_size, candidates, pset, toolbox): for i in range(pop_size - len(population)): random_candidate = toolbox.individual() population.append(random_candidate) - #print(population) return population @staticmethod diff --git a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/hybrid.py b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/hybrid.py index 9791a65..bc6b61f 100644 --- a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/hybrid.py +++ b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/hybrid/hybrid.py @@ -46,7 +46,6 @@ def validate_expressions(self, expressions, num_vars): for expression in expressions: try: expression = expression.type(torch.long)[0].tolist() - # print(f"{expression=}") expression = self.tokenizer.reverse_tokenize([expression[1:-1]])[0] #expression = prefix_to_sympy(expressions) if expression not in valid: @@ -98,10 +97,8 @@ def get_candidate_equations_single(self, X, y): num_vars = X.shape[1] x, num_array = self.format_data_for_transformer(X, y) expressions = self.generate_expressions(x, num_array) - #print(f"{expressions[0]=}") expressions = self.validate_expressions(expressions, num_vars) - #print(f"{expressions[0]=}") candidates = [] for expression in expressions: try: diff --git a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/engine/trainer.py b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/engine/trainer.py index 084640f..5ad5c8c 100644 --- a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/engine/trainer.py +++ b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/engine/trainer.py @@ -276,8 +276,6 @@ def test_seq_acc(self): y_pred = predictor.predict(src[0].unsqueeze(0), num_array[0].unsqueeze(0)) #only one example from each batch y_preds.append(y_pred.cpu().numpy()) y_true.append(np.trim_zeros(tgt[0])) - # print("pred", y_pred.cpu().tolist()) - # print("true", y_true[-1].tolist()) test_accuracy_seq = sequence_accuracy(y_true, y_preds) f= open(os.path.join(self.logs_dir, "score.txt"),"w+") diff --git a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/model/seq2seq.py b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/model/seq2seq.py index 418c822..14f1aae 100644 --- a/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/model/seq2seq.py +++ b/SYMBA_REG/SYMBREG_GP_Aryamaan_Thakur/algorithms/xval_transformers/model/seq2seq.py @@ -47,7 +47,6 @@ def __init__(self, vocab_size: int, emb_size): def forward(self, tokens, num_array): out = self.embedding(tokens.long()) * math.sqrt(self.emb_size) - # print("embeds", out.shape) out = self.layer_norm(out) out *= num_array.unsqueeze(-1) return out @@ -68,15 +67,12 @@ def forward(self, tokens, num_array): #dims = torch.tensor(out.size(1)*out.size(2)*out.size(3)) #mag_norm = 5/torch.sqrt(dims) #out += torch.zeros_like(out).uniform_(-mag_norm, mag_norm) - #print("embed", out.shape) - #print("num", num_array.shape) bs, n = out.shape[0], out.shape[1] out *= num_array.unsqueeze(-1) out = out.view(bs, n, -1) out = self.activation(self.fc1(out)) out = self.dropout(out) out = self.fc2(out) - #print("out", out.shape) return out class Model(nn.Module): diff --git a/SYMBA_REG/SymbolicGPT_Krish_Malik/src/embeddings/t_net_embeddings.py b/SYMBA_REG/SymbolicGPT_Krish_Malik/src/embeddings/t_net_embeddings.py index 3561d90..7b36bb8 100644 --- a/SYMBA_REG/SymbolicGPT_Krish_Malik/src/embeddings/t_net_embeddings.py +++ b/SYMBA_REG/SymbolicGPT_Krish_Malik/src/embeddings/t_net_embeddings.py @@ -126,6 +126,4 @@ def forward(self, x): print("✅ Embeddings saved to tnet_embeddings.json") # try: # files.download("tnet_embeddings.json") # Uncomment if in Google Colab -# print("✅ tnet_embeddings.json downloaded.") -# except Exception as e: -# print(f"❗ Could not download file (likely not in Google Colab environment): {e}") \ No newline at end of file +# except Exception as e: \ No newline at end of file