Upload folder using huggingface_hub
Browse files- ksplit1/checkpoint-1473/config.json +26 -0
- ksplit1/checkpoint-1473/model.safetensors +3 -0
- ksplit1/checkpoint-1473/optimizer.pt +3 -0
- ksplit1/checkpoint-1473/rng_state.pth +3 -0
- ksplit1/checkpoint-1473/scheduler.pt +3 -0
- ksplit1/checkpoint-1473/trainer_state.json +112 -0
- ksplit1/checkpoint-1473/training_args.bin +3 -0
- ksplit1/checkpoint-1964/config.json +26 -0
- ksplit1/checkpoint-1964/model.safetensors +3 -0
- ksplit1/checkpoint-1964/optimizer.pt +3 -0
- ksplit1/checkpoint-1964/rng_state.pth +3 -0
- ksplit1/checkpoint-1964/scheduler.pt +3 -0
- ksplit1/checkpoint-1964/trainer_state.json +143 -0
- ksplit1/checkpoint-1964/training_args.bin +3 -0
- ksplit1/checkpoint-491/config.json +26 -0
- ksplit1/checkpoint-491/model.safetensors +3 -0
- ksplit1/checkpoint-491/optimizer.pt +3 -0
- ksplit1/checkpoint-491/rng_state.pth +3 -0
- ksplit1/checkpoint-491/scheduler.pt +3 -0
- ksplit1/checkpoint-491/trainer_state.json +57 -0
- ksplit1/checkpoint-491/training_args.bin +3 -0
- ksplit1/checkpoint-982/config.json +26 -0
- ksplit1/checkpoint-982/model.safetensors +3 -0
- ksplit1/checkpoint-982/optimizer.pt +3 -0
- ksplit1/checkpoint-982/rng_state.pth +3 -0
- ksplit1/checkpoint-982/scheduler.pt +3 -0
- ksplit1/checkpoint-982/trainer_state.json +88 -0
- ksplit1/checkpoint-982/training_args.bin +3 -0
- ksplit1/config.json +26 -0
- ksplit1/many_control_pred_dict.pkl +3 -0
- ksplit1/model.safetensors +3 -0
- ksplit1/training_args.bin +3 -0
- many_control_eval_metrics_dict.pkl +3 -0
ksplit1/checkpoint-1473/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": ".venv/src/geneformer/gf-12L-95M-i4096/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.02,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "relu",
|
| 9 |
+
"hidden_dropout_prob": 0.02,
|
| 10 |
+
"hidden_size": 512,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1024,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 8,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.41.0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 20275
|
| 26 |
+
}
|
ksplit1/checkpoint-1473/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d807006345ab2fd98c4cc831ade3cc5258c2cd09bc460875c706e6d8acc5356b
|
| 3 |
+
size 151931712
|
ksplit1/checkpoint-1473/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11c9ece87661e66fd50edbdd3fbc204895ff73e403842ab92cab3ccef4243cb8
|
| 3 |
+
size 236642362
|
ksplit1/checkpoint-1473/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b95cd18f54e59353547a684ca8e9b8bcf53f2488894a074868f02bff6dd09ee
|
| 3 |
+
size 14244
|
ksplit1/checkpoint-1473/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:022d229cb0f7e505b2fe82984b4618a208917b6938b3cfe8a33bfe771e862112
|
| 3 |
+
size 1064
|
ksplit1/checkpoint-1473/trainer_state.json
ADDED
|
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3622211515903473,
|
| 3 |
+
"best_model_checkpoint": "finetuned/250416_geneformer_cellClassifier_many_control/ksplit1/checkpoint-982",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1473,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.39918533604887985,
|
| 13 |
+
"grad_norm": 1.4352002143859863,
|
| 14 |
+
"learning_rate": 1.9869391106802152e-05,
|
| 15 |
+
"loss": 0.3426,
|
| 16 |
+
"step": 196
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.7983706720977597,
|
| 20 |
+
"grad_norm": 4.147289276123047,
|
| 21 |
+
"learning_rate": 1.8813246895441423e-05,
|
| 22 |
+
"loss": 0.1893,
|
| 23 |
+
"step": 392
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.8260050659219328,
|
| 28 |
+
"eval_loss": 0.3700631558895111,
|
| 29 |
+
"eval_macro_f1": 0.7135959847549435,
|
| 30 |
+
"eval_runtime": 163.0259,
|
| 31 |
+
"eval_samples_per_second": 94.445,
|
| 32 |
+
"eval_steps_per_second": 1.969,
|
| 33 |
+
"step": 491
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.1975560081466394,
|
| 37 |
+
"grad_norm": 8.350934028625488,
|
| 38 |
+
"learning_rate": 1.6804080338412108e-05,
|
| 39 |
+
"loss": 0.1337,
|
| 40 |
+
"step": 588
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.5967413441955194,
|
| 44 |
+
"grad_norm": 17.019676208496094,
|
| 45 |
+
"learning_rate": 1.4059153092823729e-05,
|
| 46 |
+
"loss": 0.1051,
|
| 47 |
+
"step": 784
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.9959266802443993,
|
| 51 |
+
"grad_norm": 3.5994579792022705,
|
| 52 |
+
"learning_rate": 1.0875288455846521e-05,
|
| 53 |
+
"loss": 0.0962,
|
| 54 |
+
"step": 980
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.0,
|
| 58 |
+
"eval_accuracy": 0.850360459829837,
|
| 59 |
+
"eval_loss": 0.3622211515903473,
|
| 60 |
+
"eval_macro_f1": 0.7522559675724617,
|
| 61 |
+
"eval_runtime": 271.2733,
|
| 62 |
+
"eval_samples_per_second": 56.758,
|
| 63 |
+
"eval_steps_per_second": 1.183,
|
| 64 |
+
"step": 982
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 2.395112016293279,
|
| 68 |
+
"grad_norm": 7.739411354064941,
|
| 69 |
+
"learning_rate": 7.596774313729619e-06,
|
| 70 |
+
"loss": 0.0851,
|
| 71 |
+
"step": 1176
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 2.794297352342159,
|
| 75 |
+
"grad_norm": 4.228972434997559,
|
| 76 |
+
"learning_rate": 4.578133497286259e-06,
|
| 77 |
+
"loss": 0.0723,
|
| 78 |
+
"step": 1372
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.0,
|
| 82 |
+
"eval_accuracy": 0.8550366954601546,
|
| 83 |
+
"eval_loss": 0.3901084065437317,
|
| 84 |
+
"eval_macro_f1": 0.7393773894378316,
|
| 85 |
+
"eval_runtime": 234.4812,
|
| 86 |
+
"eval_samples_per_second": 65.664,
|
| 87 |
+
"eval_steps_per_second": 1.369,
|
| 88 |
+
"step": 1473
|
| 89 |
+
}
|
| 90 |
+
],
|
| 91 |
+
"logging_steps": 196,
|
| 92 |
+
"max_steps": 1964,
|
| 93 |
+
"num_input_tokens_seen": 0,
|
| 94 |
+
"num_train_epochs": 4,
|
| 95 |
+
"save_steps": 500,
|
| 96 |
+
"stateful_callbacks": {
|
| 97 |
+
"TrainerControl": {
|
| 98 |
+
"args": {
|
| 99 |
+
"should_epoch_stop": false,
|
| 100 |
+
"should_evaluate": false,
|
| 101 |
+
"should_log": false,
|
| 102 |
+
"should_save": true,
|
| 103 |
+
"should_training_stop": false
|
| 104 |
+
},
|
| 105 |
+
"attributes": {}
|
| 106 |
+
}
|
| 107 |
+
},
|
| 108 |
+
"total_flos": 5.856626086901909e+16,
|
| 109 |
+
"train_batch_size": 64,
|
| 110 |
+
"trial_name": null,
|
| 111 |
+
"trial_params": null
|
| 112 |
+
}
|
ksplit1/checkpoint-1473/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c84b1503595e1e15162fe3abad80b56a1ddfd1a495507d81bd94e5c04412d23
|
| 3 |
+
size 5176
|
ksplit1/checkpoint-1964/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": ".venv/src/geneformer/gf-12L-95M-i4096/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.02,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "relu",
|
| 9 |
+
"hidden_dropout_prob": 0.02,
|
| 10 |
+
"hidden_size": 512,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1024,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 8,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.41.0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 20275
|
| 26 |
+
}
|
ksplit1/checkpoint-1964/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bdd727d62546e0e1df8ceb24361c2af25dadf2b7a0b2a82d98dee5517d24a98a
|
| 3 |
+
size 151931712
|
ksplit1/checkpoint-1964/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0a3a57216c236077971f0a72d4ff8892c503c25c5ebf8c9f55bbf5bc9a921809
|
| 3 |
+
size 236642362
|
ksplit1/checkpoint-1964/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff1b1a6ab8d0a223133da66ecbd2b14a66d1eab924a083bf41cabbf1606c8c6e
|
| 3 |
+
size 14244
|
ksplit1/checkpoint-1964/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f41ba4e70647289959d70b8f1a1045783e7ca5fda2f4724c1399250746e9a1d1
|
| 3 |
+
size 1064
|
ksplit1/checkpoint-1964/trainer_state.json
ADDED
|
@@ -0,0 +1,143 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3622211515903473,
|
| 3 |
+
"best_model_checkpoint": "finetuned/250416_geneformer_cellClassifier_many_control/ksplit1/checkpoint-982",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1964,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.39918533604887985,
|
| 13 |
+
"grad_norm": 1.4352002143859863,
|
| 14 |
+
"learning_rate": 1.9869391106802152e-05,
|
| 15 |
+
"loss": 0.3426,
|
| 16 |
+
"step": 196
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.7983706720977597,
|
| 20 |
+
"grad_norm": 4.147289276123047,
|
| 21 |
+
"learning_rate": 1.8813246895441423e-05,
|
| 22 |
+
"loss": 0.1893,
|
| 23 |
+
"step": 392
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.8260050659219328,
|
| 28 |
+
"eval_loss": 0.3700631558895111,
|
| 29 |
+
"eval_macro_f1": 0.7135959847549435,
|
| 30 |
+
"eval_runtime": 163.0259,
|
| 31 |
+
"eval_samples_per_second": 94.445,
|
| 32 |
+
"eval_steps_per_second": 1.969,
|
| 33 |
+
"step": 491
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.1975560081466394,
|
| 37 |
+
"grad_norm": 8.350934028625488,
|
| 38 |
+
"learning_rate": 1.6804080338412108e-05,
|
| 39 |
+
"loss": 0.1337,
|
| 40 |
+
"step": 588
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.5967413441955194,
|
| 44 |
+
"grad_norm": 17.019676208496094,
|
| 45 |
+
"learning_rate": 1.4059153092823729e-05,
|
| 46 |
+
"loss": 0.1051,
|
| 47 |
+
"step": 784
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.9959266802443993,
|
| 51 |
+
"grad_norm": 3.5994579792022705,
|
| 52 |
+
"learning_rate": 1.0875288455846521e-05,
|
| 53 |
+
"loss": 0.0962,
|
| 54 |
+
"step": 980
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.0,
|
| 58 |
+
"eval_accuracy": 0.850360459829837,
|
| 59 |
+
"eval_loss": 0.3622211515903473,
|
| 60 |
+
"eval_macro_f1": 0.7522559675724617,
|
| 61 |
+
"eval_runtime": 271.2733,
|
| 62 |
+
"eval_samples_per_second": 56.758,
|
| 63 |
+
"eval_steps_per_second": 1.183,
|
| 64 |
+
"step": 982
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 2.395112016293279,
|
| 68 |
+
"grad_norm": 7.739411354064941,
|
| 69 |
+
"learning_rate": 7.596774313729619e-06,
|
| 70 |
+
"loss": 0.0851,
|
| 71 |
+
"step": 1176
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 2.794297352342159,
|
| 75 |
+
"grad_norm": 4.228972434997559,
|
| 76 |
+
"learning_rate": 4.578133497286259e-06,
|
| 77 |
+
"loss": 0.0723,
|
| 78 |
+
"step": 1372
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.0,
|
| 82 |
+
"eval_accuracy": 0.8550366954601546,
|
| 83 |
+
"eval_loss": 0.3901084065437317,
|
| 84 |
+
"eval_macro_f1": 0.7393773894378316,
|
| 85 |
+
"eval_runtime": 234.4812,
|
| 86 |
+
"eval_samples_per_second": 65.664,
|
| 87 |
+
"eval_steps_per_second": 1.369,
|
| 88 |
+
"step": 1473
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.1934826883910388,
|
| 92 |
+
"grad_norm": 5.064053535461426,
|
| 93 |
+
"learning_rate": 2.145787379450612e-06,
|
| 94 |
+
"loss": 0.0698,
|
| 95 |
+
"step": 1568
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 3.5926680244399183,
|
| 99 |
+
"grad_norm": 9.535772323608398,
|
| 100 |
+
"learning_rate": 5.627582300170908e-07,
|
| 101 |
+
"loss": 0.064,
|
| 102 |
+
"step": 1764
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 3.9918533604887982,
|
| 106 |
+
"grad_norm": 8.033270835876465,
|
| 107 |
+
"learning_rate": 2.2724601255630364e-10,
|
| 108 |
+
"loss": 0.0663,
|
| 109 |
+
"step": 1960
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 4.0,
|
| 113 |
+
"eval_accuracy": 0.8575696564265766,
|
| 114 |
+
"eval_loss": 0.38419830799102783,
|
| 115 |
+
"eval_macro_f1": 0.7422031043489273,
|
| 116 |
+
"eval_runtime": 248.2004,
|
| 117 |
+
"eval_samples_per_second": 62.035,
|
| 118 |
+
"eval_steps_per_second": 1.293,
|
| 119 |
+
"step": 1964
|
| 120 |
+
}
|
| 121 |
+
],
|
| 122 |
+
"logging_steps": 196,
|
| 123 |
+
"max_steps": 1964,
|
| 124 |
+
"num_input_tokens_seen": 0,
|
| 125 |
+
"num_train_epochs": 4,
|
| 126 |
+
"save_steps": 500,
|
| 127 |
+
"stateful_callbacks": {
|
| 128 |
+
"TrainerControl": {
|
| 129 |
+
"args": {
|
| 130 |
+
"should_epoch_stop": false,
|
| 131 |
+
"should_evaluate": false,
|
| 132 |
+
"should_log": false,
|
| 133 |
+
"should_save": true,
|
| 134 |
+
"should_training_stop": true
|
| 135 |
+
},
|
| 136 |
+
"attributes": {}
|
| 137 |
+
}
|
| 138 |
+
},
|
| 139 |
+
"total_flos": 7.81329435167825e+16,
|
| 140 |
+
"train_batch_size": 64,
|
| 141 |
+
"trial_name": null,
|
| 142 |
+
"trial_params": null
|
| 143 |
+
}
|
ksplit1/checkpoint-1964/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c84b1503595e1e15162fe3abad80b56a1ddfd1a495507d81bd94e5c04412d23
|
| 3 |
+
size 5176
|
ksplit1/checkpoint-491/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": ".venv/src/geneformer/gf-12L-95M-i4096/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.02,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "relu",
|
| 9 |
+
"hidden_dropout_prob": 0.02,
|
| 10 |
+
"hidden_size": 512,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1024,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 8,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.41.0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 20275
|
| 26 |
+
}
|
ksplit1/checkpoint-491/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae8808380b2210aac733883d5cde164a650678eea0e836542109e131417057bd
|
| 3 |
+
size 151931712
|
ksplit1/checkpoint-491/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:198b1a850b3a8656bbf14d0a7e5089d07f502afc74ef079d2a6e25bd4a49b32e
|
| 3 |
+
size 236642362
|
ksplit1/checkpoint-491/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af9d07fa7fae4b23e3abefba5f7f904596b2cc33aa4b57ecd002c3b007712e44
|
| 3 |
+
size 14244
|
ksplit1/checkpoint-491/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa0552e41bd88ad0b0e994324a403690f329640fddf6c50134bc6c53ee5e1170
|
| 3 |
+
size 1064
|
ksplit1/checkpoint-491/trainer_state.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3700631558895111,
|
| 3 |
+
"best_model_checkpoint": "finetuned/250416_geneformer_cellClassifier_many_control/ksplit1/checkpoint-491",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 491,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.39918533604887985,
|
| 13 |
+
"grad_norm": 1.4352002143859863,
|
| 14 |
+
"learning_rate": 1.9869391106802152e-05,
|
| 15 |
+
"loss": 0.3426,
|
| 16 |
+
"step": 196
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.7983706720977597,
|
| 20 |
+
"grad_norm": 4.147289276123047,
|
| 21 |
+
"learning_rate": 1.8813246895441423e-05,
|
| 22 |
+
"loss": 0.1893,
|
| 23 |
+
"step": 392
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.8260050659219328,
|
| 28 |
+
"eval_loss": 0.3700631558895111,
|
| 29 |
+
"eval_macro_f1": 0.7135959847549435,
|
| 30 |
+
"eval_runtime": 163.0259,
|
| 31 |
+
"eval_samples_per_second": 94.445,
|
| 32 |
+
"eval_steps_per_second": 1.969,
|
| 33 |
+
"step": 491
|
| 34 |
+
}
|
| 35 |
+
],
|
| 36 |
+
"logging_steps": 196,
|
| 37 |
+
"max_steps": 1964,
|
| 38 |
+
"num_input_tokens_seen": 0,
|
| 39 |
+
"num_train_epochs": 4,
|
| 40 |
+
"save_steps": 500,
|
| 41 |
+
"stateful_callbacks": {
|
| 42 |
+
"TrainerControl": {
|
| 43 |
+
"args": {
|
| 44 |
+
"should_epoch_stop": false,
|
| 45 |
+
"should_evaluate": false,
|
| 46 |
+
"should_log": false,
|
| 47 |
+
"should_save": true,
|
| 48 |
+
"should_training_stop": false
|
| 49 |
+
},
|
| 50 |
+
"attributes": {}
|
| 51 |
+
}
|
| 52 |
+
},
|
| 53 |
+
"total_flos": 1.954272982146427e+16,
|
| 54 |
+
"train_batch_size": 64,
|
| 55 |
+
"trial_name": null,
|
| 56 |
+
"trial_params": null
|
| 57 |
+
}
|
ksplit1/checkpoint-491/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c84b1503595e1e15162fe3abad80b56a1ddfd1a495507d81bd94e5c04412d23
|
| 3 |
+
size 5176
|
ksplit1/checkpoint-982/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": ".venv/src/geneformer/gf-12L-95M-i4096/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.02,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "relu",
|
| 9 |
+
"hidden_dropout_prob": 0.02,
|
| 10 |
+
"hidden_size": 512,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1024,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 8,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.41.0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 20275
|
| 26 |
+
}
|
ksplit1/checkpoint-982/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca9bea864eb95cd39b648d02edd6ff9c5170a57b18b26128dddfdb07b8e1e562
|
| 3 |
+
size 151931712
|
ksplit1/checkpoint-982/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ee5810ba82b628319afa7687077381cc86674ef1116c4d643f80ab8cd8f1bc0
|
| 3 |
+
size 236642362
|
ksplit1/checkpoint-982/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e9f4c27ee1b577c399437f9d8d00209a75fbaa684bf726ca7c491e9bc0f274f
|
| 3 |
+
size 14244
|
ksplit1/checkpoint-982/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4819be9772d193d1c97e75f988e40d3a1cc67d420862d2cb8b256fa7070d81c7
|
| 3 |
+
size 1064
|
ksplit1/checkpoint-982/trainer_state.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3622211515903473,
|
| 3 |
+
"best_model_checkpoint": "finetuned/250416_geneformer_cellClassifier_many_control/ksplit1/checkpoint-982",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 982,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.39918533604887985,
|
| 13 |
+
"grad_norm": 1.4352002143859863,
|
| 14 |
+
"learning_rate": 1.9869391106802152e-05,
|
| 15 |
+
"loss": 0.3426,
|
| 16 |
+
"step": 196
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.7983706720977597,
|
| 20 |
+
"grad_norm": 4.147289276123047,
|
| 21 |
+
"learning_rate": 1.8813246895441423e-05,
|
| 22 |
+
"loss": 0.1893,
|
| 23 |
+
"step": 392
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.8260050659219328,
|
| 28 |
+
"eval_loss": 0.3700631558895111,
|
| 29 |
+
"eval_macro_f1": 0.7135959847549435,
|
| 30 |
+
"eval_runtime": 163.0259,
|
| 31 |
+
"eval_samples_per_second": 94.445,
|
| 32 |
+
"eval_steps_per_second": 1.969,
|
| 33 |
+
"step": 491
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.1975560081466394,
|
| 37 |
+
"grad_norm": 8.350934028625488,
|
| 38 |
+
"learning_rate": 1.6804080338412108e-05,
|
| 39 |
+
"loss": 0.1337,
|
| 40 |
+
"step": 588
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.5967413441955194,
|
| 44 |
+
"grad_norm": 17.019676208496094,
|
| 45 |
+
"learning_rate": 1.4059153092823729e-05,
|
| 46 |
+
"loss": 0.1051,
|
| 47 |
+
"step": 784
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.9959266802443993,
|
| 51 |
+
"grad_norm": 3.5994579792022705,
|
| 52 |
+
"learning_rate": 1.0875288455846521e-05,
|
| 53 |
+
"loss": 0.0962,
|
| 54 |
+
"step": 980
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.0,
|
| 58 |
+
"eval_accuracy": 0.850360459829837,
|
| 59 |
+
"eval_loss": 0.3622211515903473,
|
| 60 |
+
"eval_macro_f1": 0.7522559675724617,
|
| 61 |
+
"eval_runtime": 271.2733,
|
| 62 |
+
"eval_samples_per_second": 56.758,
|
| 63 |
+
"eval_steps_per_second": 1.183,
|
| 64 |
+
"step": 982
|
| 65 |
+
}
|
| 66 |
+
],
|
| 67 |
+
"logging_steps": 196,
|
| 68 |
+
"max_steps": 1964,
|
| 69 |
+
"num_input_tokens_seen": 0,
|
| 70 |
+
"num_train_epochs": 4,
|
| 71 |
+
"save_steps": 500,
|
| 72 |
+
"stateful_callbacks": {
|
| 73 |
+
"TrainerControl": {
|
| 74 |
+
"args": {
|
| 75 |
+
"should_epoch_stop": false,
|
| 76 |
+
"should_evaluate": false,
|
| 77 |
+
"should_log": false,
|
| 78 |
+
"should_save": true,
|
| 79 |
+
"should_training_stop": false
|
| 80 |
+
},
|
| 81 |
+
"attributes": {}
|
| 82 |
+
}
|
| 83 |
+
},
|
| 84 |
+
"total_flos": 3.904358296838933e+16,
|
| 85 |
+
"train_batch_size": 64,
|
| 86 |
+
"trial_name": null,
|
| 87 |
+
"trial_params": null
|
| 88 |
+
}
|
ksplit1/checkpoint-982/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c84b1503595e1e15162fe3abad80b56a1ddfd1a495507d81bd94e5c04412d23
|
| 3 |
+
size 5176
|
ksplit1/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": ".venv/src/geneformer/gf-12L-95M-i4096/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.02,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "relu",
|
| 9 |
+
"hidden_dropout_prob": 0.02,
|
| 10 |
+
"hidden_size": 512,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1024,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 8,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.41.0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 20275
|
| 26 |
+
}
|
ksplit1/many_control_pred_dict.pkl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36a809c95a7c22650d8827b28e8d7ae304b2edfc3a877526a0e606ba33369c65
|
| 3 |
+
size 400531
|
ksplit1/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca9bea864eb95cd39b648d02edd6ff9c5170a57b18b26128dddfdb07b8e1e562
|
| 3 |
+
size 151931712
|
ksplit1/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c84b1503595e1e15162fe3abad80b56a1ddfd1a495507d81bd94e5c04412d23
|
| 3 |
+
size 5176
|
many_control_eval_metrics_dict.pkl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47be69484fca74c721838109cc58e72be4f00b5ff6631f42dc8943a96e7cb5d4
|
| 3 |
+
size 2540
|