Merge pull request #6 from Miking98/readme

README update with species

README.md

@@ -282,28 +282,26 @@ python -m train wandb=null experiment=hg38/chromatin_profile dataset.ref_genome_
 
 ### Species Classification
 
-You'll need to download fasta files for each species that you want to use (just the .zips, the dataloader wil unzip automatically).
+You'll need to download fasta files for each species that you want to use (just the .zips, the dataloader wil unzip automatically). You can download them using the following commands:
 
-
-Sample species run:  
 ```
-python -m train wandb=null experiment=hg38/species dataset.species=[human,mouse,hippo,pig,lemur] train.global_batch_size=256 optimizer.lr=6e-5 trainer.devices=1 dataset.batch_size=1 dataset.max_length=1024 dataset.species_dir=/path/to/data/species/ model.layer.l_max=1026 model.d_model=128 model.n_layer=2 trainer.max_epochs=150 decoder.mode=last train.pretrained_model_path=null train.pretrained_model_state_hook=null
+# Human
+wget -P human/ -r -nH --cut-dirs=12 --no-parent ftp://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/009/914/755/GCA_009914755.4_T2T-CHM13v2.0/GCA_009914755.4_T2T-CHM13v2.0_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/
+# Lemur
+wget -P lemur/ -r -nH --cut-dirs=11 --no-parent ftp://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Lemur_catta/latest_assembly_versions/GCA_020740605.1_mLemCat1.pri/GCA_020740605.1_mLemCat1.pri_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/
+# House mouse
+wget -P mouse/ -r -nH --cut-dirs=11 --no-parent ftp://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Mus_musculus/latest_assembly_versions/GCA_921998355.2_A_J_v3/GCA_921998355.2_A_J_v3_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/
+# Pig
+wget -P pig/ -r -nH --cut-dirs=11 --no-parent ftp://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Sus_scrofa/latest_assembly_versions/GCA_002844635.1_USMARCv1.0/GCA_002844635.1_USMARCv1.0_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/
+# Hippo
+wget -P hippo/ -r -nH --cut-dirs=11 --no-parent ftp://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Hippopotamus_amphibius/latest_assembly_versions/GCA_023065835.1_ASM2306583v1/GCA_023065835.1_ASM2306583v1_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/
 ```
 
-Let's break some of these args down:
-- `experiment=hg38/species`  # main config for this experiment
-- `dataset.species`  # list of species you want (and already downloaded their .fasta files)
-- `decoder.mode=last`  # using the last token to classify (instead of default pooling)
-- `train.pretrained_model_path`  # if using a pretrained model, point to it, if not, set to null
-- `train.pretrained_model_state_hook=null`  # if using a pretrained model, this will load the weights properly (and not head). if not, set to null
-
-
 Your folder struture should look like this:  
 
 ```
 data
 |-- species/
-    |-- README.txt: Contains links to where the `.fna` files were downloaded from.
     |-- chimpanzee/
         |-- chr1.fna
         |-- chr2.fna
@@ -329,6 +327,19 @@ data
 ```
 
 
+Sample species run: 
+```
+python -m train wandb=null experiment=hg38/species dataset.species=[human,mouse,hippo,pig,lemur] train.global_batch_size=256 optimizer.lr=6e-5 trainer.devices=1 dataset.batch_size=1 dataset.max_length=1024 dataset.species_dir=/path/to/data/species/ model.layer.l_max=1026 model.d_model=128 model.n_layer=2 trainer.max_epochs=150 decoder.mode=last train.pretrained_model_path=null train.pretrained_model_state_hook=null
+```
+
+Let's break some of these args down:
+- `experiment=hg38/species`  # main config for this experiment
+- `dataset.species`  # list of species you want (and already downloaded their .fasta files)
+- `decoder.mode=last`  # using the last token to classify (instead of default pooling)
+- `train.pretrained_model_path`  # if using a pretrained model, point to it, if not, set to null
+- `train.pretrained_model_state_hook=null`  # if using a pretrained model, this will load the weights properly (and not head). if not, set to null
+
+
 # More advanced stuff below
 
 
@@ -527,7 +538,7 @@ In the sample config, see the
 
 Things to note:
 
-Train dataset will change during training, but the test set will always be fixed.  The test len/batch size is set the normal way in your command launch, ie, `dataset.batch_size`, `dataset.
+Train dataset will change during training, but the test set will always be fixed.  The test len/batch size is set the normal way in your command launch, ie, `dataset.batch_size`, `dataset`.
 
 
 ## Citation