[init] init datasets

llm-srbench/bio_pop_growth/BPG0/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.9539969835279999 * (1 - P / 96.90688297671034) * P + 0.9539969835279999 * P ** 0.3333333333333333

llm-srbench/bio_pop_growth/BPG0/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01c47943843a79996e046cf1275d8b812dc4784f7a812313c33c34cbf81d25f7
+size 29349

llm-srbench/bio_pop_growth/BPG0/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG0
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.098819732666016]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 101.43367767333984]
+      ood_range: [101.26644897460938, 101.43058013916016]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [-0.0711578950285911, 26.604358673095703]
+    ood_range: [-0.06796845048666, 0.100564457476139]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG0/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93c3ce07e742819126077fdaea15cdde1aa73ef9f47e5e7dc338bd0e19fa50e3
+size 29596

llm-srbench/bio_pop_growth/BPG0/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:debb017346c6cb2a9482d3f13b1265eac8a23affacb9cd3a64ee128fd26829f8
+size 121465

llm-srbench/bio_pop_growth/BPG0/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54b799d9e68b5ea3f63d751da90cedcf1c863e4278676a5a938aceb70ac5fbfe
+size 196525

llm-srbench/bio_pop_growth/BPG0/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd702e003193a276823d6e376c7d664d0073924d4075ea7a1d390b6f80fe262c
+size 28075

llm-srbench/bio_pop_growth/BPG1/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.3156057016002752 * P * np.exp(-0.05405074842498068 * t) + 0.3156057016002752 * P**2 / (9.870854086995406 * P + 1)

llm-srbench/bio_pop_growth/BPG1/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6f0ff146586742bb3e03633d1ca8f1c206dfb3f9385ea611922084892f8d360
+size 28215

llm-srbench/bio_pop_growth/BPG1/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG1
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.087017059326165]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 996.0541381835936]
+      ood_range: [997.2052612304688, 1311.25830078125]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [0.3280320763587951, 48.74041748046875]
+    ood_range: [48.775184631347656, 58.080604553222656]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG1/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ede72550f415b40592a86109ba46722611b203c7530a05843923d24be286dcbe
+size 27874

llm-srbench/bio_pop_growth/BPG1/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22b36c295a46ebc0d4de00a9d12ab093aeff72ef2877fc4bce8440ff1def92fa
+size 123056

llm-srbench/bio_pop_growth/BPG1/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9816f249f24220b78daf3c409f094361ddf3df0d2b74c6d61efebbc4562663a0
+size 191963

llm-srbench/bio_pop_growth/BPG1/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3deb61204dc215e1ea317690f3221df25a4c6018a7adcd421ba2cd74c4fb849
+size 27417

llm-srbench/bio_pop_growth/BPG10/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.10100810033209855 * P**0.333333333333333 + 0.10100810033209855 * P

llm-srbench/bio_pop_growth/BPG10/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0131b3c66aade84a7e0a0fa47565e71c97f5cbd82425d120471ce6125d0ed187
+size 28390

llm-srbench/bio_pop_growth/BPG10/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG10
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.087017059326165]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 590.2333374023438]
+      ood_range: [591.6644287109375, 1080.2445068359375]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [0.2020162045955658, 60.46563720703125]
+    ood_range: [60.61087417602539, 110.1498565673828]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG10/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7baea24e2c77870fcec4c92340b66edbe0668de2547a30ee1056069097d9862
+size 27803

llm-srbench/bio_pop_growth/BPG10/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b083961b7d7107b154c9d281cda311f46a2e8288757903182eff361fe0b82a97
+size 126062

llm-srbench/bio_pop_growth/BPG10/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:321538db9b9943af63c7f354a94dcbe5e837c115529012b3efb0284cb7b95db7
+size 192720

llm-srbench/bio_pop_growth/BPG10/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a044ccde58a46e7292213ef3ae6f8d7e2ab1125a0c64b7d65590a338b6653c62
+size 27547

llm-srbench/bio_pop_growth/BPG11/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.7124493718672418*(1 - P/68.90600928530216)*P + 0.7124493718672418*P**0.333333333333333 + 0.7124493718672418*P*np.exp(-0.03459335742983884*t)

llm-srbench/bio_pop_growth/BPG11/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57af93a6c37afacacf5f2ae5ccf621f951f6118ca1cf4b9dee44e05fb6549ad8
+size 28998

llm-srbench/bio_pop_growth/BPG11/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG11
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.098819732666016]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 125.01062774658205]
+      ood_range: [81.60037994384764, 83.52909088134764]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [-1.6351751089096067, 45.735366821289055]
+    ood_range: [-0.3851576149463653, -0.3096195757389068]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG11/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47004537d37cb14a43df51a2d3e5b340288b8f0fe23caf517bed4f554cc2329b
+size 29237

llm-srbench/bio_pop_growth/BPG11/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d14b223d43c1fcf61de8381d7d6a14676b70a3ae508e809dd2cb1ae9be2cc32
+size 125562

llm-srbench/bio_pop_growth/BPG11/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33cfde67dd1430f2e926b2c95a528add2c77a66ac6e2556e20478104945fd636
+size 196566

llm-srbench/bio_pop_growth/BPG11/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddd635688175fff047cde4dc66150482b864a88340484ddd6d5ce80bfa8a4cc3
+size 28062

llm-srbench/bio_pop_growth/BPG12/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.8770002846146427 * P * np.sin(0.5671736066225963 * t) + 0.7013915186995147 * (1 - P / 65.7541311408653) * P

llm-srbench/bio_pop_growth/BPG12/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dd754fc64f4efe6a5fd5e0ce14ea9682fb5582f7af714fceec66986c387efe4
+size 28601

llm-srbench/bio_pop_growth/BPG12/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG12
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.087017059326165]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 136.5778350830078]
+      ood_range: [17.544462203979492, 136.52293395996094]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [-33.35061264038086, 50.45180892944336]
+    ood_range: [-23.597190856933597, 50.378955841064446]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG12/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:389bce1f960522a3cf75b333401dff3773c1c6023e6558dca27cf63bf738b966
+size 28427

llm-srbench/bio_pop_growth/BPG12/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d056c7443eb30973f3da21675ff24810e95b265c4ddf50957b23940b7426965d
+size 124198

llm-srbench/bio_pop_growth/BPG12/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46ea366723d0bfaa6cad5257c26831f1e55f82af87cd64f613a565fd024a3c9d
+size 194501

llm-srbench/bio_pop_growth/BPG12/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3b4d791d03d94fac11f316dc4d657c7e2109c3f060f7e5c097118b527980a83
+size 27835

llm-srbench/bio_pop_growth/BPG13/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.20107390997009728*(-1 + P/5.636576533003584)*(1 - P/10.232561336048414)*P + 0.20107390997009728*P + 0.20107390997009728*P/(1 + np.exp(-5.636576533003584*(-0.6341847122108325 + P)))

llm-srbench/bio_pop_growth/BPG13/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd7f0cd165921e717141caadc563bfa541d208d7a6b771bd8ac4fd86622a2b4a
+size 29631

llm-srbench/bio_pop_growth/BPG13/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG13
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.098819732666016]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 18.93589401245117]
+      ood_range: [18.906984329223636, 18.923608779907227]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [-0.0260864794254302, 4.357088088989258]
+    ood_range: [-0.008245201781392, 0.0158262960612773]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG13/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09994520af3ef190a2b12432513f6acbf5c05c06b06245b6d69f8479159a3be1
+size 29959

llm-srbench/bio_pop_growth/BPG13/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a894e6a06fb250700ed366fe51011ad933b6c7a75b2db262482ebfa0b78b3af
+size 123808

llm-srbench/bio_pop_growth/BPG13/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbf9aa76d6d000c0cd95b79d343187c978c2566d22e5f68d3bf91e73186652bd
+size 196003

llm-srbench/bio_pop_growth/BPG13/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a637c7d8bff48e315a76515b3fb9378fd2b8af76a98c6c8330d1ce837d00411f
+size 28010

llm-srbench/bio_pop_growth/BPG14/formula.py

@@ -0,0 +1,2 @@
+def dP_dt(t, P):
+    return 0.11423898258902421*(1 - P/40.69810220267985)*P + 0.11423898258902421*(1 - np.exp(-0.08368363514073095*P))*P + 0.11423898258902421*P*np.exp(-0.08368363514073095*t)

llm-srbench/bio_pop_growth/BPG14/id_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a5db4a2de38a706ee3e9101df47ec41aa42f5acc05334f5e1541b7142ccfd9b
+size 28835

llm-srbench/bio_pop_growth/BPG14/metadata.yaml

@@ -0,0 +1,45 @@
+dataset:
+  name: llm-srbench__bio_pop_growth__BPG14
+  description: 
+  splits:
+    train:
+      file: train.csv
+      samples: 3500
+      ratio: 7
+    valid:
+      file: valid.csv
+      samples: 500
+      ratio: 1
+    id_test:
+      file: id_test.csv
+      samples: 500
+      ratio: 1
+    ood_test:
+      file: ood_test.csv
+      samples: 500
+      ratio: 1
+  ground_truth_formula:
+    file: formula.py
+  features:
+    - name: t
+      type: continuous
+      train_range: [1.0, 54.098819732666016]
+      ood_range: [54.11062240600586, 60.0]
+    - name: P
+      type: continuous
+      train_range: [1.0, 82.64566802978516]
+      ood_range: [81.76631927490233, 82.00167083740233]
+
+  target:
+    name: dP_dt
+    type: continuous
+    train_range: [-0.0608358271420001, 5.34712028503418]
+    ood_range: [-0.0479926317930221, -0.0332890152931213]
+
+  license: MIT
+  citation:
+    arXiv: 2504.10415
+    title: LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
+  resources:
+    - https://github.com/deep-symbolic-mathematics/llm-srbench
+    - https://huggingface.co/datasets/nnheui/llm-srbench

llm-srbench/bio_pop_growth/BPG14/ood_test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:991ae422cd3193e892deec5294675d2023ce503486c4a29bc380b69ba608c574
+size 29756

llm-srbench/bio_pop_growth/BPG14/package.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a9fd100bad587004ea163cfd9bf96dafba880649856fd496df0a9d156beffa3
+size 126652

llm-srbench/bio_pop_growth/BPG14/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:befa71562001d40c85fe08114acf2698acbe48d3ea8aea52998eac73e163229d
+size 193278

llm-srbench/bio_pop_growth/BPG14/valid.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a6f8c4fea11d0ce41dd533b33eae18b601c1eb48ead3347ef4ec1258668910d
+size 27675