Upload folder using huggingface_hub (#7)

- Upload folder using huggingface_hub (5eb61aac1645ce750ef978bd9366a1b2a21de995)

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48e02db927d5123e3ae1c830db167827df07a85226a2c7faf31e17af0a0366b8
+oid sha256:4373ca3f29c182403e6a4581c43a863b8b58292755d198fe1fc834ccd084a6ce
 size 686525061

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a51a5a533fd7542d18145c6853de93f25ee3d709d53ec729f2684e7565340f3d
+oid sha256:de22ee000a33cc4c27096c9cfabd9c972fb27bbd412e0581dbdd28cd7229ca79
 size 343271789

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab62043be50b93d4eb28964be2d945176db3d64fe73ddd052a7656ba9141c683
+oid sha256:6dd3a816ab8628e6038ecf426e93a907752049203fbc39b63fcde557182a866f
 size 14575

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7be74ea536e79eb53304c8ad7665b1808b95e8a71eefc3a101477593346ec95
+oid sha256:b081c596a28e61a0c2c71b36c558646038f30339afa5e1f02c6c82300b54ce5e
 size 627

trainer_state.json

@@ -1,610 +1,247 @@
 {
-  "best_metric": 0.03899691626429558,
-  "best_model_checkpoint": "./vit-base-beans/checkpoint-720",
-  "epoch": 1.9565217391304348,
-  "global_step": 720,
+  "best_metric": 0.07841455936431885,
+  "best_model_checkpoint": "./vit-base-beans/checkpoint-280",
+  "epoch": 0.7608695652173914,
+  "global_step": 280,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.03,
-      "learning_rate": 0.0001972826086956522,
-      "loss": 0.5574,
+      "learning_rate": 0.00019322033898305085,
+      "loss": 0.6554,
       "step": 10
     },
     {
       "epoch": 0.05,
-      "learning_rate": 0.00019456521739130434,
-      "loss": 0.3197,
+      "learning_rate": 0.0001864406779661017,
+      "loss": 0.306,
       "step": 20
     },
     {
       "epoch": 0.08,
-      "learning_rate": 0.00019184782608695653,
-      "loss": 0.271,
+      "learning_rate": 0.00017966101694915257,
+      "loss": 0.267,
       "step": 30
     },
     {
       "epoch": 0.11,
-      "learning_rate": 0.0001891304347826087,
-      "loss": 0.2484,
+      "learning_rate": 0.00017288135593220342,
+      "loss": 0.2825,
       "step": 40
     },
     {
       "epoch": 0.11,
-      "eval_accuracy": 0.8801089918256131,
-      "eval_loss": 0.21790249645709991,
-      "eval_runtime": 16.4161,
-      "eval_samples_per_second": 44.712,
-      "eval_steps_per_second": 5.604,
+      "eval_accuracy": 0.9318801089918256,
+      "eval_loss": 0.2294050008058548,
+      "eval_runtime": 16.8062,
+      "eval_samples_per_second": 43.674,
+      "eval_steps_per_second": 5.474,
       "step": 40
     },
     {
       "epoch": 0.14,
-      "learning_rate": 0.0001864130434782609,
-      "loss": 0.2372,
+      "learning_rate": 0.00016610169491525423,
+      "loss": 0.3097,
       "step": 50
     },
     {
       "epoch": 0.16,
-      "learning_rate": 0.00018369565217391304,
-      "loss": 0.1945,
+      "learning_rate": 0.00015932203389830508,
+      "loss": 0.2772,
       "step": 60
     },
     {
       "epoch": 0.19,
-      "learning_rate": 0.00018097826086956522,
-      "loss": 0.2186,
+      "learning_rate": 0.00015254237288135592,
+      "loss": 0.1711,
       "step": 70
     },
     {
       "epoch": 0.22,
-      "learning_rate": 0.0001782608695652174,
-      "loss": 0.3033,
+      "learning_rate": 0.00014576271186440677,
+      "loss": 0.1483,
       "step": 80
     },
     {
       "epoch": 0.22,
-      "eval_accuracy": 0.8678474114441417,
-      "eval_loss": 0.28713682293891907,
-      "eval_runtime": 18.6159,
-      "eval_samples_per_second": 39.429,
-      "eval_steps_per_second": 4.942,
+      "eval_accuracy": 0.944141689373297,
+      "eval_loss": 0.16020944714546204,
+      "eval_runtime": 17.3798,
+      "eval_samples_per_second": 42.233,
+      "eval_steps_per_second": 5.294,
       "step": 80
     },
     {
       "epoch": 0.24,
-      "learning_rate": 0.00017554347826086956,
-      "loss": 0.2893,
+      "learning_rate": 0.00013898305084745764,
+      "loss": 0.2356,
       "step": 90
     },
     {
       "epoch": 0.27,
-      "learning_rate": 0.00017282608695652174,
-      "loss": 0.3135,
+      "learning_rate": 0.00013220338983050849,
+      "loss": 0.1144,
       "step": 100
     },
     {
       "epoch": 0.3,
-      "learning_rate": 0.00017010869565217392,
-      "loss": 0.2673,
+      "learning_rate": 0.00012542372881355933,
+      "loss": 0.2216,
       "step": 110
     },
     {
       "epoch": 0.33,
-      "learning_rate": 0.0001673913043478261,
-      "loss": 0.2709,
+      "learning_rate": 0.00011864406779661017,
+      "loss": 0.2209,
       "step": 120
     },
     {
       "epoch": 0.33,
-      "eval_accuracy": 0.9073569482288828,
-      "eval_loss": 0.23233647644519806,
-      "eval_runtime": 16.7819,
-      "eval_samples_per_second": 43.738,
-      "eval_steps_per_second": 5.482,
+      "eval_accuracy": 0.829700272479564,
+      "eval_loss": 0.29767370223999023,
+      "eval_runtime": 17.13,
+      "eval_samples_per_second": 42.849,
+      "eval_steps_per_second": 5.371,
       "step": 120
     },
     {
       "epoch": 0.35,
-      "learning_rate": 0.00016467391304347828,
-      "loss": 0.2171,
+      "learning_rate": 0.00011186440677966102,
+      "loss": 0.2153,
       "step": 130
     },
     {
       "epoch": 0.38,
-      "learning_rate": 0.00016195652173913046,
-      "loss": 0.2393,
+      "learning_rate": 0.00010508474576271188,
+      "loss": 0.1693,
       "step": 140
     },
     {
       "epoch": 0.41,
-      "learning_rate": 0.00015923913043478262,
-      "loss": 0.1934,
+      "learning_rate": 9.830508474576272e-05,
+      "loss": 0.1625,
       "step": 150
     },
     {
       "epoch": 0.43,
-      "learning_rate": 0.0001565217391304348,
-      "loss": 0.1878,
+      "learning_rate": 9.152542372881357e-05,
+      "loss": 0.1384,
       "step": 160
     },
     {
       "epoch": 0.43,
-      "eval_accuracy": 0.9318801089918256,
-      "eval_loss": 0.1636728048324585,
-      "eval_runtime": 16.6111,
-      "eval_samples_per_second": 44.187,
-      "eval_steps_per_second": 5.538,
+      "eval_accuracy": 0.9645776566757494,
+      "eval_loss": 0.10578873753547668,
+      "eval_runtime": 16.6083,
+      "eval_samples_per_second": 44.195,
+      "eval_steps_per_second": 5.539,
       "step": 160
     },
     {
       "epoch": 0.46,
-      "learning_rate": 0.00015380434782608698,
-      "loss": 0.2186,
+      "learning_rate": 8.474576271186441e-05,
+      "loss": 0.1371,
       "step": 170
     },
     {
       "epoch": 0.49,
-      "learning_rate": 0.00015108695652173913,
-      "loss": 0.1952,
+      "learning_rate": 7.796610169491526e-05,
+      "loss": 0.1355,
       "step": 180
     },
     {
       "epoch": 0.52,
-      "learning_rate": 0.0001483695652173913,
-      "loss": 0.1546,
+      "learning_rate": 7.11864406779661e-05,
+      "loss": 0.1557,
       "step": 190
     },
     {
       "epoch": 0.54,
-      "learning_rate": 0.0001456521739130435,
-      "loss": 0.1655,
+      "learning_rate": 6.440677966101695e-05,
+      "loss": 0.1551,
       "step": 200
     },
     {
       "epoch": 0.54,
-      "eval_accuracy": 0.8678474114441417,
-      "eval_loss": 0.24226485192775726,
-      "eval_runtime": 17.096,
-      "eval_samples_per_second": 42.934,
-      "eval_steps_per_second": 5.381,
+      "eval_accuracy": 0.9114441416893733,
+      "eval_loss": 0.1725204885005951,
+      "eval_runtime": 17.0671,
+      "eval_samples_per_second": 43.007,
+      "eval_steps_per_second": 5.39,
       "step": 200
     },
     {
       "epoch": 0.57,
-      "learning_rate": 0.00014293478260869567,
-      "loss": 0.1843,
+      "learning_rate": 5.76271186440678e-05,
+      "loss": 0.2091,
       "step": 210
     },
     {
       "epoch": 0.6,
-      "learning_rate": 0.00014021739130434783,
-      "loss": 0.12,
+      "learning_rate": 5.0847457627118643e-05,
+      "loss": 0.1313,
       "step": 220
     },
     {
       "epoch": 0.62,
-      "learning_rate": 0.0001375,
-      "loss": 0.1686,
+      "learning_rate": 4.4067796610169495e-05,
+      "loss": 0.1056,
       "step": 230
     },
     {
       "epoch": 0.65,
-      "learning_rate": 0.0001347826086956522,
-      "loss": 0.2393,
+      "learning_rate": 3.728813559322034e-05,
+      "loss": 0.1608,
       "step": 240
     },
     {
       "epoch": 0.65,
-      "eval_accuracy": 0.94141689373297,
-      "eval_loss": 0.1487017571926117,
-      "eval_runtime": 16.7397,
-      "eval_samples_per_second": 43.848,
-      "eval_steps_per_second": 5.496,
+      "eval_accuracy": 0.9673024523160763,
+      "eval_loss": 0.09211871027946472,
+      "eval_runtime": 16.6063,
+      "eval_samples_per_second": 44.2,
+      "eval_steps_per_second": 5.54,
       "step": 240
     },
     {
       "epoch": 0.68,
-      "learning_rate": 0.00013206521739130434,
-      "loss": 0.2159,
+      "learning_rate": 3.050847457627119e-05,
+      "loss": 0.0761,
       "step": 250
     },
     {
       "epoch": 0.71,
-      "learning_rate": 0.00012934782608695652,
-      "loss": 0.1608,
+      "learning_rate": 2.3728813559322036e-05,
+      "loss": 0.1044,
       "step": 260
     },
     {
       "epoch": 0.73,
-      "learning_rate": 0.0001266304347826087,
-      "loss": 0.1508,
+      "learning_rate": 1.694915254237288e-05,
+      "loss": 0.1141,
       "step": 270
     },
     {
       "epoch": 0.76,
-      "learning_rate": 0.00012391304347826086,
-      "loss": 0.1432,
+      "learning_rate": 1.016949152542373e-05,
+      "loss": 0.12,
       "step": 280
     },
     {
       "epoch": 0.76,
-      "eval_accuracy": 0.9209809264305178,
-      "eval_loss": 0.1679653823375702,
-      "eval_runtime": 19.1578,
-      "eval_samples_per_second": 38.313,
-      "eval_steps_per_second": 4.802,
-      "step": 280
-    },
-    {
-      "epoch": 0.79,
-      "learning_rate": 0.00012119565217391304,
-      "loss": 0.1852,
-      "step": 290
-    },
-    {
-      "epoch": 0.82,
-      "learning_rate": 0.00011847826086956522,
-      "loss": 0.1845,
-      "step": 300
-    },
-    {
-      "epoch": 0.84,
-      "learning_rate": 0.00011576086956521739,
-      "loss": 0.1238,
-      "step": 310
-    },
-    {
-      "epoch": 0.87,
-      "learning_rate": 0.00011304347826086956,
-      "loss": 0.0986,
-      "step": 320
-    },
-    {
-      "epoch": 0.87,
-      "eval_accuracy": 0.9604904632152589,
-      "eval_loss": 0.11554770916700363,
-      "eval_runtime": 16.799,
-      "eval_samples_per_second": 43.693,
-      "eval_steps_per_second": 5.477,
-      "step": 320
-    },
-    {
-      "epoch": 0.9,
-      "learning_rate": 0.00011032608695652174,
-      "loss": 0.1781,
-      "step": 330
-    },
-    {
-      "epoch": 0.92,
-      "learning_rate": 0.0001076086956521739,
-      "loss": 0.1229,
-      "step": 340
-    },
-    {
-      "epoch": 0.95,
-      "learning_rate": 0.0001048913043478261,
-      "loss": 0.1238,
-      "step": 350
-    },
-    {
-      "epoch": 0.98,
-      "learning_rate": 0.00010217391304347828,
-      "loss": 0.0729,
-      "step": 360
-    },
-    {
-      "epoch": 0.98,
-      "eval_accuracy": 0.9577656675749319,
-      "eval_loss": 0.10841722786426544,
-      "eval_runtime": 17.3667,
-      "eval_samples_per_second": 42.265,
-      "eval_steps_per_second": 5.297,
-      "step": 360
-    },
-    {
-      "epoch": 1.01,
-      "learning_rate": 9.945652173913043e-05,
-      "loss": 0.1156,
-      "step": 370
-    },
-    {
-      "epoch": 1.03,
-      "learning_rate": 9.673913043478261e-05,
-      "loss": 0.0934,
-      "step": 380
-    },
-    {
-      "epoch": 1.06,
-      "learning_rate": 9.402173913043478e-05,
-      "loss": 0.082,
-      "step": 390
-    },
-    {
-      "epoch": 1.09,
-      "learning_rate": 9.130434782608696e-05,
-      "loss": 0.0553,
-      "step": 400
-    },
-    {
-      "epoch": 1.09,
-      "eval_accuracy": 0.9632152588555858,
-      "eval_loss": 0.09444452077150345,
-      "eval_runtime": 16.7805,
-      "eval_samples_per_second": 43.741,
-      "eval_steps_per_second": 5.483,
-      "step": 400
-    },
-    {
-      "epoch": 1.11,
-      "learning_rate": 8.858695652173914e-05,
-      "loss": 0.1067,
-      "step": 410
-    },
-    {
-      "epoch": 1.14,
-      "learning_rate": 8.586956521739131e-05,
-      "loss": 0.0544,
-      "step": 420
-    },
-    {
-      "epoch": 1.17,
-      "learning_rate": 8.315217391304349e-05,
-      "loss": 0.1047,
-      "step": 430
-    },
-    {
-      "epoch": 1.2,
-      "learning_rate": 8.043478260869566e-05,
-      "loss": 0.0395,
-      "step": 440
-    },
-    {
-      "epoch": 1.2,
-      "eval_accuracy": 0.9673024523160763,
-      "eval_loss": 0.07385484129190445,
-      "eval_runtime": 17.3745,
-      "eval_samples_per_second": 42.246,
-      "eval_steps_per_second": 5.295,
-      "step": 440
-    },
-    {
-      "epoch": 1.22,
-      "learning_rate": 7.771739130434783e-05,
-      "loss": 0.0566,
-      "step": 450
-    },
-    {
-      "epoch": 1.25,
-      "learning_rate": 7.500000000000001e-05,
-      "loss": 0.0625,
-      "step": 460
-    },
-    {
-      "epoch": 1.28,
-      "learning_rate": 7.228260869565217e-05,
-      "loss": 0.0961,
-      "step": 470
-    },
-    {
-      "epoch": 1.3,
-      "learning_rate": 6.956521739130436e-05,
-      "loss": 0.0178,
-      "step": 480
-    },
-    {
-      "epoch": 1.3,
-      "eval_accuracy": 0.9673024523160763,
-      "eval_loss": 0.10838904976844788,
-      "eval_runtime": 16.8091,
-      "eval_samples_per_second": 43.667,
-      "eval_steps_per_second": 5.473,
-      "step": 480
-    },
-    {
-      "epoch": 1.33,
-      "learning_rate": 6.684782608695652e-05,
-      "loss": 0.0641,
-      "step": 490
-    },
-    {
-      "epoch": 1.36,
-      "learning_rate": 6.413043478260869e-05,
-      "loss": 0.0497,
-      "step": 500
-    },
-    {
-      "epoch": 1.39,
-      "learning_rate": 6.141304347826087e-05,
-      "loss": 0.0364,
-      "step": 510
-    },
-    {
-      "epoch": 1.41,
-      "learning_rate": 5.869565217391305e-05,
-      "loss": 0.0273,
-      "step": 520
-    },
-    {
-      "epoch": 1.41,
-      "eval_accuracy": 0.9727520435967303,
-      "eval_loss": 0.07216904312372208,
-      "eval_runtime": 17.0234,
-      "eval_samples_per_second": 43.117,
-      "eval_steps_per_second": 5.404,
-      "step": 520
-    },
-    {
-      "epoch": 1.44,
-      "learning_rate": 5.5978260869565226e-05,
-      "loss": 0.0558,
-      "step": 530
-    },
-    {
-      "epoch": 1.47,
-      "learning_rate": 5.32608695652174e-05,
-      "loss": 0.0379,
-      "step": 540
-    },
-    {
-      "epoch": 1.49,
-      "learning_rate": 5.054347826086957e-05,
-      "loss": 0.0569,
-      "step": 550
-    },
-    {
-      "epoch": 1.52,
-      "learning_rate": 4.782608695652174e-05,
-      "loss": 0.0273,
-      "step": 560
-    },
-    {
-      "epoch": 1.52,
       "eval_accuracy": 0.9754768392370572,
-      "eval_loss": 0.0689365416765213,
-      "eval_runtime": 16.7035,
-      "eval_samples_per_second": 43.943,
-      "eval_steps_per_second": 5.508,
-      "step": 560
-    },
-    {
-      "epoch": 1.55,
-      "learning_rate": 4.5108695652173916e-05,
-      "loss": 0.0087,
-      "step": 570
-    },
-    {
-      "epoch": 1.58,
-      "learning_rate": 4.239130434782609e-05,
-      "loss": 0.0347,
-      "step": 580
-    },
-    {
-      "epoch": 1.6,
-      "learning_rate": 3.9673913043478264e-05,
-      "loss": 0.0196,
-      "step": 590
-    },
-    {
-      "epoch": 1.63,
-      "learning_rate": 3.695652173913043e-05,
-      "loss": 0.0271,
-      "step": 600
-    },
-    {
-      "epoch": 1.63,
-      "eval_accuracy": 0.9795640326975477,
-      "eval_loss": 0.07317620515823364,
-      "eval_runtime": 16.8063,
-      "eval_samples_per_second": 43.674,
-      "eval_steps_per_second": 5.474,
-      "step": 600
-    },
-    {
-      "epoch": 1.66,
-      "learning_rate": 3.423913043478261e-05,
-      "loss": 0.0951,
-      "step": 610
-    },
-    {
-      "epoch": 1.68,
-      "learning_rate": 3.152173913043479e-05,
-      "loss": 0.0726,
-      "step": 620
-    },
-    {
-      "epoch": 1.71,
-      "learning_rate": 2.8804347826086957e-05,
-      "loss": 0.01,
-      "step": 630
-    },
-    {
-      "epoch": 1.74,
-      "learning_rate": 2.608695652173913e-05,
-      "loss": 0.0294,
-      "step": 640
-    },
-    {
-      "epoch": 1.74,
-      "eval_accuracy": 0.9822888283378747,
-      "eval_loss": 0.057586103677749634,
-      "eval_runtime": 17.373,
-      "eval_samples_per_second": 42.249,
-      "eval_steps_per_second": 5.296,
-      "step": 640
-    },
-    {
-      "epoch": 1.77,
-      "learning_rate": 2.3369565217391306e-05,
-      "loss": 0.017,
-      "step": 650
-    },
-    {
-      "epoch": 1.79,
-      "learning_rate": 2.065217391304348e-05,
-      "loss": 0.0271,
-      "step": 660
-    },
-    {
-      "epoch": 1.82,
-      "learning_rate": 1.793478260869565e-05,
-      "loss": 0.0572,
-      "step": 670
-    },
-    {
-      "epoch": 1.85,
-      "learning_rate": 1.5217391304347828e-05,
-      "loss": 0.0449,
-      "step": 680
-    },
-    {
-      "epoch": 1.85,
-      "eval_accuracy": 0.9877384196185286,
-      "eval_loss": 0.04121558368206024,
-      "eval_runtime": 16.5738,
-      "eval_samples_per_second": 44.287,
-      "eval_steps_per_second": 5.551,
-      "step": 680
-    },
-    {
-      "epoch": 1.88,
-      "learning_rate": 1.25e-05,
-      "loss": 0.0261,
-      "step": 690
-    },
-    {
-      "epoch": 1.9,
-      "learning_rate": 9.782608695652175e-06,
-      "loss": 0.0091,
-      "step": 700
-    },
-    {
-      "epoch": 1.93,
-      "learning_rate": 7.065217391304347e-06,
-      "loss": 0.0799,
-      "step": 710
-    },
-    {
-      "epoch": 1.96,
-      "learning_rate": 4.347826086956522e-06,
-      "loss": 0.0076,
-      "step": 720
-    },
-    {
-      "epoch": 1.96,
-      "eval_accuracy": 0.9877384196185286,
-      "eval_loss": 0.03899691626429558,
-      "eval_runtime": 17.1545,
-      "eval_samples_per_second": 42.788,
-      "eval_steps_per_second": 5.363,
-      "step": 720
+      "eval_loss": 0.07841455936431885,
+      "eval_runtime": 16.7459,
+      "eval_samples_per_second": 43.832,
+      "eval_steps_per_second": 5.494,
+      "step": 280
     }
   ],
-  "max_steps": 736,
-  "num_train_epochs": 2,
-  "total_flos": 8.915533311547699e+17,
+  "max_steps": 295,
+  "num_train_epochs": 1,
+  "total_flos": 3.471672249955123e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29d42765cdc82ba9c501fd08fe3051fcacc5260575abfef3749cf9ec8c08e6be
+oid sha256:f07575c36fbac5b1ce76fcb42fdadefe90c95ce5be20b51ab48ba4f71f7e83f5
 size 3963